computer science Recently Published Documents

Total documents.

  • Latest Documents
  • Most Cited Documents
  • Contributed Authors
  • Related Sources
  • Related Keywords

Hiring CS Graduates: What We Learned from Employers

Computer science ( CS ) majors are in high demand and account for a large part of national computer and information technology job market applicants. Employment in this sector is projected to grow 12% between 2018 and 2028, which is faster than the average of all other occupations. Published data are available on traditional non-computer science-specific hiring processes. However, the hiring process for CS majors may be different. It is critical to have up-to-date information on questions such as “what positions are in high demand for CS majors?,” “what is a typical hiring process?,” and “what do employers say they look for when hiring CS graduates?” This article discusses the analysis of a survey of 218 recruiters hiring CS graduates in the United States. We used Atlas.ti to analyze qualitative survey data and report the results on what positions are in the highest demand, the hiring process, and the resume review process. Our study revealed that a software developer was the most common job the recruiters were looking to fill. We found that the hiring process steps for CS graduates are generally aligned with traditional hiring steps, with an additional emphasis on technical and coding tests. Recruiters reported that their hiring choices were based on reviewing resume’s experience, GPA, and projects sections. The results provide insights into the hiring process, decision making, resume analysis, and some discrepancies between current undergraduate CS program outcomes and employers’ expectations.

A Systematic Literature Review of Empiricism and Norms of Reporting in Computing Education Research Literature

Context. Computing Education Research (CER) is critical to help the computing education community and policy makers support the increasing population of students who need to learn computing skills for future careers. For a community to systematically advance knowledge about a topic, the members must be able to understand published work thoroughly enough to perform replications, conduct meta-analyses, and build theories. There is a need to understand whether published research allows the CER community to systematically advance knowledge and build theories. Objectives. The goal of this study is to characterize the reporting of empiricism in Computing Education Research literature by identifying whether publications include content necessary for researchers to perform replications, meta-analyses, and theory building. We answer three research questions related to this goal: (RQ1) What percentage of papers in CER venues have some form of empirical evaluation? (RQ2) Of the papers that have empirical evaluation, what are the characteristics of the empirical evaluation? (RQ3) Of the papers that have empirical evaluation, do they follow norms (both for inclusion and for labeling of information needed for replication, meta-analysis, and, eventually, theory-building) for reporting empirical work? Methods. We conducted a systematic literature review of the 2014 and 2015 proceedings or issues of five CER venues: Technical Symposium on Computer Science Education (SIGCSE TS), International Symposium on Computing Education Research (ICER), Conference on Innovation and Technology in Computer Science Education (ITiCSE), ACM Transactions on Computing Education (TOCE), and Computer Science Education (CSE). We developed and applied the CER Empiricism Assessment Rubric to the 427 papers accepted and published at these venues over 2014 and 2015. Two people evaluated each paper using the Base Rubric for characterizing the paper. An individual person applied the other rubrics to characterize the norms of reporting, as appropriate for the paper type. Any discrepancies or questions were discussed between multiple reviewers to resolve. Results. We found that over 80% of papers accepted across all five venues had some form of empirical evaluation. Quantitative evaluation methods were the most frequently reported. Papers most frequently reported results on interventions around pedagogical techniques, curriculum, community, or tools. There was a split in papers that had some type of comparison between an intervention and some other dataset or baseline. Most papers reported related work, following the expectations for doing so in the SIGCSE and CER community. However, many papers were lacking properly reported research objectives, goals, research questions, or hypotheses; description of participants; study design; data collection; and threats to validity. These results align with prior surveys of the CER literature. Conclusions. CER authors are contributing empirical results to the literature; however, not all norms for reporting are met. We encourage authors to provide clear, labeled details about their work so readers can use the study methodologies and results for replications and meta-analyses. As our community grows, our reporting of CER should mature to help establish computing education theory to support the next generation of computing learners.

Light Diacritic Restoration to Disambiguate Homographs in Modern Arabic Texts

Diacritic restoration (also known as diacritization or vowelization) is the process of inserting the correct diacritical markings into a text. Modern Arabic is typically written without diacritics, e.g., newspapers. This lack of diacritical markings often causes ambiguity, and though natives are adept at resolving, there are times they may fail. Diacritic restoration is a classical problem in computer science. Still, as most of the works tackle the full (heavy) diacritization of text, we, however, are interested in diacritizing the text using a fewer number of diacritics. Studies have shown that a fully diacritized text is visually displeasing and slows down the reading. This article proposes a system to diacritize homographs using the least number of diacritics, thus the name “light.” There is a large class of words that fall under the homograph category, and we will be dealing with the class of words that share the spelling but not the meaning. With fewer diacritics, we do not expect any effect on reading speed, while eye strain is reduced. The system contains morphological analyzer and context similarities. The morphological analyzer is used to generate all word candidates for diacritics. Then, through a statistical approach and context similarities, we resolve the homographs. Experimentally, the system shows very promising results, and our best accuracy is 85.6%.

A genre-based analysis of questions and comments in Q&A sessions after conference paper presentations in computer science

Gender diversity in computer science at a large public r1 research university: reporting on a self-study.

With the number of jobs in computer occupations on the rise, there is a greater need for computer science (CS) graduates than ever. At the same time, most CS departments across the country are only seeing 25–30% of women students in their classes, meaning that we are failing to draw interest from a large portion of the population. In this work, we explore the gender gap in CS at Rutgers University–New Brunswick, a large public R1 research university, using three data sets that span thousands of students across six academic years. Specifically, we combine these data sets to study the gender gaps in four core CS courses and explore the correlation of several factors with retention and the impact of these factors on changes to the gender gap as students proceed through the CS courses toward completing the CS major. For example, we find that a significant percentage of women students taking the introductory CS1 course for majors do not intend to major in CS, which may be a contributing factor to a large increase in the gender gap immediately after CS1. This finding implies that part of the retention task is attracting these women students to further explore the major. Results from our study include both novel findings and findings that are consistent with known challenges for increasing gender diversity in CS. In both cases, we provide extensive quantitative data in support of the findings.

Designing for Student-Directedness: How K–12 Teachers Utilize Peers to Support Projects

Student-directed projects—projects in which students have individual control over what they create and how to create it—are a promising practice for supporting the development of conceptual understanding and personal interest in K–12 computer science classrooms. In this article, we explore a central (and perhaps counterintuitive) design principle identified by a group of K–12 computer science teachers who support student-directed projects in their classrooms: in order for students to develop their own ideas and determine how to pursue them, students must have opportunities to engage with other students’ work. In this qualitative study, we investigated the instructional practices of 25 K–12 teachers using a series of in-depth, semi-structured interviews to develop understandings of how they used peer work to support student-directed projects in their classrooms. Teachers described supporting their students in navigating three stages of project development: generating ideas, pursuing ideas, and presenting ideas. For each of these three stages, teachers considered multiple factors to encourage engagement with peer work in their classrooms, including the quality and completeness of shared work and the modes of interaction with the work. We discuss how this pedagogical approach offers students new relationships to their own learning, to their peers, and to their teachers and communicates important messages to students about their own competence and agency, potentially contributing to aims within computer science for broadening participation.

Creativity in CS1: A Literature Review

Computer science is a fast-growing field in today’s digitized age, and working in this industry often requires creativity and innovative thought. An issue within computer science education, however, is that large introductory programming courses often involve little opportunity for creative thinking within coursework. The undergraduate introductory programming course (CS1) is notorious for its poor student performance and retention rates across multiple institutions. Integrating opportunities for creative thinking may help combat this issue by adding a personal touch to course content, which could allow beginner CS students to better relate to the abstract world of programming. Research on the role of creativity in computer science education (CSE) is an interesting area with a lot of room for exploration due to the complexity of the phenomenon of creativity as well as the CSE research field being fairly new compared to some other education fields where this topic has been more closely explored. To contribute to this area of research, this article provides a literature review exploring the concept of creativity as relevant to computer science education and CS1 in particular. Based on the review of the literature, we conclude creativity is an essential component to computer science, and the type of creativity that computer science requires is in fact, a teachable skill through the use of various tools and strategies. These strategies include the integration of open-ended assignments, large collaborative projects, learning by teaching, multimedia projects, small creative computational exercises, game development projects, digitally produced art, robotics, digital story-telling, music manipulation, and project-based learning. Research on each of these strategies and their effects on student experiences within CS1 is discussed in this review. Last, six main components of creativity-enhancing activities are identified based on the studies about incorporating creativity into CS1. These components are as follows: Collaboration, Relevance, Autonomy, Ownership, Hands-On Learning, and Visual Feedback. The purpose of this article is to contribute to computer science educators’ understanding of how creativity is best understood in the context of computer science education and explore practical applications of creativity theory in CS1 classrooms. This is an important collection of information for restructuring aspects of future introductory programming courses in creative, innovative ways that benefit student learning.

CATS: Customizable Abstractive Topic-based Summarization

Neural sequence-to-sequence models are the state-of-the-art approach used in abstractive summarization of textual documents, useful for producing condensed versions of source text narratives without being restricted to using only words from the original text. Despite the advances in abstractive summarization, custom generation of summaries (e.g., towards a user’s preference) remains unexplored. In this article, we present CATS, an abstractive neural summarization model that summarizes content in a sequence-to-sequence fashion while also introducing a new mechanism to control the underlying latent topic distribution of the produced summaries. We empirically illustrate the efficacy of our model in producing customized summaries and present findings that facilitate the design of such systems. We use the well-known CNN/DailyMail dataset to evaluate our model. Furthermore, we present a transfer-learning method and demonstrate the effectiveness of our approach in a low resource setting, i.e., abstractive summarization of meetings minutes, where combining the main available meetings’ transcripts datasets, AMI and International Computer Science Institute(ICSI) , results in merely a few hundred training documents.

Exploring students’ and lecturers’ views on collaboration and cooperation in computer science courses - a qualitative analysis

Factors affecting student educational choices regarding oer material in computer science, export citation format, share document.


Publishing our work allows us to share ideas and work collaboratively to advance the field of computer science.

People looking at a screen

  • Algorithms and Optimization 323
  • Applied science 186
  • Climate and Sustainability 10
  • Cloud AI 46
  • Language 235
  • Perception 291

Research Area

  • Algorithms and Theory 1317
  • Data Management 166
  • Data Mining and Modeling 353
  • Distributed Systems and Parallel Computing 340
  • Economics and Electronic Commerce 340
  • Education Innovation 68
  • General Science 329
  • Hardware and Architecture 145
  • Health & Bioscience 359
  • Human-Computer Interaction and Visualization 803
  • Information Retrieval and the Web 414
  • Machine Intelligence 3781
  • Machine Perception 1449
  • Machine Translation 145
  • Mobile Systems 107
  • Natural Language Processing 1068
  • Networking 312
  • Quantum Computing 124
  • Responsible AI 221
  • Robotics 198
  • Security, Privacy and Abuse Prevention 491
  • Software Engineering 199
  • Software Systems 448
  • Speech Processing 542
  • Title, desc

We believe open collaboration is essential for progress

We're proud to work with academic and research institutions to push the boundaries of AI and computer science. Learn more about our student and faculty programs, as well as our global outreach initiatives.


Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • View all journals
  • Explore content
  • About the journal
  • Publish with us
  • Sign up for alerts
  • Review Article
  • Published: 04 March 2010

Quantum computers

  • T. D. Ladd 1   nAff11 ,
  • F. Jelezko 2 ,
  • R. Laflamme 3 , 4 , 5 ,
  • Y. Nakamura 6 , 7 ,
  • C. Monroe 8 , 9 &
  • J. L. O’Brien 10  

Nature volume  464 ,  pages 45–53 ( 2010 ) Cite this article

76k Accesses

2266 Citations

60 Altmetric

Metrics details

  • Computer science
  • Quantum information
  • Quantum mechanics

Over the past several decades, quantum information science has emerged to seek answers to the question: can we gain some advantage by storing, transmitting and processing information encoded in systems that exhibit unique quantum properties? Today it is understood that the answer is yes, and many research groups around the world are working towards the highly ambitious technological goal of building a quantum computer, which would dramatically improve computational power for particular tasks. A number of physical systems, spanning much of modern physics, are being developed for quantum computation. However, it remains unclear which technology, if any, will ultimately prove successful. Here we describe the latest developments for each of the leading approaches and explain the major challenges for the future.

This is a preview of subscription content, access via your institution

Access options

Subscribe to this journal

Receive 51 print issues and online access

185,98 € per year

only 3,65 € per issue

Rent or buy this article

Prices vary by article type

Prices may be subject to local taxes which are calculated during checkout

research paper on computers

Nielsen, M. A. & Chuang, I. L. Quantum Computation and Quantum Information (Cambridge University Press, 2000)

MATH   Google Scholar  

Knill, E. Quantum computing with realistically noisy devices. Nature 434 , 39–44 (2005)

ADS   CAS   PubMed   Google Scholar  

DiVincenzo, D. P. The physical implementation of quantum computation. Fortschr. Phys. 48 , 771–783 (2000)

Mizel, A., Lidar, D. A. & Mitchell, M. Simple proof of equivalence between adiabatic quantum computation and the circuit model. Phys. Rev. Lett. 99 , 070502 (2007)

ADS   PubMed   Google Scholar  

Raussendorf, R. & Briegel, H. J. A one-way quantum computer. Phys. Rev. Lett. 86 , 5188–5191 (2001)

Cory, D. G., Fahmy, A. F. & Havel, T. F. Ensemble quantum computing by NMR-spectroscopy. Proc. Natl Acad. Sci. USA 94 , 1634–1639 (1997)

ADS   CAS   PubMed   PubMed Central   Google Scholar  

Gershenfeld, N. A. & Chuang, I. L. Bulk spin resonance quantum computation. Science 275 , 350–356 (1997)

MathSciNet   CAS   PubMed   MATH   Google Scholar  

Ryan, C. A., Moussa, O., Baugh, J. & Laflamme, R. Spin based heat engine: demonstration of multiple rounds of algorithmic cooling. Phys. Rev. Lett. 100 , 140501 (2008)

Shor, P. W. & Jordan, S. P. Estimating Jones polynomials is a complete problem for one clean qubit. Quant. Inform. Comput. 8 , 681–714 (2008)

MathSciNet   MATH   Google Scholar  

Braunstein, S. L. & van Loock, P. Quantum information with continuous variables. Rev. Mod. Phys. 77 , 513–577 (2005)

ADS   MathSciNet   MATH   Google Scholar  

Schmidt, H. & Imamoglu, A. Giant Kerr nonlinearities obtained by electromagnetically induced transparency. Opt. Lett. 21 , 1936–1938 (1996)

Duan, L. M. & Kimble, H. J. Scalable photonic quantum computation through cavity-assisted interactions. Phys. Rev. Lett. 92 , 127902 (2004)

Knill, E., Laflamme, R. & Milburn, G. J. A scheme for efficient quantum computation with linear optics. Nature 409 , 46–52 (2001)

Politi, A., Matthews, J. C. F. & O’Brien, J. L. Shor’s quantum factoring algorithm on a photonic chip. Science 325 , 1221 (2009)

ADS   MathSciNet   CAS   PubMed   MATH   Google Scholar  

O’Brien, J. L. Optical quantum computing. Science 318 , 1567–1570 (2007)

Migdal, A. & Dowling, J. eds. Single-photon detectors, applications, and measurement. J. Mod. Opt. 51 , (2004)

Hadfield, R. H. Single-photon detectors for optical quantum information applications. Nature Photon. 3 , 696–705 (2009)

ADS   CAS   Google Scholar  

Grangier, P., Sanders, B. & Vuckovic, J. eds. Focus on single photons on demand. New J. Phys. 6 , (2004)

Shields, A. J. Semiconductor quantum light sources. Nature Photon. 1 , 215–223 (2007)

Matthews, J. C. F., Politi, A., Stefanov, A. & O’Brien, J. L. Manipulation of multiphoton entanglement in waveguide quantum circuits. Nature Photon. 3 , 346–350 (2009)

Kistner, C. et al. Demonstration of strong coupling via electro-optical tuning in high-quality QD-micropillar systems. Opt. Express 16 , 15006–15012 (2008)

Fushman, I. et al. Controlled phase shifts with a single quantum dot. Science 320 , 769–772 (2008)

Gruber, A. et al. Scanning confocal optical microscopy and magnetic resonance on single defect centers. Science 276 , 2012–2014 (1997)

CAS   Google Scholar  

Devitt, S. J. et al. Photonic module: an on-demand resource for photonic entanglement. Phys. Rev. A 76 , 052312 (2007)

ADS   Google Scholar  

Wineland, D. J. et al. Experimental issues in coherent quantum-state manipulation of trapped atomic ions. J. Res. Natl. Inst. Stand. Technol. 103 , 259–328 (1998)

CAS   PubMed   PubMed Central   Google Scholar  

Wineland, D. & Blatt, R. Entangled states of trapped atomic ions. Nature 453 , 1008–1014 (2008)

Ospelkaus, C. et al. Trapped-ion quantum logic gates based on oscillating magnetic fields. Phys. Rev. Lett. 101 , 090502 (2008)

Garcia-Ripoll, J. J., Zoller, P. & Cirac, J. I. Speed optimized two-qubit gates with laser coherent control techniques for ion trap quantum computing. Phys. Rev. Lett. 91 , 157901 (2003)

Leibfried, D., Blatt, R., Monroe, C. & Wineland, D. Quantum dynamics of single trapped ions. Rev. Mod. Phys. 75 , 281–324 (2003)

Home, J. P. et al. Complete methods set for scalable ion trap quantum information processing. Science 325 , 1227–1230 (2009)

Olmschenk, S. et al. Quantum teleportation between distant matter qubits. Science 323 , 486–489 (2009)

Dür, W., Briegel, H. J., Cirac, J. I. & Zoller, P. Quantum repeaters based on entanglement purification. Phys. Rev. A 59 , 169–181 (1999)

Duan, L.-M. & Raussendorf, R. Efficient quantum computation with probabilistic quantum gates. Phys. Rev. Lett. 95 , 080503 (2005)

ADS   MathSciNet   PubMed   Google Scholar  

Morsch, O. & Oberthaler, M. Dynamics of Bose-Einstein condensates in optical lattices. Rev. Mod. Phys. 78 , 179–215 (2006)

Anderlini, M. et al. Controlled exchange interaction between pairs of neutral atoms in an optical lattice. Nature 448 , 452–456 (2007)

Urban, E. et al. Observation of Rydberg blockade between two atoms. Nature Phys. 5 , 110–114 (2009)

Gaëtan, A. et al. Observation of collective excitation of two individual atoms in the Rydberg blockade regime. Nature Phys. 5 , 115–118 (2009)

Negrevergne, C. et al. Benchmarking quantum control methods on a 12-qubit system. Phys. Rev. Lett. 96 , 170501 (2006)

Vandersypen, L. M. K. et al. Experimental realization of Shor’s quantum factoring algorithm using nuclear magnetic resonance. Nature 414 , 883–887 (2001)

Khaneja, N., Reiss, T., Kehlet, C., Schulte-Herbruggen, T. & Glaser, S. J. Optimal control of coupled spin dynamics: design of NMR pulse sequences by gradient ascent algorithms. J. Magn. Reson. 172 , 296–305 (2005)

Braunstein, S. L. et al. Separability of very noisy mixed states and implications for NMR quantum computing. Phys. Rev. Lett. 83 , 1054–1057 (1999)

Mehring, M., Mende, J. & Scherer, W. Entanglement between an electron and a nuclear spin 1/2. Phys. Rev. Lett. 90 , 153001 (2003)

Hanson, R., Kouwenhoven, L. P., Petta, J. R., Tarucha, S. & Vandersypen, L. M. K. Spins in few-electron quantum dots. Rev. Mod. Phys. 79 , 1217–1265 (2007)

Uhrig, S. G. Keeping a quantum bit alive by optimized π-pulse sequences. Phys. Rev. Lett. 98 , 100504 (2007)

Liu, H. W. et al. A gate-defined silicon quantum dot molecule. Appl. Phys. Lett. 92 , 222104 (2008)

Simmons, C. B. et al. Charge sensing and controllable tunnel coupling in a Si/SiGe double quantum dot. Nano Lett. 9 , 3234–3238 (2009)

Kane, B. E. A silicon-based nuclear spin quantum computer. Nature 393 , 133–137 (1998)

Vrijen, R. et al. Electron-spin-resonance transistors for quantum computing in silicon-germanium heterostructures. Phys. Rev. A 62 , 012306 (2000)

Tyryshkin, A. M. & Lyon, S. A. Data presented at the Silicon Qubit Workshop, 24–25 August (University of California, Berkeley; sponsored by Lawrence Berkeley National Laboratory and Sandia National Laboratory, 2009)

Ladd, T. D., Maryenko, D., Yamamoto, Y., Abe, E. & Itoh, K. M. Coherence time of decoupled nuclear spins in silicon. Phys. Rev. B 71 , 14401 (2005)

Yang, A. et al. Simultaneous subsecond hyperpolarization of the nuclear and electron spins of phosphorus in silicon by optical pumping of exciton transitions. Phys. Rev. Lett. 102 , 257401 (2009)

Batra, A., Weis, C. D., Reijonen, J., Persaud, A. & Schenkel, T. Detection of low energy single ion impacts in micron scale transistors at room temperature. Appl. Phys. Lett. 91 , 193502 (2007)

O’Brien, J. L. et al. Towards the fabrication of phosphorus qubits for a silicon quantum computer. Phys. Rev. B 64 , 161401 (2001)

Schneider, C. et al. Lithographic alignment to site-controlled quantum dots for device integration. Appl. Phys. Lett. 92 , 183101 (2008)

Atatüre, M. et al. Quantum-dot spin-state preparation with near-unity fidelity. Science 312 , 551–553 (2006)

Gerardot, B. D. et al. Optical pumping of a single hole spin in a quantum dot. Nature 451 , 441–444 (2008)

Press, D., Ladd, T. D., Zhang, B. Y. & Yamamoto, Y. Complete quantum control of a single quantum dot spin using ultrafast optical pulses. Nature 456 , 218–221 (2008)

Berezovsky, J. et al. Nondestructive optical measurements of a single electron spin in a quantum dot. Science 314 , 1916–1920 (2006)

Harrison, J., Sellars, M. J. & Manson, N. B. Measurement of the optically induced spin polarisation of N-V centres in diamond. Diamond Related Mater. 15 , 586–588 (2006)

Dutt, M. V. G. et al. Quantum register based on individual electronic and nuclear spin qubits in diamond. Science 316 , 1312–1316 (2007)

PubMed   Google Scholar  

Neumann, P. et al. Multipartite entanglement among single spins in diamond. Science 320 , 1326–1329 (2008)

Jiang, L. et al. Repetitive readout of a single electronic spin via quantum logic with nuclear spin ancillae. Science 326 , 267–272 (2009)

Hanson, R., Dobrovitski, V. V., Feiguin, A. E., Gywat, O. & Awschalom, D. D. Coherent dynamics of a single spin interacting with an adjustable spin bath. Science 320 , 352–355 (2008)

Takahashi, S., Hanson, R., van Tol, J., Sherwin, M. S. & Awschalom, D. D. Quenching spin decoherence in diamond through spin bath polarization. Phys. Rev. Lett. 101 , 047601 (2008)

Balasubramanian, G. et al. Ultralong spin coherence time in isotopically engineered diamond. Nature Mater. 8 , 383–387 (2009)

Neumann, P. et al. Scalable quantum register based on coupled electron spins in a room temperature solid. Nature Phys. 10.1038/nphys1536 (in the press)

Wang, C. F. et al. Fabrication and characterization of two-dimensional photonic crystal microcavities in nanocrystalline diamond. Appl. Phys. Lett. 91 , 201112 (2007)

Wu, E. et al. Room temperature triggered single-photon source in the near infrared. New J. Phys. 9 , 434 (2007)

Wang, C., Kurtsiefer, C., Weinfurter, H. & Burchard, B. Single photon emission from SiV centres in diamond produced by ion implantation. J. Phys. At. Mol. Opt. Phys. 39 , 37–41 (2006)

Sanaka, K., Pawlis, A., Ladd, T. D., Lischka, K. & Yamamoto, Y. Indistinguishable photons from independent semiconductor nanostructures. Phys. Rev. Lett. 103 , 053601 (2009)

Nakamura, Y., Pashkin, Yu. A. & Tsai, J. S. Coherent control of macroscopic quantum states in a single-Cooper-pair box. Nature 398 , 786–788 (1999)

Vion, D. et al. Manipulating the quantum state of an electrical circuit. Science 296 , 886–889 (2002)

Schreier, J. A. et al. Suppressing charge noise decoherence in superconducting charge qubits. Phys. Rev. B 77 , 180502 (2008)

Chiorescu, I., Nakamura, Y., Harmans, C. J. P. M. & Mooij, J. E. Coherent quantum dynamics of a superconducting flux qubit. Science 299 , 1869–1871 (2003)

Martinis, J. M., Nam, S., Aumentado, J. & Urbina, C. Rabi oscillations in a large Josephson-junction qubit. Phys. Rev. Lett. 89 , 117901 (2002)

Niskanen, A. O. et al. Quantum coherent tunable coupling of superconducting qubits. Science 316 , 723–726 (2007)

Harris, R. et al. Experimental demonstration of a robust and scalable flux qubit. Preprint at 〈 〉 (2009)

Wallraff, A. et al. Strong coupling of a single photon to a superconducting qubit using circuit quantum electrodynamics. Nature 431 , 162–167 (2004)

DiCarlo, L. et al. Demonstration of two-qubit algorithms with a superconducting quantum processor. Nature 260 , 240–244 (2009)

Ansmann, M. et al. Violation of Bell’s inequality in Josephson phase qubits. Nature 461 , 504–506 (2009)

Chow, J. M. et al. Entanglement metrology using a joint readout of superconducting qubits. Preprint at 〈 〉 (2009)

Lupascu, A. et al. Quantum non-demolition measurement of a superconducting two-level system. Nature Phys. 3 , 119–123 (2007)

Micheli, A., Brennen, G. K. & Zoller, P. A toolbox for lattice-spin models with polar molecules. Nature Phys. 2 , 341–347 (2006)

Rippe, L., Julsgaard, B., Walther, A., Ying, Y. & Kroll, S. Experimental quantum-state tomography of a solid-state qubit. Phys. Rev. A 77 , 022307 (2008)

de Riedmatten, H., Afzelius, M., Staudt, M. U., Simon, C. & Gisin, N. A solid-state light-matter interface at the single-photon level. Nature 456 , 773–777 (2008)

Morton, J. J. L. et al. Bang-bang control of fullerene qubits using ultrafast phase gates. Nature Phys. 2 , 40–43 (2006)

Mason, N., Biercuk, M. J. & Marcus, C. M. Local gate control of a carbon nanotube double quantum dot. Science 303 , 655–658 (2004)

Trauzettel, B., Bulaev, D. V., Loss, D. & Burkard, G. Spin qubits in graphene quantum dots. Nature Phys. 3 , 192–196 (2007)

Platzman, P. M. & Dykman, M. I. Quantum computing with electrons floating on liquid helium. Science 284 , 1967–1969 (1999)

CAS   PubMed   Google Scholar  

Leuenberger, M. N. & Loss, D. Quantum computing in molecular magnets. Nature 410 , 789–793 (2001)

Tian, L., Rabl, P., Blatt, R. & Zoller, P. Interfacing quantum-optical and solid-state qubits. Phys. Rev. Lett. 92 , 247902 (2004)

Andre, A. et al. A coherent all-electrical interface between polar molecules and mesoscopic superconducting resonators. Nature Phys. 2 , 636–642 (2006)

Recher, P., Sukhorukov, E. V. & Loss, D. Andreev tunneling, Coulomb blockade, and resonant transport of nonlocal spin-entangled electrons. Phys. Rev. B 63 , 165314 (2001)

Privman, V., Vagner, I. D. & Kventsel, G. Quantum computation in quantum-Hall systems. Phys. Lett. A 239 , 141–146 (1998)

ADS   MathSciNet   CAS   MATH   Google Scholar  

Smelyanskiy, V. N., Petukhov, A. G. & Osipov, V. V. Quantum computing on long-lived donor states of Li in Si. Phys. Rev. B 72 , 081304 (2005)

Tian, L. & Zoller, P. Coupled ion-nanomechanical systems. Phys. Rev. Lett. 93 , 266403 (2004)

Piermarocchi, C., Chen, P., Sham, L. J. & Steel, D. G. Optical RKKY interaction between charged semiconductor quantum dots. Phys. Rev. Lett. 89 , 167402 (2002)

Quinteiro, G. F., Fernandez-Rossier, J. & Piermarocchi, C. Long-range spin-qubit interaction mediated by microcavity polaritons. Phys. Rev. Lett. 97 , 097401 (2006)

Khitun, A., Ostroumov, R. & Wang, K. L. Spin-wave utilization in a quantum computer. Phys. Rev. A 64 , 062304 (2001)

Barnes, C. H. W., Shilton, J. M. & Robinson, A. M. Quantum computation using electrons trapped by surface acoustic waves. Phys. Rev. B 62 , 8410–8419 (2000)

Chang, D. E., Sørensen, A. S., Hemmer, P. R. & Lukin, M. D. Quantum optics with surface plasmons. Phys. Rev. Lett. 97 , 053002 (2006)

Raussendorf, R. & Harrington, J. Fault-tolerant quantum computation with high threshold in two dimensions. Phys. Rev. Lett. 98 , 190504 (2007)

Nayak, C., Simon, S. H., Stern, A., Freedman, M. & Das Sarma, S. Non-abelian anyons and topological quantum computation. Rev. Mod. Phys. 80 , 1083–1159 (2008)

Langer, C. et al. Long-lived qubit memory using atomic ions. Phys. Rev. Lett. 95 , 060502 (2005)

Knill, E. et al. Randomized benchmarking of quantum gates. Phys. Rev. A 77 , 012307 (2008)

Benhelm, J., Kirchmair, G., Roos, C. F. & Blatt, R. Towards fault-tolerant quantum computing with trapped ions. Nature Phys. 4 , 463–466 (2008)

Treutlein, P., Hommelhoff, P., Steinmetz, T., Hänsch, T. W. & Reichel, J. Coherence in microchip traps. Phys. Rev. Lett. 92 , 203005 (2004)

Ryan, C. A., Laforest, M. & Laflamme, R. Randomized benchmarking of single- and multi-qubit control in liquid-state NMR quantum information processing. New J. Phys. 11 , 013034 (2009)

Bertet, P. et al. Dephasing of a superconducting qubit induced by photon noise. Phys. Rev. Lett. 95 , 257002 (2005)

Emerson, J. et al. Symmetrized characterization of noisy quantum processes. Science 317 , 1893–1896 (2007)

Hanson, R. & Awschalom, D. D. Coherent manipulation of single spins in semiconductors. Nature 453 , 1043–1049 (2008)

Download references


We thank R. Hanson, M. D. Lukin, and W. D. Oliver for comments. We acknowledge support from NSF, EPSRC, QIP IRC, IARPA, ERC, the Leverhulme Trust, CREST-JST, DFG, BMBF and Landesstiftung BW. J.L.O’B. acknowledges a Royal Society Wolfson Merit Award.

Author Contributions All authors contributed to all aspects of this work.

Author information

Present address: Present address: HRL Laboratories, LLC, 3011 Malibu Canyon Road, Malibu, California 90265, USA.,

Authors and Affiliations

Edward L. Ginzton Laboratory, Stanford University, Stanford, California 94305-4088, USA ,

3. Physikalisches Institut, Universität Stuttgart, Pfaffenwaldring 57, D-70550, Germany ,

Institute for Quantum Computing,,

R. Laflamme

Department of Physics and Astronomy, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, N2L 3G1, Canada,

Perimeter Institute, 31 Caroline Street North, Waterloo, Ontario, N2L 2Y5, Canada ,

Nano Electronics Research Laboratories, NEC Corporation, Tsukuba, Ibaraki 305-8501, Japan ,

Y. Nakamura

The Institute of Physical and Chemical Research (RIKEN), Wako, Saitama 351-0198, Japan ,

University of Maryland Department of Physics,, Joint Quantum Institute,

National Institute of Standards and Technology, College Park, Maryland 20742, USA ,

H. H. Wills Physics Laboratory and Department of Electrical and Electronic Engineering, Centre for Quantum Photonics, University of Bristol, Merchant Venturers Building, Woodland Road, Bristol, BS8 1UB, UK,

J. L. O’Brien

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to J. L. O’Brien .

Ethics declarations

Competing interests.

The authors declare no competing financial interests.

PowerPoint slides

Powerpoint slide for fig. 1, powerpoint slide for fig. 2, powerpoint slide for fig. 3, powerpoint slide for fig. 4, powerpoint slide for fig. 5, rights and permissions.

Reprints and permissions

About this article

Cite this article.

Ladd, T., Jelezko, F., Laflamme, R. et al. Quantum computers. Nature 464 , 45–53 (2010).

Download citation

Issue Date : 04 March 2010


Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

This article is cited by

Thermal multiferroics in all-inorganic quasi-two-dimensional halide perovskites.

  • Xue-Zeng Lu
  • Hiroshi Kageyama

Nature Materials (2024)

Quantum entanglement of ions for light dark matter detection

  • Ryuichiro Kitano
  • Ryoto Takai

Journal of High Energy Physics (2024)

Quantum frequency conversion and single-photon detection with lithium niobate nanophotonic chips

  • Xufeng Jiao
  • Jian-Wei Pan

npj Quantum Information (2023)

A high-fidelity quantum matter-link between ion-trap microchip modules

  • W. K. Hensinger

Nature Communications (2023)

A photonic entanglement filter with Rydberg atoms

  • Gen-Sheng Ye

Nature Photonics (2023)

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

  • Explore articles by subject
  • Guide to authors
  • Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

research paper on computers

research paper on computers no longer supports Internet Explorer.

To browse and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

  •  We're Hiring!
  •  Help Center
  • Computer Science
  • Most Cited Papers
  • Most Downloaded Papers
  • Newest Papers
  • Save to Library
  • Last »
  • Artificial Intelligence Follow Following
  • Software Engineering Follow Following
  • Computer Vision Follow Following
  • Human Computer Interaction Follow Following
  • Machine Learning Follow Following
  • Data Mining Follow Following
  • Computer Graphics Follow Following
  • Distributed Computing Follow Following
  • Computer Networks Follow Following
  • Cloud Computing Follow Following

Enter the email address you signed up with and we'll email you a reset link.

  • Publishing
  •   We're Hiring!
  •   Help Center
  • Find new research papers in:
  • Health Sciences
  • Earth Sciences
  • Cognitive Science
  • Mathematics
  • Academia ©2024

U.S. flag

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

  • Publications
  • Account settings
  • Advanced Search
  • Journal List
  • Sensors (Basel)

Logo of sensors

Brain-Computer Interface: Advancement and Challenges

M. f. mridha.

1 Department of Computer Science and Engineering, Bangladesh University of Business and Technology, Dhaka 1216, Bangladesh; db.ude.tbub@zorif (M.F.M.); [email protected] (S.C.D.); moc.liamg@ibakmdm (M.M.K.); [email protected] (A.A.L.)

Sujoy Chandra Das

Muhammad mohsin kabir, aklima akter lima, md. rashedul islam.

2 Department of Computer Science and Engineering, University of Asia Pacific, Dhaka 1216, Bangladesh

Yutaka Watanobe

3 Department of Computer Science and Engineering, University of Aizu, Aizu-Wakamatsu 965-8580, Japan;

Associated Data

There is no statement regarding the data.

Brain-Computer Interface (BCI) is an advanced and multidisciplinary active research domain based on neuroscience, signal processing, biomedical sensors, hardware, etc. Since the last decades, several groundbreaking research has been conducted in this domain. Still, no comprehensive review that covers the BCI domain completely has been conducted yet. Hence, a comprehensive overview of the BCI domain is presented in this study. This study covers several applications of BCI and upholds the significance of this domain. Then, each element of BCI systems, including techniques, datasets, feature extraction methods, evaluation measurement matrices, existing BCI algorithms, and classifiers, are explained concisely. In addition, a brief overview of the technologies or hardware, mostly sensors used in BCI, is appended. Finally, the paper investigates several unsolved challenges of the BCI and explains them with possible solutions.

1. Introduction

The quest for direct communication between a person and a computer has always been an attractive topic for scientists and researchers. The Brain-Computer Interface (BCI) system has directly connected the human brain and the outside environment. The BCI is a real-time brain-machine interface that interacts with external parameters. The BCI system employs the user’s brain activity signals as a medium for communication between the person and the computer, translated into the required output. It enables users to operate external devices that are not controlled by peripheral nerves or muscles via brain activity.

BCI has always been a fascinating domain for researchers. Recently, it has become a charming area of scientific inquiry and has become a possible means of proving a direct connection between the brain and technology. Many research and development projects have implemented this concept, and it has also become one of the fastest expanding fields of scientific inquiry. Many scientists tried and applied various communication methods between humans and computers in different BCI forms. However, it has progressed from a simple concept in the early days of digital technology to extremely complex signal recognition, recording, and analysis techniques today. In 1929, Hans Berger [ 1 ] became the first person to record an Electroencephalogram (EEG) [ 2 ], which shows the electrical activity of the brain that is measured through the scalp of a human brain. The author tried it on a boy with a brain tumor; since then, EEG signals have been used clinically to identify brain disorders. Vidal [ 3 ] made the first effort to communicate between a human and a computer using EEG in 1973, coining the phrase “Brain-Computer Interface”. The author listed all of the components required to construct a functional BCI. He made an experiment room that was separated from the control and computer rooms. In the experiment room, three screens were required; the subject’s EEG was to be sent to an amplifier the size of an entire desk in the control area, including two more screens and a printer.

The concept of combining brains and technology has constantly stimulated people’s interest, and it has become a reality because of recent advancements in neurology and engineering, which have opened the pathway to repairing and possibly enhancing human physical and mental capacities. The sector flourishing the most based on BCI is considered the medical application sector. Cochlear implants [ 4 ] for the deaf and deep brain stimulation for Parkinson’s illness are examples of medical uses becoming more prevalent. In addition to these medical applications, security, lie detection, alertness monitoring, telepresence, gaming, education, art, and human enhancement are just a few uses for brain–computer interfaces (BCIs), also known as brain–machine interfaces or BMIs [ 5 ]. Every application based on BCI follows different approaches and methods. Each method has its own set of benefits and drawbacks. The degree to which a performance can be enhanced while minute-to-minute and day-to-day volatility are reduced is crucial for the future of BCI technology. Such advancements rely on the capacity to systematically evaluate and contrast different BCI techniques, allowing for the most promising approaches to be discovered. In addition, this versatility around BCI technologies in different sectors and their applications can seem so complex yet so structured. Most of the BCI applications follow a standard structure and system. This basic structure of BCI consists of signal acquisition, pre-processing, feature extraction, classification, and control of the devices. The signal acquisition paves the way to connecting a brain and a computer and to gathering knowledge from signals. The three parts of pre-processing, feature extraction, and classification are responsible for making the associated signal more usable. Lastly, control of the devices points out the primary motivation: to use the signals in an application, prosthetic, etc.

The outstanding compatibility of various methods and procedures in BCI systems demands extensive research. A few research studies on specific features of BCI have also been conducted. Given all of the excellent BCI research, a comprehensive survey is now necessary. Therefore, an extensive survey analysis was attempted and focused on nine review papers featured in this study. Most surveys, however, do not address contemporary trends and application as well as the purpose and limits of BCI methods. Now, an overview and comparisons of the known reviews of the literature on BCI are shown in Table 1 .

A summary of recent surveys/reviews on various BCI technologies, signals, algorithms, classifiers, etc.

Abiri, R. et al. [ 6 ] evaluated the current review on EEG-based various experimental paradigms used by BCI systems. For each experimental paradigm, the researchers experimented with different EEG decoding algorithms and classification methods. The researchers overviewed the paradigms such as Motor imagery paradigms, Body kinematics, Visual P300, Evoked potential, and Error related potential and the hybrid paradigms analyzed with the classification methods and their applications. Researchers have already faced some severe issues while exploring BCI paradigms, including training time and fatigue, signal processing, and novel decoders; shared control to supervisory control in closed-loop; etc. Tiwari, N. et al. [ 7 ] provided a complete assessment of the evolution of BCI and a fundamental introduction to brain functioning. An extensive comprehensive revision of the anatomy of the human brain, BCI, and its phases; the methods for extracting signals; and the algorithms for putting the extracted information to use was offered. The authors explained the steps of BCI, which consisted of signal acquisition, feature extraction, and signal classification. As the human brain is complex, human-generated thoughts are non-stationary, and generated signals are nonlinear. Thus, the challenging aspect is to develop a system to find deeper insights from the human brain; then, BCI application will perform better with these deeper insights. Vasiljevic, G.A.M. et al. [ 8 ] presented a Systematic Literature Review (SLR) conclusion of BCI games employing consumer-grade gadgets. The authors analyzed the collected data to provide a comprehensive picture of the existing reality and obstacles for HCI of BCI-based games utilizing consumer-grade equipment. According to the observations, numerous games with more straightforward commands were designed for research objectives, and there was a growing amount of more user-friendly BCI games, particularly for recreation. However, this study is limited to the process of search and classification. Martini, M.L. et al. [ 9 ] investigated existing BCI sensory modalities to convey perspectives as technology improves. The sensor element of a BCI circuit determines the quality of brain pattern recognition, and numerous sensor modalities are presently used for system applications, which are generally either electrode-based or functional neuroimaging-based. Sensors differed significantly in their inherent spatial and temporal capabilities along with practical considerations such as invasiveness, mobility, and maintenance. Bablani, A. et al. [ 10 ] examined brain reactions utilizing invasive and noninvasive acquisition techniques, which included electrocorticography (ECoG), electroencephalography (EEG), magnetoencephalography (MEG), and magnetic resonance imaging (MRI). For operating any application, such responses must be interpreted utilizing machine learning and pattern recognition technologies. A short analysis of the existing feature extraction techniques and classification algorithms applicable to brain data has been presented in this study.

Fleury, M. et al. [ 11 ] described various haptic interface paradigms, including SMR, P300, and SSSEP, and approaches for designing relevant haptic systems. The researchers found significant trends in utilizing haptics in BCIs and NF and evaluated various solutions. Haptic interfaces could improve productivity and could improve the relevance of feedback delivered, especially in motor restoration using the SMR paradigm. Torres, E.P. et al. [ 12 ] conducted an overview of relevant research literature from 2015 to 2020. It provides trends and a comparison of methods used in new implementations from a BCI perspective. An explanation of datasets, emotion elicitation methods, feature extraction and selection, classification algorithms, and performance evaluation is presented. Zhang, X. et al. [ 13 ] discussed the classification of noninvasive brain signals and the fundamentals of deep learning algorithms. This study significantly gives an overview of brain signals and deep learning approaches to enable users to understand BCI research. The prominent deep learning techniques and cutting-edge models for brain signals are presented in this paper, together with specific ideas for selecting the best deep learning models. Gu, X. et al. [ 14 ] investigated the most current research on EEG signal detection technologies and computational intelligence methodologies in BCI systems that filled in the loopholes in the five-year systematic review (2015–2019). The authors demonstrated sophisticated signal detecting and augmentation technologies for collecting and cleaning EEG signals. The researchers also exhibited computational intelligence techniques, such as interpretable fuzzy models, transfer learning, deep learning, and combinations for monitoring, maintaining, or tracking human cognitive states and the results of operations in typical applications.

The study necessitated a compendium of scholarly studies covering 1970 to 2021 since we analyze BCI in detail in this literature review. We specialized in the empirical literature on BCI from 2000 to 2021. For historical purposes, such as the invention of BCI systems and their techniques, we selected some publications before 2000. Kitchenham [ 15 , 16 ] established the Systematic Literature Review (SLR) method, which is applied in the research and comprises three phases: organizing, executing, and documenting the review. The SLR methodologies attempted to address all possible questions that could arise as the current research progresses. The recent study’s purpose is to examine the findings of numerous key research areas. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines were used to put together the essential materials for this study, which consists of four parts: identification, scanning, eligibility testing, and inclusion. We gathered 577 papers from a variety of sources and weeded out duplicates and similar articles. Finally, we carefully chose 361 articles and sources for monitoring and review. The PRISMA process is presented in Figure 1 .

An external file that holds a picture, illustration, etc.
Object name is sensors-21-05746-g001.jpg

The PRISMA process that is followed in this article.

However, this research looks at the present challenges and difficulties in this BCI field. Furthermore, this study generates ideas and suggestions for future research subjects. The following are the research’s total contributions:

  • The paper explicitly illustrates Brain-Computer Interface’s (BCI) present, past, and future trends and technologies.
  • The paper presents a taxonomy of BCI and elaborates on the few traditional BCI systems with workflow and architectural concepts.
  • The paper investigates some BCI tools and datasets. The datasets are also classified on different BCI research domains.
  • In addition, the paper demonstrates the application of BCI, explores a few unsolved challenges, and analyzes the opportunities.

After reading this section, one should understand BCI and how to get started with it. Our motivation to work with BCI started from a desire to learn more about this domain. Furthermore, the BCI has a bright future ahead of it, as it has a lot to offer in the medical field and in everyday life. BCI can change one’s incapability and can make life and work easy, as detailed in the following section. The applications, problems, future, and social consequences of BCI have also fueled our enthusiasm for this research.

The remainder of the paper is constructed as follows. The motivation of this work and diverse applications of BCI systems are illustrated in Section 2 . Section 3 describes the structure of BCI and briefly reviews the most popular techniques of BCI. In Section 5 , different categories of datasets available publicly are displayed. In Section 7 , the most widely used methods for signal enhancement and feature extraction of BCI are discussed. The most commonly known classifiers are reviewed in Section 8 . A broad discussion on the evaluation metrics for BCI is given in Section 9 . The challenges faced most commonly during the BCI process are reviewed in Section 10 . Lastly, this paper provides a conclusion in Section 11 .

2. Applications of BCI

BCIs may be used for various purposes and the application determines the design of a BCI. According to Nijholt [ 17 ], applications based on BCI have two methods of usability. One can command whether the other one can be observed or monitored. The majority of command applications concentrate on manipulating brain impulses using electrodes to control an external device. On the other hand, applications that involve observation focus on recognizing a subject’s mental and emotional state to behave appropriately depending on their surroundings. Some applications of BCI [ 18 ] based on usability are described below:

2.1. Biomedical Applications

The majority of BCI integrations and research have been focused on medical applications, with many BCIs aiming to replace or restore Central Nervous System (CNS) functioning lost with sickness or by accident. Other BCIs are more narrowly targeted. In diagnostic applications, on treatment and motor rehabilitation following CNS disease or trauma, BCIs for biological purposes are also employed in affective application domains. Biomedical technologies and applications can minimize extended periods of sickness, can provide supervision and protection by empowering persons with mobility difficulties, and can support their rehabilitation. The necessity to build accurate technology that can cope with potentially abnormal brain responses that might occur due to diseases such as brain stroke is a significant challenge in developing such platforms [ 19 ]. The following subsections go through each of these applications in further detail.

2.1.1. Substitute to CNS

These substitution means that it can repair or replace CNS functioning lost due to diseases such as paralysis and spinal cord injury due to stroke or trauma. In addition, due to changed brain functions, individuals with such illnesses might suffer and developing such technology can be difficult. Myoelectrics, known as a motor action potential, which captures electrical impulses in muscles, is now used in several robotic prosthetics. Bousseta, R. et al. [ 20 ] provided an experimental technology for controlling the movement of a robotic prosthetic arm with mental imagery and using cognitive tasks, which can move in four directions like left, right, up, and down.

2.1.2. Assessment and Diagnosis

The usage of BCIs in a clinical context can also help with assessment and diagnosis. Perales [ 21 ] suggested a BCI for assessing the attention of youngsters with cerebral palsy while playing games. Another research [ 22 ] looked into using BCI to capture EEG characteristics as a tool for diagnosing schizophrenia. There are also various diagnostic methods such as the detection of brain tumors [ 23 ], the identification of breast cancer [ 24 ], parkinson’s disease [ 25 ] etc. Diagnoses of several diseases in children including epilepsy, neurodegenerative disorders, motor disabilities, inattentiveness, or different types of ADHD [ 26 ] are possible. Assessment and diagnosis technologies are essential to patient well-being. Their functioning must be fine-tuned to guarantee that they are safe, acceptable, and accurate to industry standards.

2.1.3. Therapy or Rehabilitation

BCI is being used in therapeutic applications besides neurological application and prosthetics nowadays. Among the many applications, post-stroke motor rehabilitation shows promising results using BCI. Stroke is a disease that causes long-term disability to the human body and hampers all kinds of motor or vigorous activity due to an impediment of blood flow. Stroke rehabilitation application has promised to aid these activities or user imaginations through a robot or other types of machinery [ 27 , 28 , 29 ]. Some other applications treat neurological disorders such as Parkinson’s disease (PD), cluster headaches, tinnitus, etc. Deep Brain Stimulation (DBS) is an established treatment for PD as it delivers electrical impulses to a targeted area of the brain responsible for the symptoms [ 30 ]. Some stimulation BCI devices are used to process calmness during migraine attacks and cluster headaches. Lastly, a CNS disorder known as tinnitus is also in development to provide treatment by identifying brain patterns that are changed due to the disease [ 31 ]. Lastly, treatment for auditory verbal hallucinations (AVHs), best known as schizophrenia, is a possibility besides diagnosis [ 32 , 33 ].

2.1.4. Affective Computing

Users’ emotions and state of mind are observed in affective computing BCIs, with the possibility of altering their surrounding environment to improve or change that emotion. Ehrlich, S. et al. [ 34 ] created a closed-loop system in which music is generated and then replayed to listeners based on their emotional state. Human emotional states and sensory connections can be studied with a device that is related to BCI system. Patients suffering neurological diseases also can benefit from affective computing to help them convey their feelings to others [ 35 ].

2.2. Non-Biomedical Applications

BCI technologies have shown economic promise in recent years, notably in the field of non-biomedical applications. Most of these applications consist of entertaining applications, games, and emotional computation. In comparison, researchers focus on robustness and high efficiency in medical and military applications, and innovations targeted at leisure or lifestyle demand a greater emphasis on enjoyment and social elements. The most challenging aspect of this entertainment application is that it must be a user favorite to be commercially successful. As an example, some of the most popular forms of amusement are as follows:

2.2.1. Gaming

BCIs focused mainly on the gaming sector have grown in importance as a research topic. However, gaming BCIs are currently a poor substitute for standard game control methods [ 36 ]. BCI in gaming is an area where further research is needed to make games more user-friendly. In some cases, EEG data make BCI games more utilizable and increase engagement, and the system tracks each player’s enthusiasm level and activates dynamic difficulty adjustment (DDA) when the players’ excitement drops [ 37 ]. When developing such systems, fine-tuning the algorithms that regulate the game’s behavior is a big challenge. Some other games are based on BCI, as it is not visually intense and the graphics are not compatible with the recent generation. With setbacks, there is an engaging future for an Adaptation of P300 based Brain-Computer Interface for Gaming [ 38 ], which is gaining more popularity as these are very flexible to play.

2.2.2. Industry

EEG-based BCIs can also be used in industrial robotics, increasing worker safety by keeping people away from potentially demanding jobs. These technologies could substitute the time-consuming button and joystick systems used to teach robots in industrial applications; can detect when a person is too tired or ill to operate the machinery; and can take the necessary precautions to avoid injury, such as stopping the machinery [ 38 ].

2.2.3. Artistic Application

The four types of artistic applications recognized by BCIs are passive, selective, direct, and collaborative. Passive artistic BCIs need not require active user input to use the user’s brain activity to determine which pre-programmed responses to produce. Every user has had some limited control over the process within selective systems. Still, they will never be in charge of the creative product. Direct artistic BCIs provide users with far more flexibility, generally allowing them to choose items from extensive menus, such as brush type and managing brush stroke movements [ 39 ]. Lastly, the collaborative system is controlled by different users [ 40 ].

2.2.4. Transport

BCI is used in transportation monitoring which tracks awareness to assess driver weariness and to enhance airline pilot performances. In the BCI system, mistakes can be costly regarding lives and monetary obligations on the entities involved when such technologies are utilized in critical applications [ 41 , 42 ].

3. Structure of BCI

The BCI system operates with a closed-loop system. Every action taken by the user is met with some feedback. For example, an imagined hand movement might result in a command that causes a robotic arm to move. This simple movement of this arm needs a lot of processes inside it. It starts from the brain, which is one of our body’s most extensive and most complicated organs. It is made up of billions of nerves that link billions of synapses to communicate. The processes from taking signals from the human brain to transforming into a workable command are shown in Figure 2 and described below:

  • Signal acquisition: In the case of BCI, it is a process of taking samples of signals that measure the brain activity and turning them into commands that can control a virtual or real-world application. The various techniques of BCI for signal acquisition are described later.
  • Pre-processing: After the signal acquisition, the pre-processing of signals is needed. In most cases, the collected signals from the brain are noisy and impaired with artifacts. This step helps to clean this noise and artifacts with different methods and filtering. That is why it is named signal enhancement.
  • Feature extraction: The next stage is feature extraction, which involves analyzing the signal and extracting data. As the brain activity signal is complicated, it is hard to extract useful information just by analyzing it. It is thus necessary to employ processing algorithms that enable the extraction of features of a brain, such as a person’s purpose.
  • Classification: The next step is to apply classification techniques to the signal, free of artifacts. The classification aids in determining the type of mental task the person is performing or the person’s command.
  • Control of devices: The classification step sends a command to the feedback device or application. It may be a computer, for example, where the signal is used to move a cursor, or a robotic arm, where the signal is utilized to move the arm.

An external file that holds a picture, illustration, etc.
Object name is sensors-21-05746-g002.jpg

Basic architecture of a BCI system.

The basic architecture of the BCI system was explained in the preceding section. It prompts us to investigate the classification of BCI system. Based upon various techniques, BCI system is classified. The BCI techniques are discussed in following parts.

From the above Figure 3 , we can classify BCI from different aspects such as dependability, invasiveness, and autonomy.

  • Dependability: BCI can be classified as dependent or independent. Dependent BCIs necessitate certain types of motor control from the operator or healthy subjects, such as gaze control. On the other hand, independent BCIs do not enable the individual to exert any form of motor control; this type of BCI is appropriate for stroke patients or seriously disabled patients.
  • Invasiveness: BCI is also classified into three types according to invasiveness: invasive, partially invasive, and non-invasive. Invasive BCIs are by far the most accurate as they are implanted directly into the cortex, allowing researchers to monitor the activity of every neuron. Invasive varieties of BCI are inserted directly into the brain throughout neurosurgery. There are two types of invasive BCIs: single unit BCIs, which detect signals from a single place of brain cells, and multi-unit BCIs, which detect signals from several areas. Semi-invasive BCIs use Electrocorticography (ECoG), a kind of signal platform that enables electrodes to be placed on the attainable edge of the brain to detect electrical impulses originating from the cerebral cortex. Although this procedure is less intrusive, it still necessitates a surgical opening in the brain. Noninvasive BCIs use external sensing rather than brain implants. Electroencephalography (EEG), Magnetoencephalography (MEG), Positron emission tomography (PET), Functional magnetic resonance imaging (fMRI), and Functional near-infrared spectroscopy (fNIRS) are all noninvasive techniques used it to analyze the brain. However, because of the low cost and portability of the gear, EEG is the most commonly used.
  • Autonomy: BCI can operate either in a synchronous or asynchronous manner. Time-dependent or time-independent interactions between the user and system are possible. The system is known as synchronous BCI if the interaction is carried out within a particular amount of time in response to a cue supplied by the system. In asynchronous BCI, the subject can create a mental task at a certain time to engage with the system. Synchronous BCIs are less user-friendly than asynchronous BCIs; however, designing one is substantially easier than developing an asynchronous BCI.

An external file that holds a picture, illustration, etc.
Object name is sensors-21-05746-g003.jpg

The classification/taxonomy of the BCI system.

As the motive of this research work is to focus on advancements of BCI, the most advanced and mostly used techniques that is based on invasiveness are described in the following part. Based on invasiveness, BCI is classified into three categories that are more familiar. In the consequent section, we address these three categories and describe them elaborately.

3.1. Invasive

The types of BCI that are invasive are inserted directly into the brain with neurosurgery. Invasive BCIs seem to be the most accurate even though they are implanted directly into the cortex as it is allowed to track every neuron’s action. Invasive BCI also has two units rather than parts. The first unit is single-unit BCIs that detect signals from a single location of brain cells, whereas multi-unit BCIs detect numerous areas, the second unit [ 43 ]. However, the neurosurgery treatment has various flaws, such as the possibility of scar tissue formation. The body responds to the foreign object by forming a scar around the electrodes, leading the signal to deteriorate. Since neurosurgery is a dangerous and costly procedure, invasive BCI is mainly used on blind and paralyzed patients.

3.2. Partially Invasive

Although this approach is not as intrusive, it still involves brain surgery. Electrocorticography (ECoG) is a sort of partially invasive BCI monitoring system that places electrodes in the cortex surface of the brain to produce signals with electrical activity. For example, blinking allows your brain to discharge electrical activity. When investigating signals, though, these involuntary actions are generally not of interest since they are in the way of what we search for. It is a form of noise. ECoGs are less considered with noise than non-invasive BCI, making interpretation easier [ 44 ].

Electrocorticography (ECoG)

Electrocorticography (ECoG) [ 45 ] is an partially invasive method that measures the brain’s electrical activity. In another sense, the participant’s skull must be evacuated, and the electrodes must be placed right at the brain’s service. Consequently, this electrode is located on the skull. The particular resolution of the recorded signals is considerably better than EEG. The signal-to-noise ratio is superior compared with the closer proximity to cerebral activity. Furthermore, motion artifacts such as blinks and eye movement have a significantly lower impact on ECoG signals. However, ECoG would only be helpful in the accessible brain area and is close to impossible to utilize outside of a surgical setting [ 46 ].

3.3. Noninvasive

Noninvasive neuroimaging technologies have also been used as interfaces in human research. Noninvasive EEG-based BCIs account for the vast bulk of published BCI research. EEG-based noninvasive technologies and interfaces have been employed in a considerably more comprehensive range of applications. Noninvasive apps and technologies are becoming increasingly popular in recent years since they do not require any brain surgery. In the noninvasive mode, a headpiece or helmet-like electrode is utilized outside the skull to measure the signal by causing electrical activity in the brain. There are some well-known and widely used ways for measuring these electrical activity or potentials, such as Electroencephalography (EEG), Magnetoencephalography (MEG), Functional Magnetic Resonance Imaging (fMRI), Facial Near Infrared Spectroscopy (fNIRS), and Positron Emission Tomography (PET). An elaborate description of BCI techniques is given below:

3.3.1. Electroencephalography (EEG)

EG monitors electrical activity in the scalp generated by activating a few of the brain’s neurons. Several electrodes implanted on the scalp directly, mainly on the cortex, are often used to record these electrical activities quickly. For its excellent temporal resolution, ease of use, safety, and affordability, EEG is the most used technology for capturing brain activity. Active electrodes and passive electrodes are indeed the two types of electrodes that can be utilized. Active electrodes usually feature an integrated amplifier, whereas passive electrodes require an external amplifier to magnify the detected signals. The prime objective of implementing either embedded or external amplifiers is to lessen the impact of background noise and other signal weaknesses caused by cable movement. One of the issues with EEG is that it necessitates the use of gel or saline solutions to lower the resistance of skin-electrode contact. Furthermore, the signal quality is poor, and it is altered by background noise. The International 10–20 system [ 47 ] is often used to implant electrodes over the scalp surface for recording purposes. The electrical activities across various frequency bands are used to describe EEG in general.

3.3.2. Magnetoencephalography (MEG)

The magnetic fields created by current flow in the brain are measured using MEG (Magnetoencephalography). Electric fields have significantly more interrupted travel via the skull than magnetic fields, therefore it has superior spatial resolution than EEG. A functional neuroimaging technique is applied to measure and evaluate the brain’s magnetic field. MEG operates on the outside of the head and is now a part of the clinical treatment regularly. David Choen [ 48 , 49 ] was the first to invent it in 1968 by utilizing a conduction copper detector inside a shielded chamber to reduce background noise. Improved MEG signals have recently been produced using more sensitive sensors such as superconducting quantum interference devices (SQUID) [ 50 ]. MEG has become significant, especially for patients with epilepsy and brain tumors. It may aid in detecting regions of the brain with average function in individuals with epilepsy, tumors, or other mass lesions. MEG operates with magnetic waves rather than electrical waves so that it could contribute additional information to EEG. MEG is also capable of capturing signals with high temporal and spatial resolution. Therefore, to detect cerebral activity that creates tiny magnetic fields the scanners must be closer to the brain’s surface. As a result, specific sensors are required for MEG, such as superconducting quantum interference (SQUID) sensors [ 51 ].

3.3.3. Functional Magnetic Resonance Imaging (fMRI)

Noninvasive functional magnetic resonance imaging (fMRI) is used to evaluate the fluctuation in blood oxygen levels throughout brain activities. fMRI has an excellent spatial resolution, which makes it ideal for identifying active areas of the brain [ 52 ]. The time resolution of fMRI is comparatively low, ranging from 1 to 2 s [ 53 ]. It also has low resolution when it comes to head movements, which could result in artifacts. In the 1990s, functional magnetic resonance imaging (fMRI) was created. It is a noninvasive and safe technology that does not include the use of radiation, is simple to use, and has great spatial and temporal resolution. Hemoglobin in capillary red blood cells in the brain transports oxygen to the neurons. As a result of the increased demand for oxygen, blood flow increases. If haemoglobin is oxygenated, its magnetic properties vary. The MRI equipment, which is a cylindrical tube with a strong electromagnet, can determine which regions of the brain are activated because of this difference. That is how fMRI works. There is also a specific application or software known as diffusion MRI, which generates images from the data or results that use water molecules’ diffusion. Diffusion-weighted and diffusion tensor imaging (DWI/DTI) facilitates this exploration of the microarchitecture of the brain. Diffusion-weighted magnetic resonance imaging (DWI or DW-MRI) imaging renders picture variation depending on variances in the degree of diffusion of water particles inside the brain. Diffusion depicts the stochastic thermic mobility of particles. Diffusion inside the brain is defined by several agents, including representing particles beneath study, the temperature, and the microenvironmental structure in which the diffusion occurs [ 54 ]. Diffusion tensor imaging (DTI) investigates the three-dimensional form of the diffusion, also recognized as diffusion tensor. It is a powerful MRI modality that produces directional knowledge about the water motion in a voxel. It exhibits noninvasively microscopic tissue features that surpass the ability of any other imaging methods [ 55 ].

3.3.4. Functional Near-Infrared Spectroscopy (fNIRS)

The infrared radiation is projected into the brain using fNIRS equipment [ 53 , 56 ] to monitor improvements in specific wavelengths as the light is reflected. fNIRS often detects changes in regional blood volume and oxygenation. When a particular area of the brain works, it requires additional oxygen, which is given to the neurons via capillary red blood cells—the increased blood flow in the brain areas that would be most active at a given time. fMRI is a technique that monitors variations in oxygen levels caused by various activities. As a result, images with a high spatial resolution (1 cm) but lower temporal resolution (>2–5 s) could be obtained, comparable with standard functional magnetic resonance imaging.

3.3.5. Positron Emission Tomography (PET)

PET (positron emission tomography) is a sophisticated imaging tool for examining brain activities in real-time. It enables noninvasive measurement of cerebral blood flow, metabolism, and receptor binding in the brain. Due to the relatively high prices and complexity of the accompanying infrastructure, including cyclotrons, PET scanners, and radio chemistry laboratories, PET was previously only used in research. PET has been widely employed in clinical neurology in recent years due to technological improvements and the proliferation of PET scanners to better our understanding of disease etiology, to help in diagnosis, and to monitor disease progression and response to therapy [ 57 ]. PET medications such as radiolabeled choline, fluciclovine (18F-FACBC), and compounds targeting prostate-specific membrane antigen are now being researched and explored to improve noninvasive prostate cancer localization diagnostic performance [ 58 ].

4. Brain Control Signals

The brain-computer interface (BCI) is based on signal amplification that comes directly from the brain. Several of these signals are simple to extract, while others are more difficult and require additional preprocessing [ 53 ]. These control signals can be classified into one of three groups: (1) evoked signals, (2) spontaneous signals, and (3) hybrid signals. A detailed overview of the three categories is given below. The control signals classification is shown in Figure 4 .

An external file that holds a picture, illustration, etc.
Object name is sensors-21-05746-g004.jpg

The basic architecture of BCI control signals.

4.1. Visual Evoked Potentials

Electrical potentials evoked by short visual stimuli are known as VEPs. The visual cortex’s potentials are monitored, and the waveforms are derived from the EEG. VEPs are generally used to assess the visual pathways from the eye to the brain’s visual cortex. Middendorf et al. published a procedure for measuring the position of the user’s gaze using VEPs in 2000 [ 59 ]. The user is confronted with a screen that displays several virtual buttons that flash at varied rates. The frequency of the photic driving reflex over the user’s visual brain is determined after the user focuses their gaze on a button. Whenever the frequency of a shown button equals the frequency of the user, the system concludes that the user wants to pick it. Steady-State Evoked Potentials (SSEP) and P300 are two of the most well-evoked signals. External stimulation is required for evoked signals that can be unpleasant, awkward, and exhausting for the individual.

4.1.1. Steady-State Evoked Potential (SSEP)

SSEP signals are produced when a patient experiences periodic stimuli such as a flickering picture, modulated sound, or even vibrations [ 60 , 61 ]. The strength of the EEG signal in the brain must grow to meet the stimulus frequency. Signals in many brain locations are observed in terms of the sensory process. SSEP signals of different forms, such as steady-state visual potentials (SSVEPs), somatosensory SSEP, and auditory SSEP, are found. SSVEP is widely used in a variety of applications. These are normal brain reactions to repeating stimuli, which vary depending on the frequency with which they are presented. Although there are instances of BCI paradigms utilizing somatosensory (SSSEP) or auditory (SSAEP) stimuli, they are generally induced using visual stimuli (steady-state visually evoked potentials, SSVEP) [ 62 ].

4.1.2. P300 Evoked Potentials (P300)

The peaks in an EEG generated by infrequent visual, auditory, or somatosensory inputs are known as P300 evoked potentials. Without the need for training to use P300-based BCI systems. A matrix of symbols, in which selection is dependent on the participant’s gaze, is a prominent use of P300-based BCI systems. Such a signal is typically produced using an “odd-ball” paradigm. The user is asked to respond to a random succession of stimuli, which is less frequent than others [ 63 ]. The P300-based EEG waves are triggered when this unusual stimulus is significant to the person. P300 does not reasonably require any subject training, although, it does need repetitive stimulation, which may tire the subject and may cause inconsistencies.

4.2. Spontaneous Signals

With no external cues, the person produces random signals willingly. These signals are produced without any external stimuli (somatosensory, aural, or visual). Motor and sensorimotor rhythms, Slow Cortical Potentials (SCPs), and non-motor cognitive signals are some of the most prominent spontaneous signals [ 53 ].

4.2.1. Motor and Sensorimotor Rhythms

Motor activities are linked to motor and sensorimotor rhythms. Sensorimotor rhythms are rhythmic oscillations in electrophysiological brain activity in the mu (Rolandic band, 7–13 Hz) and beta (13–30 Hz) frequencies. Motor imagery is the process of converting a participant’s motor intentions into control signals employing motor imagery conditions [ 64 ]. The left-hand motion, in an instance, may result in EEG signals in the and rhythms and a decrease in certain motor cortex areas (8–12 Hz) and (18–26 Hz). Depending on the motor imagery rhythms, various applications can be used such as controlling a mouse or playing a game.

4.2.2. Slow Cortical Potentials (SCP)

SCP is indeed an EEG signal with a frequency less than 1 Hz [ 65 ]. It is a low-frequency potential observed in the frontal and central portions of the cortex and depolarization level variations throughout the cortical dendrites. SCP is a highly gradual change in brain activity, either positive or negative, that can only last milliseconds to several seconds. Through operant conditioning, the subject can control the movement of such signals. As a result, extensive training may be required in addition to that needed for motor rhythms. Many studies no longer choose SCP, and motor and sensorimotor rhythms have taken their place.

4.2.3. Non-Motor Cognitive Tasks

Cognitive objectives are utilized to drive the BCI in non-motor cognitive tasks. Several tasks, such as musical imagination, visual counting, mental rotation, and mathematical computation, might be completed [ 66 ]. Penny, W.D. et al. [ 67 ] used a pattern classifier with unclear parameters. The individual performed simple subtraction in one of their non-motor cognitive activities.

4.3. Hybrid Signals

The term “hybrid signals” refers to the utilization of a mixture of brain-generated signals for control. As a result, instead of measuring and using only one signal in the BCI system, a mix of signals is used. The fundamental goal of using two or more types of brain signals as input to a BCI system is to increase dependability while avoiding the drawbacks of each signal type [ 68 ].

Some research is addressed that the types of brain signals are classified into two categories [ 10 ]. These are event-related potentials and evoked brain potential. Three varieties are organized for evoked brain potential: Visual Evoked Potential (VEP), Tactile Evoked Potential (TEP), and Auditory Evoked Potential (AEP) [ 69 ].

While analyzing the literature on BCI systems, we discovered various often used datasets that researchers used while implementing these techniques. In terms of the research, EEG is now the most frequent method for collecting brain data in BCI. As this is a noninvasive method and has convenient handling for most datasets, an EEG signal is used. However, for a variety of reasons, EEG does not provide a comprehensive method of data collection. It needs a variety of fixed things to acquire the data. Firstly, the signal must be acquired and stored by some subject, participants, or patients. It is unsuitable when only one subject requires the same arrangement as multiple subjects to obtain data. After the subjects are prepared, the electrodes (a gear mounted on the scalp) are attached to the individuals to capture and measure data. This data collection method lasted for several sessions, with a particular recording period determined by the work’s purpose. The saved data in these sessions and recordings are primarily brain signals measured by a brain’s action on a sure thing, such as a video or a picture. EEG signals differ from one participant to the next and from one session to the next. In this section, the datasets as well as the subjects and electrodes, channels, and sessions are described. The explanation is tabulated in Table 2 , Table 3 , Table 4 , Table 5 , Table 6 , Table 7 and Table 8 . In Table 2 , some popular motor imagery datasets are illustrated. The most beneficial option for creating BCIs is motor imagery (MI) impulses captured via EEG, which offers a great degree of mobility. It enables people with motor disabilities to communicate with the device by envisioning motor movements without any external stimuli generated from the motor cortex. A few datasets based on error-related potentials (ErrPs) are exhibited in Table 3 . ErrPs is an EEG dataset that utilizes a P300-based BCI speller to boost the performance of BCIs. Detecting and fixing errors of the neuronal signature of a user’s knowledge linked to a brain pattern is known as error-related potentials (ErrPs). Affective computing improves human–machine communication by identifying human emotions. Some mostly used emotion recognition datasets are shown in Table 4 . Various EEG-based BCI devices can detect the user’s emotional states to make contact effortless, more useable, and practical. The emotions extracted in emotion-recognition datasets are valence, arousal, calm, positive, exciting, happy, sad, neutral, and fear. In addition, it is certainly clear by now that brain signals or memory are a mixed emotion. The part where all of these mixed emotions are gathered from different body parts is known as a miscellaneous part of the brain. Therefore, miscellaneous datasets include memory signals, brain images, brain signals, etc. Some miscellaneous datasets are represented in Table 5 . In EEG-based BCI, the signals can detect eye movement such as eye blinks, eye states, etc. The BCI datasets of eye blinks or movements include voluntary and involuntary eye states, blinks, and activities are illustrated in Table 6 . Subsequently, the electrical response in the brain to a specific motor or cognitive event such as a stimulus is known as an event-related potential (ERP). An unwanted sound, a sparking light, or a blinking eye can be an example of a stimulus. BCI utilizing ERPs attempts to track attention, weariness, and the brain’s reaction to this event-related stimulus. Table 7 is encapsulated with popular ERP datasets around. Moreover, the visual information-processing mechanism in the brain is reflected in Visually Evoked Potentials (VEPs). Flashing objects in the form of shifting colors or a reversing grid are frequent visual stimulators. The CRT/LCD monitor or flash tube/infrared diode (LED) is utilized for stimulus display in VEP-based BCIs. Frequently used VEP-based datasets with these utilized objects are represented in Table 8 .

A table of different types of motor imagery datasets of BCI.

A table of different types of Error-Related Potentials (ErrP) dataset of BCI.

A table of different types emotion recognition dataset of BCI.

A table of different types of miscellaneous datasets.

A table of different types of eye-blink or movement datasets in BCI.

A table of different types Event-Related Potential (ERP) datasets in BCI. These datasets are collected from [ 229 ].

A table of different types of Visually Evoked Potential (VEP) datasets in BCI. These datasets are collected from [ 229 ].

However, the dataset covers information recorded from the beginning of BCI. To extract information from datasets, feature extraction methods are necessary, which is reviewed in the following section.

6. Signal Preprocessing and Signal Enhancement

In most situations, the signal or data measured or extracted from datasets are filled with noise. With a natural human activity such as eye blinks and heartbeats, the collected data might become noisy. These noises are eliminated during the pre-processing step to produce clean data that may subsequently process the feature extraction and classification. This pre-processing unit is also known as signal enhancement since it cleans the signal in BCI. Some methods are used for signal enhancement in the BCI system, and these are explained elaborately in the following subsections.

6.1. Independent Component Analysis (ICA)

The noises and EEG signals are isolated in ICA by treating them as distinct entities. Furthermore, the data are retained during the removal of noises. This method divides the EEG data into spatially fixed and temporally independent components. In the case of computing and noise demonstrable, the ICA shows more efficiency [ 256 ].

6.2. Common Average Reference (CAR)

It is most commonly employed as a basic dimensionality reduction technique. This approach decreases noise across all recorded channels, but this does not address channel-specific noise and may inject noise into an otherwise clean channel. It is a spatial filter that can be thought of as the subtraction of shared EEG activity, retaining only the idle action of each EEG particular electrode [ 256 ].

6.3. Adaptive Filters

The adaptive filter is a computational device for mathematical processes. It connects the adaptive filter’s input/output signals iteratively. There are filter coefficients that are self-adjusted utilizing an adaptive algorithm. It works by altering signal properties depending on the characteristics of the signals under investigation [ 257 ].

6.4. Principal Component Analysis (PCA)

PCA is a technique for detecting patterns in data represented by a rotation of the coordinate axes. These axes are not aligned with single time points, but they depict a signal pattern with linear combinations of sets of time points. PCA keeps the axes orthogonal while rotating them to maximize variance along the first axis. It reduces feature dimensions and aids in data classification by completing ranking. In comparison with ICA, PCA compresses separate data better whether noise is eliminated with it or not [ 258 ].

6.5. Surface Laplacian (SL)

SL refers to a method of displaying EEG data with a high spatial resolution. SL can be generated using any EEG recording reference scheme as their estimates are reference-free. Based on the volume conductor’s exterior shape, it is a general estimate of the current density entering or exiting the scalp through the skull, and it does not require volume conduction details. The advantage of SL is that it improves the spatial resolution of the EEG signal. However, SL seems not to demand additional operative neuroanatomy premises as it is sensitive to spline patterns and artifacts [ 259 ].

6.6. Signal De-Noising

Artefacts frequently corrupt EEG signals taken from brain. These artifacts must be removed from EEG data to obtain valuable information from it. The technique of eliminating sounds or artefacts from EEG signals is known as de-noising [ 260 ]. Some de-noising methods are given below:

  • Wavelet de-noising and thresholding: The multi-resolution analysis is used to transfer the EEG signal to the discrete wavelet domain. The contrasting or adaptive threshold level is used to reduce particular coefficients associated with the noise signal [ 261 ]. Shorter coefficients would tend to define noise characteristics throughout time and scale in a well-matched wavelet representation. In contrast, threshold selection is one of the most critical aspects of successful wavelet de-noising. Thresholding can isolate the signal from the noise in this case; hence, thresholding approaches come in several shapes and sizes. All coefficients underneath a predetermined threshold value are set to zero in hard thresholding. Soft thresholding is a method of reducing the value of the remaining coefficients by a factor of two [ 262 ].
  • Empirical mode decomposition (EMD): It is a signal analysis algorithm for multivariate signals. It breaks the signal down into a series of frequency and amplitude-regulated zero-mean signals, widely known as intrinsic mode functions (IMFs). Wavelet decomposition, which decomposes a signal into multiple numbers of Intrinsic Mode Functions (IMFs), is compared by EMD. It decomposes these IMFs using a shifting method. An IMF is a function with a single maximum between zero crossings and a mean value of zero. It produces a residue after degrading IMFs. These IMFs are sufficient to characterize a signal [ 263 ].

Most of our datasets mentioned in the previous section are a part of various BCI paradigms and follow these signal enhancement techniques as well. The motor imagery datasets represent paradigms such as sensorimotor activity or rhythms. In addition, error-related potentials datasets and datasets such as event-related potentials or visually evoke potentials signify their own BCI paradigm. Some other paradigms, such as overt attention, eye movement, miscellaneous, and emotion recognition, identify their datasets. Indeed, these paradigms become bigger in number as the measurement of different brain movements and emotions are attempted. More than 100 BCI designs are required to use signal enhancement techniques to extract features from the signal. In comparison, Reference [ 264 ] shows that 32% of BCI designs use surface Laplacian (SL) to extract features, principal component analysis (PCA) or independent component analysis (ICA) was used in 22%, and common spatial patterns (CSP) and common average referencing (CAR) techniques are used in 14% and 11%, respectively.

7. Feature Extraction

Now, it is necessary to understand what the features represent, their qualities, and how to use them for a BCI system to select the best appropriate classifier. A classification system’s accuracy or efficiency is primarily determined by the feature(s) of the samples to be categorized [ 265 ]; therefore, feature extraction has been crucial stage in BCI. The majority of noninvasive BCI devices use neuroimaging techniques such as MEG and MRI. However, EEG is the most widely utilized method, owing to its high temporal resolution and inexpensive cost [ 266 ]. The EEG signal feature extraction method is one of the essential components of a BCI system because of its involvement in successfully executing the classification stage at discriminating mental states. Nevertheless, the feature extraction methods based on both EEG and ECoG are discussed elaborately in the subsequent section.

7.1. EEG-Based Feature Extraction

Typically, BCI focuses on identifying acquired events using various neuroimage techniques, the most common of which is electroencephalography (EEG). Since its involvement in successfully executing the classification stage at discriminating mental states, the EEG signal feature extraction method is one of the essential components of a BCI system. According to [ 267 ] on EEG, three types of feature extraction are discussed in detail in the following sections. These features are the time domain, the frequency domain, and the time–frequency domain. The following subsection address the feature domains elaborately.

7.1.1. Time Domain

The time–frequency domain integrates analyses in the time and frequency domains. It depicts the signal energy distribution in the Time–Frequency plane (t-f plane) [ 268 ]. When it comes to deciphering rhythmic information in EEG data, a time–frequency analysis comes in handy. EEG’s time-domain properties are straightforward to fix, but they have the disadvantage of containing non-stationary signals that alter over time. Features are usually derived using signal amplitude values in time-domain approaches that can be distorted by interference as noise during EEG recording.

  • Event related potentials: Event-related potentials (ERPs) are very low voltages generated in brain regions in reaction to specific events or stimuli. They are time-locked EEG alterations that provide a safe and noninvasive way to research psychophysiological aspects of mental activities. A wide range of sensory, cognitive, or motor stimuli can trigger event-related potentials [ 269 , 270 ]. ERPs are useful to measure the time to process a stimulus and a response to be produced. The temporal resolution of event-related potentials is remarkable, but it has a low spatial resolution. ERPs were used by Changoluisa, V. et al. [ 271 ] to build an adaptive strategy for identifying and detecting changeable ERPs. Continuous monitoring of the curve in ERP components takes account of their temporal and spatial information. Some limitations of ERPs are that it shows poor spatial resolution, whether it is suitable with temporal resolution [ 272 ]. Furthermore, a significant drawback of ERP is the difficulty in determining where the electrical activity originates in the brain.
  • − Mean absolute value: M A V = 1 N ∑ n = 1 N x n (1)
  • − Power: P = 1 N ∑ n = 1 N x n 2 (2)
  • − Standard deviation: S D = 1 N ∑ n = 1 N x ( n ) − μ n (3)
  • − Root mean square (RMS): RMS = 1 N ∑ i = 1 N x i 2 1 / 2 (4)
  • − Square root of amplitude (SRA): SRA = 1 N ∑ i = 1 N x i 2 (5)
  • − Skewness value (SV): SV = 1 N ∑ i = 1 N x l − x ¯ σ 3 (6)
  • − Kurtosis value (KV): KV = 1 N ∑ i = 1 N x l − x ¯ σ 4 (7)
  • Hjorth features: Bo Hjorth introduced the Hjorth parameters in 1970 [ 276 ]; the three statistical parameters employed in time-domain signal processing are activity, mobility, and complexity. Dagdevir, E. et al. [ 277 ] proposed a motor imagery-based BCI system where the features were extracted from the dataset using the Hjorth algorithm. The Hjorth features have advantages in real-time analyses as it has a low computation cost. However, it has a statistical bias over signal parameter calculation.
  • Phase lag index (PLI): The functional connectivity is determined by calculating the PLI for two pairs of channels. Since it depicts the actual interaction between sources, this index may help estimate phase synchronization in EEG time series. PLI measures the asymmetry of the distribution of phase differences between two signals. The advantage of PLI is that it is less affected by phase delays. It quantifies the nonzero phase lag between the time series of two sources, making it less vulnerable to signals. The effectiveness of functional connectivity features evaluated by phase lag index (PLI), weighted phase lag index (wPLI), and phase-locking value (PLV) on MI classification was studied by Feng, L.Z. et al. [ 278 ].

7.1.2. Frequency Domain

When analyzing any signal in terms of frequency instead of just time, the frequency domain properties are considered. Any signal’s frequency domain representation displays how much of it falls inside a specific frequency range. The frequency domain properties are commonly acquired using power spectral density (PSD). The discussion about these properties is presented below in the following section.

  • Fast fourier transform (FFT): The Fourier transform is a mathematical transformation that converts any time-domain signal into its frequency domain. Discrete Fourier Transform (DFT) [ 279 ], Short Time Fourier Transform (STFT) [ 280 , 281 ], and Fast Fourier Transform are the most common Fourier transform utilized for EEG-based emotion identification (FFT) [ 282 ]. Djamal, E.C. et al. [ 283 ] developed a wireless device that is used to record a player’s brain activity and extracts each action using Fast Fourier Transform. FFT is faster than any other method available, allowing it to be employed in real-time applications. It is a valuable instrument for signal processing at a fixed location. A limitation of FFT is that it can convert the limited range of waveform data and the requirement to add a window weighting function to the waveform to compensate for spectral leakage.

An external file that holds a picture, illustration, etc.
Object name is sensors-21-05746-g005.jpg

The basic structure of CSP [ 286 ].

In Figure 5 , CSP provides spatial filters that minimize the variance of an individual class while concurrently maximizing the variance of other classes. These filters are mainly used to choose the frequency from the multichannel EEG signal. After frequency filtering, spatial filtering is performed using spatial filters that are employed to extract spatial information from the signal. Spatial information is significantly necessary to differentiate intent patterns in multichannel EEG recordings for BCI. The performance of this spatial filtering depends on the operational frequency band of EEG. Therefore, CSP is categorized as a frequency domain feature. However, CSP acts as signal enhancement while it requires no preceding excerpt or information of sub-specific bands.

  • Higher-order Spectral (HOS): Second-order signal measurements include the auto-correlation function and the power spectrum. Second-order measures operate satisfactorily if the signal resembles a Gaussian probability distribution function. However, most of the real-world signals are non-Gaussian. Therefore, Higher-Order Spectral (HOS) [ 285 ] is an extended version of the second-order measure that works well for non-Gaussian signals, when it comes into the equation. In addition, most of the physiological signals are nonlinear and non-stationary. HOS are considered favorable to detect these deviations from the signal’s linearity or stationarity. It is calculated using the Fourier Transform at various frequencies. H O S = X ( K ) X ( l ) X ∗ ( k + l ) (8) where X ( K ) is the Fourier transform of the raw EEG signal x ( n ) and l is a shifting parameter.

7.1.3. Time–Frequency Domain

In the time-frequency domain, the signal is evaluated both in the time and frequency domains simultaneously. The wavelet transform is one of many advanced approaches for analyzing the time-frequency representation. There are some other widely used models for utilizing the time-frequency domain. These models are addressed with a proper explanation in the subsequent section.

The AR parameters are a p ( i ) , the observations are x ( n ) and the excitation white noise is v ( n ) . Lastly, the most challenging part of AR EEG modeling is choosing the correct model to represent and following the changing spectrum correctly.

  • Wavelet Transform (WT): The WT technique encodes the original EEG data using wavelets, which are known as simple building blocks. It looks at unusual data patterns using variable windows with expansive windows for low frequencies and narrow windows for high frequencies. In addition, WT is considered an advanced approach as it offers a simultaneous localization in the time-frequency domain, which is a significant advantage. These wavelets can be discrete or continuous and describe the signal’s characteristics in a time-domain frequency. The Discrete Wavelet Transform (DWT) and the Continuous Wavelet Transform (CWT) are used frequently in EEG analysis [ 289 ]. DWT is now a more widely used signal processing method than CWT as CWT is very redundant. DWT decomposes any signal into approximation and detail coefficients corresponding to distinct frequency ranges maintaining the temporal information in the signal. However, most researchers try all available wavelets before choosing the optimal one that produces the best results, as selecting a mother wavelet is challenging. In wavelet-based feature extraction, the Daubechies wavelet of order 4 (db4) is the most commonly employed [ 290 ].

7.2. ECoG-Based Features

Electrocorticography (ECoG) generates a reliable signal through electrodes placed on the surface of the human brain, which decodes movement, vision, and speech. Decoding ECoG signal processing gives immediate patient feedback and controls a computer cursor or perhaps an exoskeleton. The ECoG signal feature extraction approach is a crucial element of the BCI system since it is involved in accomplishing the classification phase during decoding. Some of the widely used feature extraction methods are discussed below.

7.2.1. Linear Filtering

It is typically employed to filter out noise in the form of signals that are not in the frequency range of the brain’s messages. Low-pass filters and high-pass filters are the two types of linear filters. This typical linear filtering is used to removed ECOG, EOG, and EMG artifacts from EEG signals. Low pass filtering is used to remove EMG artifacts, and high pass filtering is used to remove EOG artifacts [ 291 ]. These artifacts are noises produced by either physiological processes such as muscle, eye, or other biological movement or exogenous (external) sources such as machinery faults. There are three approaches for dealing with artifacts in EEG signal acquisition. Avoiding artifacts by keeping an eye on the subject’s movements and the machine’s operation. Contaminated trials are discarded due to artifact rejection. Pre-processing techniques are used to remove artifacts. The advantage of linear filtering is that signals are considered a controlled scaling of the signal’s frequency domain components. High pass filtering is used to raise the relative importance of the high-frequency components by reducing the features in the frequency domain’s center.

7.2.2. Spatial Filtering

Spatial filtering is a technique for improving decoding by leveraging information about the electrode positions. The spatial filter aims to lessen the influence of spatial distortion in the raw signal; various ECoG channels are treated as coordinates for multivariate data sampling through spatial filters. The filtering transforms that coordinate system to facilitate decoding. Spatial filtering can use to minimize data dimensionality or to increase the dissimilarity of various observations. The referencing systems used during ECoG recordings are frequently utilized for preliminary spatial filtering. Equation ( 10 ) determines the spatial filter [ 292 ].

where x ′ is the spatially filtered signal, x i is the EEG signal from channel i , and w i is the weight of that channel. With the aid of relevant information acquired from multiple EEG channels, spatial filtering contributes to recovering the brain’s original signal. Simultaneously, it reduces dimensionality by lowering EEG channel size to smaller spatially filtered signals.

Thus far, feature extraction involves extracting new features from existing ones to minimize feature measurement costs, to improve classifier efficiency, and to improve classification accuracy. Now in the following section, the extracted feature classifiers are briefly described.

8. BCI Classifiers

BCI always needs a subject to use its device, and similarly, the subject must produce several types of data to use a BCI device. In addition, to use a BCI system, the subject must develop various brain activity patterns that the system can recognize and convert into commands. To achieve this mentioned conversion, some regression or classification algorithms can be used. The classification step’s design comprises selecting one or more classification algorithms from a variety of options. In this section, some commonly known classifiers [ 293 ], which are classified in Figure 6 , as well as some new classifiers [ 294 ] are described below.

An external file that holds a picture, illustration, etc.
Object name is sensors-21-05746-g006.jpg

Classification of commonly used classifiers in BCI.

8.1. Linear Classifiers

Linear classifiers are discriminant algorithms that discriminate classes using linear functions. It is most likely the most widely used algorithm in BCI systems. Two types of linear classifiers are used during BCI design: linear discriminant analysis (LDA) and support vector machine (SVM).

8.1.1. Linear Discriminant Analysis (LDA)

The objective of Linear Discriminant Analysis is to separate data from diverse classes using a hyperplane. The side of hyperplane determinded through the category of a feature vector in a two-class problem. LDA requires that the data has a normal distribution and that both classes have the same covariance matrix. The separation hyper-plane is based on looking for a projection that maximizes the margin between the means of two classes while minimizing intraclass variance [ 295 ]. Furthermore, this classifier is straightforward to apply and generally produces excellent results and soundly implemented in various BCI system, including MI-based BCI, P300 speller, multiclass, and asynchronous BCI. The disadvantage of LDA is its linearity, which might lead to unsatisfactory results when faced with various nonlinear EEG data.

8.1.2. Support Vector Machine (SVM)

A Support Vector Machine (SVM) uses a discriminant hyperplane to identify classes. The determined hyperplane in SVM is the one that maximizes the margins, i.e., the distance between both the nearest training samples. The ability to generalize is believed to improve when margins are maximized [ 296 ]. Linear SVM [ 297 ] is a type of SVM that allows for classification utilizing linear decision bounds. This classifier has been used to solve a substantial number of synchronous BCI tasks with tremendous success. The SVM classifier also works by projecting the input vector X onto a scalar value f(X), as shown in Equation ( 11 ).

Gaussian SVM or RBF SVM is the term applied to the equivalent SVM. RBF and SVM have also produced remarkable outcomes in BCI applications. SVM is used to solve multiclass BCI problems that use the OVR approach, similar to LDA.

8.2. Neural Networks (NN)

Neural networks (NN) and linear classifiers are the two types of classifiers most usually employed in BCI systems, considering that a NN is a collection of artificial neurons that allows us to create nonlinear decision limits [ 298 ]. The multilayer perceptron (MLP) is the most extensively used NN for BCI, as described in this section. Afterward, it briefly discusses other neural network architectures utilized in BCI systems.

8.2.1. Deep Learning (DL) Models

Deep learning has been widely used in BCI applications nowadays compared with machine learning technologies because most BCI applications require a high level of accuracy. Deep learning models perform better in recognizing changing signals from the brain, which changes swiftly. Some popular DL models such as CNN, GNN, RNN, and LSTM are described below:

  • Convolutional Neural Network (CNN): A convolutional neural network (CNN) is an ANN intended primarily to analyze visual input used in image recognition and processing. The convolutional layer, pooling layer, and fully connected layer are the three layers that comprise CNN. Using a CNN, the input data may be reduced to instant response formations with a minimum loss, and the characteristic spatial relationships of EEG patterns can be recorded. Fatigue detection, sleep stage classification, stress detection, motor imagery data processing, and emotion recognition are among the EEG-based BCI applications using CNNs. In BCI, the CNN models are used in the input brain signals to exploit the latent semantic dependencies.
  • Generative Adversarial Network (GAN): Generative adversarial networks are a recent ML technique. The GAN used two ANN models for competing to train each other simultaneously. GANs allow machines to envision and develop new images on their own. EEG-based BCI techniques recorded the signals first and then moved to the GAN techniques to regenerate the images [ 299 ]. The significant application of GAN-based BCI systems is data augmentation. Data augmentation increases the amount of training data available and allows for more complicated DL models. It can also reduce overfitting and can increase classifier accuracy and robustness. In the context of BCI, generative algorithms, including GAN, are frequently used to rebuild or generate a set of brain signal recordings to improve the training set.
  • Recurrent Neural Network (RNN): RNNs’ basic form is a layer with the output linked to the input. Since it has access to the data from past time-stamps, and the architecture of an RNN layer allows for the model to store memory [ 300 , 301 ]. Since RNN and CNN have strong temporal and spatial feature extraction abilities in most DL approaches, it is logical to mix them for temporal and spatial feature learning. RNN can be considered a more powerful version of hidden Markov models (HMM), which classifies EEG correctly [ 302 ]. LSTM is a kind of RNN with a unique architecture that allows it to acquire long-term dependencies despite the difficulties that RNNs confront. It contains a discrete memory cell, a type of node. To manage the flow of data, LSTM employs an architecture with a series of “gates”. When it comes to modeling time series of tasks such as writing and voice recognition, RNN and LSTM have been proven to be effective [ 303 ].

8.2.2. Multilayer Perceptron (MLP)

An Multilayer Perceptron (MLP) [ 304 ] comprises multiple layers of neurons along with an input layer, one or more hidden layers, and an output layer. The input of each neuron is linked to the output of the neurons in the preceding layer. Meanwhile, the output layer neurons evaluate the classification of the input feature vector. MLP and neural networks can approximate, meaning they can compare continuous functions if they have sufficient neurons and layers. The challenging factor behind MLPs is that they are susceptible to over-training, particularly containing noisy and non-stationary data. As a result, significant selection and regularization of the architecture are necessary. Perceptron is a multilayer with no hidden layers comparable with LDA. It has been used in BCI applications on occasion [ 293 ]. Sunny, M.S.H. et al. [ 305 ] used Multilayer Perceptron (MLP) to distinguish distinct frequency bands from EEG signals to extract features more effectively.

8.2.3. Adaptive Classifiers

As new EEG data become accessible, adaptive classifiers’ parameters, such as the weights allocated to each feature in a linear discriminant hyperplane, are gradually re-estimated and updated. Adaptive classifiers can use supervised and unsupervised adaptation, that is, with or without knowledge of the input data’s real class labels. The true class labels of the receiving EEG signals are obtained using supervised adaptation. The classifier is either reassigned on the existing training data, enhanced with these updated, labeled incoming data, or updated solely on this new data. Supervised user testing is essential for supervised BCI adaptation. The label of the receiving EEG data is vague with unsupervised adaptation. As a result, unsupervised adaptation is based on class-unspecific adaptation, such as updating the generalized classes EEG data mean or a co-variance matrix in the classifier model or estimating the data class labels for additional training [ 306 ].

8.3. Nonlinear Bayesian Classifiers

This section discusses the Bayes quadratic and hidden Markov models (HMM), two Bayesian classifiers used in BCI. Although Bayesian graphical networks (BGN) have been used for BCI, they are not covered here since they are not widely used [ 307 ].

8.3.1. Bayes Quadratic

The objective of Bayesian classification is to provide the highest probability class to a feature vector. The Bayes rule is often used to calculate the a posteriori probability of a feature vector assigned to a single class. The class of this feature vector can be calculated by using the MAP (maximum a posteriori) rule with these probabilities. The Bayes quadratic assumption is that the data have a distinct normal distribution. The result is quadratic decision boundaries that justify the classifier’s name [ 308 ]. Although this classifier is not extensively utilized for BCI, it has been successfully used to classify motor imagery and mental tasks.

8.3.2. Hidden Markov Model

A Bayesian classifier that generates a nonlinear cost function is known as a Hidden Markov Model (HMM). An HMM is a statistical algorithm that calculates the chances of seeing a given set of feature variables [ 309 ]. These statistical probabilities from HMM are generally Gaussian Mixture Models (GMM) in case of BCI [ 310 ]. HMM may be used to categorize temporal patterns of BCI characteristics (Obermaier, B. et al. [ 302 ]), even raw EEG data, since the EEG elements required to control BCI have particular time sequences. Although HMM is not widely used in the BCI world, this research demonstrated that they could be helpful to classification on BCI systems such as EEG signals [ 311 ].

8.4. Nearest Neighbor Classifiers

In this section, some classifiers with distance vectors are described. Classifiers such as K nearest neighbors (KNN) and Mahalanobis distance are common among them as they are nonlinear discriminative classifiers [ 312 ].

8.4.1. K Nearest Neighbors

K nearest neighbor method aims to identify the dominant class amongst an unseen point within the dataset habituated for training. Nearest neighbors are typically estimated using a metric that has some intervals during the signal acquisition of BCI. KNN can construct nonlinear decision boundaries by evaluating any function with enough training data with an inflated k value. The usability of KNN algorithms is less in the BCI field as their condescending sensitivity hampers the capacity, which causes them to fail in multiple BCI research. KNN is efficient in BCI systems with some feature vectors, but low power can cause failure in BCI research [ 313 ].

8.4.2. Mahalanobis Distance

For each prototype of class c , Mahalanobis distance-based classifiers [ 314 ] assume a Gaussian distribution N ( c , M c ) . Subsequently, using the Mahalanobis distance d c , a feature vector x is allocated to the class that corresponds to the closest prototype ( x ).

This results in a basic yet reliable classifier; it has been shown to work in multiclass and asynchronous BCI systems. Considering its excellent results, it is still rarely mentioned in BCI literature [ 315 ].

8.5. Hybrid

In several BCI papers, classification is implemented with a single classifier. Furthermore, a current tendency is to combine many classifiers in various ways [ 316 ]. The following are indeed the classifier combination strategies utilized in BCI systems:

8.5.1. Boosting

Boosting is the process of using multiple classifiers in a cascade, and each focused on the errors made by the one before it. It can combine numerous weak classifiers to form a powerful one; thereforem it is unlikely to overtrain. Moreover, it is susceptible to mislabeling, illustrating why it failed in one BCI trial [ 293 ].

8.5.2. Voting

Multiple classifiers are employed for voting, each of which allocates the input feature vector to a class. The majority class becomes the final class. In BCI systems, voting is the most preferred process of combining classifiers due to its simplicity and efficiency [ 293 ].

8.5.3. Stacking

Stacking is the process of utilizing multiple classifiers to categorize the input feature vector. Level-0 classifiers are what it is named. Each one of these classifiers’ output would then feed into a “meta-classifier” (or “level-1 classifier”), which makes a final decision [ 293 ].

Aforementioned in this section, some other classifiers are utilized in the recent BCI research. Since 2016 transfer learning is used for using MI classification tasks [ 317 ]. Some ground-breaking architectures are established in recent years, such as EEG-inception, an end-to-end Neural network [ 318 ], cluster decomposing, and multi-object optimization-based-ensemble learning framework [ 319 ]; RFNet is a fusion network that learns from attention weights and used in embedding-specific features for decision making [ 179 ].

Now, a better understanding of the performance of commonly known classifiers with some popular datasets are given in Table 9 .

Comparison of classifiers based on popular datasets and features.

9. Evaluation Measurement

To evaluate the performance of BCI systems, researchers employed several evaluation metrics. The most common is accuracy, commonly known as error rate. Although accuracy is not always an acceptable criterion due to specific rigorous requirements, various evaluation criteria have been offered. An overview of BCI research evaluation criteria is provided below.

9.1. Generally Used Evaluation Metrics

In this section, we sorted the most commonly used evaluation metrics for measuring the BCI system performances. The evaluation measures are explained carefully in the following subsections.

9.1.1. The Confusion Matrix

The confusion matrix represents the relationship between the actual class’s user-intentioned output classes and the actual predicted class. True positives rate (TPR), False negative rate (FNR), False positives rate (FPR), Positive predictive value (PPV), and negative predictive value (PPV) are used to describe sensitivity or recall, specificity, (1-specificity), precision, etc. [ 325 ].

9.1.2. Classification Accuracy and Error Rate

Classification accuracy is one of the important metrics in BCI systems; this study evaluates performance using classification accuracy as well as sensitivity and specificity. This measure determines how frequently the BCI makes a right pick or what proportion of all selections are accurate. It is the most obvious indicator of BCI accomplishment, implying that it increase in a linear fashion with decision time, so it takes a long time. The following is the mathematical formula for calculating accuracy:

9.1.3. Information Transfer Rate

Shannon [ 326 ] proposed the Information Transfer Rate (ITR) as the rate that makes up both of these metrics. This rate represents the quantity of data that may pass through the system in one unit of time. In [ 327 ], the information transmission rate in bits per minute ( b i t s / m i n ) and accuracy (ACC) in percentage (%) were used to evaluate performance. They made demographic data (age and gender) as well as the performance outcomes of 10 participants, and the ITR was computed using the Formula ( 14 ), which is as follows:

where N is the number of targets and p is the classification accuracy (ACC). Based on four cursor movements and the choose command, this resulted in a N of 5. Bits per trial were used to compute B t .

According to ITR [ 328 ] also has some important parameters that are used to evaluate BCI. A description of them is given below:

  • Target detection accuracy: The accuracy of target identification may be enhanced by increasing the Signal-to-Noise Ratio (SNR) and the separability of several classes. Several techniques, such as trial averaging, spatial filtering, and eliciting increased task-related EEG signals, are employed in the preprocessing step to reduce the SNR. Many applications utilize trail averaging across topics to improve the performance of a single BCI. These mental states may be used to lower the SNR [ 53 ].
  • Number of classes: The number of classes is raised and more sophisticated applications are built with a high ITR. TDMA, FDMA, and CDMA are among the stimulus coding techniques that have been adopted for BCI systems [ 243 , 329 ]. P300, for example, uses TDMA to code the target stimulus. In VEP-based BCI systems, FDMA and CDMA have been used.
  • Target detection time: The detection time is when a user first expresses their purpose and when the system makes a judgment. One of the goals of BCI systems is to improve the ITR by reducing target detection time. Adaptive techniques, such as the “dynamic halting” method, might be used to minimize the target detection time [ 330 ].

9.1.4. Cohen’s Kappa Coefficient

Cohen’s Kappa measures the agreement between two observers; it measures the contract between the proper output and the command of BCI domain in a BCI-based AAC system. Cohen’s kappa coefficient resolves many of the accuracy measure’s objections [ 331 ]. The general agreement p 0 = A C C , which is equivalent to the classification accuracy and the chance agreement p e , with n i and n i i being the column i t h and row i t h , correspondingly, are used to calculate K .

where posteriori and priori probability are n : i , n i : respectively. The estimated kappa Coefficient K and standard error e ( K ) are acquired by

When there is no correlation between the expected and actual classes, the kappa coefficient becomes zero. A perfect categorization is indicated by a kappa coefficient of 1. If the Kappa value is less than zero, the classifier offers an alternative assignment for the output and actual classes [ 332 ].

9.2. Continuous BCI System Evaluation

Continuous BCI performance was measured using a variety of parameters. Different measures may be even more appropriate depending on whether the study is conducted online or offline. The section goes through some of the most commonly used metrics in this field, including the correlation coefficient, accuracy, and Fitts’s Law [ 333 ].

9.2.1. Correlation Coefficient

The correlation coefficient could be a useful statistic for determining whether an intracortical implant receives task-relevant neurons. There are two essential stipulations: one is scale-invariant, which implies that the cursor might miss the mark substantially while still generating high values if the sign of the actual and anticipated movements coincide [ 334 ]; the other is that a decoder can yield a high value if it simply generates a signal that fluctuates with the repetitions [ 333 ].

9.2.2. Accuracy

Task characteristics such as target size and dwell time have a significant impact on accuracy. As a result, it is more of a sign that the task was is good enough for the subject and modality than a performance measure [ 333 ].

9.2.3. Fitts’s Law

Fitts’s law asserts that the time taken for a person to move a mouse cursor to a targeted object of the target’s distance is divided by its size. The longer it takes, the greater the distance and the narrower the target [ 335 , 336 ]. Fitts’s law requires using a method to calculate the “index of difficulty” of a particular change.

9.3. User-Centric BCI System Evaluation

Users are an essential element of the BCI product life cycle. Their interactions and experiences influence whether BCI systems are acceptable and viable. The four criteria or User Experience (UX) factors are used to evaluate user-centric BCI systems. These are usability, affects, ergonomics, and quality of life, shown below in the following subsection.

9.3.1. Usability

The amount that can be utilized to fulfill specific objectives with effectiveness, efficiency, learnability, and satisfaction in a given context is referred to as usability [ 337 ]. In usability measure, we can include four metrics, such as,

  • Effectiveness or accuracy: It depicts the overall accuracy of the BCI system as experienced from the end user’s perspective [ 333 ].
  • Efficiency or information transfer rate: It refers to the speed and timing at which a task is accomplished. Therefore, it depicts the overall BCI system’s speed, throughput, and latency seen through the eyes of the end user’s perspective [ 333 ].
  • Learnability: The BCI system can make users feel as if they can use the product effectively and quickly learn additional features. Both the end-user and the provider are affected by learnability [ 338 ].
  • Satisfaction: It is based on participants’ reactions to actual feelings while using BCI systems, showing the user’s favorable attitude regarding utilizing the system. To measure satisfaction, we can use rating scales or qualitative methods [ 333 ].

9.3.2. Affect

Regarding BCIs, it might refer to how comfortable the system is, particularly for long periods, and how pleasant or uncomfortable the stimuli are to them. EEG event-related possibilities, spectral characteristics, galvanic skin responses, or heart rates could be used to quantitatively monitor user’s exhaustion, valence, and arousal levels [ 339 ].

9.3.3. Ergonomics

Ergonomics studies are the study of how people interact with their environments. The load on the user’s memory is represented by the cognitive task load, a multidimensional entity. In addition, physiological markers including eye movement, EEG, ERP, and spectral characteristics could also be employed to evaluate cognitive stress objectively [ 340 ].

9.3.4. Quality of Life

It expresses the user’s overall perception of the system’s utility and acceptance and its influence on their well-being. The Return on Investment (ROI) is an economic measure of the perceived benefit derived from it. The overall quality of experience is a measure of how satisfied a user is with their expertise [ 333 ].

Other assessment methods, such as Mutual Information, Written symbol rate (WSR), and Practical bit rate (PBR), are utilized to a lesser extent.

10. Limitations and Challenges

The brain-computer interface is advancing towards a more dynamic and accurate solution of the connection between brain and machine. Still, few factors are resisting achieving the ultimate goal. Therefore, we analyzed a few core research on BCI in this section and found the limitations exhibited in Table 10 . Then, we demonstrated the significant challenges of the BCI domain.

A summary of some research papers proposing new methods of BCI.

The challenges and difficulties of the BCI domain are divided into three categories: challenges based on usability, technical challenges, and ethical challenges. The rest of the section briefly explains these challenges.

10.1. Based on Usability

This section describes the challenges that users have in accepting BCI technology [ 350 ]. They include concerns relating to the requisite training for class discrimination.

10.1.1. Training Time

Usually, training a user, either leading the user through the procedure or the total quantity of the documented manual, takes time. The majority of the time, the user also requests the system to be simpler to use. The users often despise a complicated system that is difficult to manage. It is a challenging effort to create such a sophisticated, user-friendly system [ 351 ].

10.1.2. Fatigue

The majority of present BCIs generate a lot of fatigue since they need a lot of concentration, focus, and awareness to a rapid and intermittent input. In addition to the annoyance of weariness of electrodes, BCI may fail to operate because the user cannot maintain a sufficient degree of focus. As in BCI, mental activity is continually monitored and the user’s attention point alters the input. The concentration necessary for stimuli results in a combination of input and output [ 352 , 353 ]. Rather than relaxing, the user must concentrate on a single point as an input and then look at the outcome. At some point, the interaction has a forced quality to it, rather than the natural quality that would be there if the user could choose whatever part of the visual output to focus on [ 6 ].

10.1.3. Mobility to Users

Across most situations, users are not allowed to move around or to have mobility in BCIs. During the test application, users must stay motionless and quiet, ideally sitting down. However, in a real-world setting, a user may need to utilize BCI while walking down the street, for example, to manage a smartphone. Additionally, BCIs cannot ensure user comfort. Usually, the EEG headset is not lightweight and easy to carry, which hampers the user experience.

10.1.4. Psychophysiological and Neurological Challenges

Emotional and mental mechanisms, cognition-related neurophysiology, and neurological variables, such as functionality and architecture, play vital roles in BCI performance, resulting in significant intra- and inter-individual heterogeneity. Immediate brain dynamics are influenced by psychological elements such as attention; memory load; weariness; conflicting cognitive functions; and users’ specific characteristics such as lifestyle, gender, and age. Participants with weaker empathy engage less emotionally in a P300-BCI paradigm and generate larger P300 wave amplitudes than someone with greater empathy involvement [ 354 ].

10.2. Technical Challenges

Non-linearity, non-stationarity, and noise as well as limited training sets and the accompanying dimensionality curse are difficulties relating to the recorded electrophysiological characteristics of brain impulses.

10.2.1. Non-Linearity

The brain is a very complex nonlinear system in which chaotic neuronal ensemble activity may be seen. Nonlinear dynamic techniques can thus better describe EEG data than linear ones.

10.2.2. Non-Stationarity

The non-stationarity of electrophysiological brain signals to recognize human recognition is a significant challenge in developing a BCI system. It results in a constant shift in the signals utilized with time, either between or within transition time. EEG signal variability can be influenced by the mental and emotional state backdrop across sessions. In addition, various emotional states such as sadness, happiness, anxiety, and fear can vary on daily basis that reflects non-stationarity [ 355 ]. Noise is also a significant contribution to the non-stationarity problems that BCI technology faces. Noises and other external interferences are always present in raw EEG data of emotion recognition that is most robust [ 356 ]. It comprises undesired signals generated by changes in electrode location as well as noise from the surroundings [ 357 ].

10.2.3. Transfer Rate of Signals

In BCIs, the system must continuously adjust to the signals of the user. This modification must be made quickly and precisely. Current BCIs have an extremely slow information transfer rate, taking almost two minutes to “digitalize” a single phrase, for example. Furthermore, BCI accuracy does not always reach a desirable level, particularly in visual stimulus-based BCI. Actions must sometimes be repeated or undone, producing pain or even dissatisfaction in using interactive systems using this type of interface [ 358 ].

10.2.4. Signal Processing

Recently, a variety of decoding techniques, signal processing algorithms, and classification algorithms have been studied. Despite this, the information retrieved from EEG waves does not have a high enough signal-to-noise ratio to operate a device with some extent of liberty, such as a prosthetic limb. Algorithms that are more resilient, accurate, and quick are required to control BCI.

10.2.5. Training Sets

In BCI, the training process is mainly impacted by usability concerns, but training sets are tiny in most cases. Although the subjects find the training sessions time-consuming and challenging, they give the user the required expertise to interact with the system and to learn to manage their neurophysiological signals. As a result, balancing the technological complexity of decoding the user’s brain activity with the level of training required for the proper functioning of the interfaces is a crucial issue in building a BCI [ 359 ].

10.2.6. Lack of Data Analysis Method

The classifiers should be evaluated online since every BCI implementation is in an online situation. Additionally, it should be validated to ensure that it has low complexity and can be calibrated rapidly in real-time. Domain adaptation and transfer learning could be an acceptable solution for developing calibration-free BCIs, where even the integration of unique feature sets, such as covariance matrices with domain adaptation algorithms, can strengthen the invariance performance of BCIs.

10.2.7. Performance Evaluation Metrics

A variety of performance evaluation measures are used to evaluate BCI systems. However, when different evaluation metrics are used to assess BCI systems, it is nearly impossible to compare systems. As a result, the BCI research community should establish a uniform and systematic approach to quantify a particular BCI application or a particular metric. For example, to test the efficiency of a BCI wheelchair control, the number of control commands, categories of control commands, total distance, time consumed, the number of collisions, classification accuracy, and the average success rate need to be evaluated, among other factors [ 360 ].

10.2.8. Low ITR of BCI Systems

The information transfer rate is one of the extensively used processes for the performance evaluation metrics of BCI systems. The number of classes, target detection accuracy, and target detection time are all factors of this rate. By increasing the Signal-to-Noise Ratio (SNR), it can improve the target detection accuracy [ 53 , 328 ]. Several techniques are typically used for the preprocessing phase to optimize the SNR. When a high ITR has been attained, more complicated applications can be created by expanding the number of classes available. CDMA, TDMA, and FDMA [ 243 , 361 ] are only a few of the stimulus coding schemes that have already been developed for BCI systems. TDMA was used with P300 to code the required stimuli, while CDMA and FDMA have been used with BCIs that interact with VEP. Furthermore, the essential aspect of BCIs is reducing the target recognition period, which helps to increase the ITR. Adaptive techniques, such as “dynamic stopping”, could be an effective option for accomplishing this.

10.2.9. Specifically Allocated Lab for BCI Technology

Most of the BCI systems are trialed in a supervised lab rather than in the actual surroundings of the users. When designing a BCI system, it is essential to think about the environment in which the technology may be used. It is critical to thoroughly investigate the system’s requirements, environmental factors, circumstances, and target users mostly during the system design phase.

10.3. Ethical Challenges

There are many thoughts surrounding the ethical issues behind BCI as it considers physical, psychological, and social factors. In biological factors, BCI always finds a human body to identify signals that must be acquainted with electrodes. As humans need to wear these electrodes, it is always risky for them and can harm the human body to some worse extent. BCI also requires strict maintenance of the human body during signal acquisition, so the subject must sit for a long time in his place. Adding to that, a user or participant must act what the electrodes need, so they cannot do anything willingly. This fact can have a substantial impact on the human body.

11. Conclusions

The brain-computer interface is a communication method that joins the wired brain and external applications and devices directly. The BCI domain includes investigating, assisting, augmenting, and experimenting with brain signal activities. Due to transatlantic documentation, low-cost amplifiers, greater temporal resolution, and superior signal analysis methods, BCI technologies are available to researchers in diverse domains. Moreover, It is an interdisciplinary area that allows for biology, engineering, computer science, and applied mathematics research. However, an architectural and constructive investigation of the brain–computer interface is exhibited in this article. It is aimed at novices who would like to learn about the current state of BCI systems and methodologies. The fundamental principles of BCI techniques are discussed elaborately. It describes the architectural perspectives of certain unique taxons and gives a taxonomy of BCI systems. The paper also covered feature extraction, classification, evaluation procedures, and techniques as the research continues. It presents a summary of the present methods for creating various types of BCI systems. The study looks into the different types of datasets that are available for BCI systems as well. The article also explains the challenges and limitations of the described BCI systems, along with possible solutions. Lastly, BCI technology advancement is accomplished in four stages: primary scientific development, preclinical experimentation, clinical investigation, and commercialization. At present, most of the BCI techniques are in the preclinical and clinical phases. The combined efforts of scientific researchers and the tech industries are needed to avail the benefit of this great domain to ordinary people through commercialization.


We would like to thank Bangladesh University of Business & Technology (BUBT), University of Asia Pacific (UAP), and University of Aizu (UoA) for supporting this research. Also, special thanks to the Advanced Machine Learning lab, BUBT; Computer Vision & Pattern Recognition Lab, UAP; Database System Lab, UoA; for giving facilities to research and publish.

Author Contributions

Conceptualization, M.F.M.; Data curation, M.F.M., S.C.D., M.M.K. and A.A.L.; Formal analysis, M.F.M.; Investigation, M.R.I. and Y.W.; Methodology, M.F.M., S.C.D., M.M.K., A.A.L., M.R.I. and Y.W.; Software, S.C.D., M.M.K. and A.A.L.; Supervision, M.R.I.; Validation, M.F.M., M.R.I. and Y.W.; Visualization, M.F.M., S.C.D., M.M.K. and A.A.L.; Writing—original draft, M.F.M., S.C.D., M.M.K., A.A.L., M.R.I. and Y.W.; Writing—review & editing, M.F.M., M.R.I. and Y.W. All authors have read and agreed to the published version of the manuscript.

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Data availability statement, conflicts of interest.

The authors declare no conflict of interest.

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Quantum Computing Review: A Decade of Research

Ieee account.

  • Change Username/Password
  • Update Address

Purchase Details

  • Payment Options
  • Order History
  • View Purchased Documents

Profile Information

  • Communications Preferences
  • Profession and Education
  • Technical Interests
  • US & Canada: +1 800 678 4333
  • Worldwide: +1 732 981 0060
  • Contact & Support
  • About IEEE Xplore
  • Accessibility
  • Terms of Use
  • Nondiscrimination Policy
  • Privacy & Opting Out of Cookies

A not-for-profit organization, IEEE is the world's largest technical professional organization dedicated to advancing technology for the benefit of humanity. © Copyright 2024 IEEE - All rights reserved. Use of this web site signifies your agreement to the terms and conditions.

Computer Technology Research Paper Topics

Academic Writing Service

This list of computer technology research paper topics  provides the list of 33 potential topics for research papers and an overview article on the history of computer technology.

1. Analog Computers

Paralleling the split between analog and digital computers, in the 1950s the term analog computer was a posteriori projected onto pre-existing classes of mechanical, electrical, and electromechanical computing artifacts, subsuming them under the same category. The concept of analog, like the technical demarcation between analog and digital computer, was absent from the vocabulary of those classifying artifacts for the 1914 Edinburgh Exhibition, the first world’s fair emphasizing computing technology, and this leaves us with an invaluable index of the impressive number of classes of computing artifacts amassed during the few centuries of capitalist modernity. True, from the debate between ‘‘smooth’’ and ‘‘lumpy’’ artificial lines of computing (1910s) to the differentiation between ‘‘continuous’’ and ‘‘cyclic’’ computers (1940s), the subsequent analog–digital split became possible by the multitudinous accumulation of attempts to decontextualize the computer from its socio-historical use alternately to define the ideal computer technically. The fact is, however, that influential classifications of computing technology from the previous decades never provided an encompassing demarcation compared to the analog– digital distinction used since the 1950s. Historians of the digital computer find that the experience of working with software was much closer to art than science, a process that was resistant to mass production; historians of the analog computer find this to have been typical of working with the analog computer throughout all its aspects. The historiography of the progress of digital computing invites us to turn to the software crisis, which perhaps not accidentally, surfaced when the crisis caused by the analog ended. Noticeably, it was not until the process of computing with a digital electronic computer became sufficiently visual by the addition of a special interface—to substitute for the loss of visualization that was previously provided by the analog computer—that the analog computer finally disappeared.

Academic Writing, Editing, Proofreading, And Problem Solving Services

Get 10% off with 24start discount code, 2. artificial intelligence.

Artificial intelligence (AI) is the field of software engineering that builds computer systems and occasionally robots to perform tasks that require intelligence. The term ‘‘artificial intelligence’’ was coined by John McCarthy in 1958, then a graduate student at Princeton, at a summer workshop held at Dartmouth in 1956. This two-month workshop marks the official birth of AI, which brought together young researchers who would nurture the field as it grew over the next several decades: Marvin Minsky, Claude Shannon, Arthur Samuel, Ray Solomonoff, Oliver Selfridge, Allen Newell, and Herbert Simon. It would be difficult to argue that the technologies derived from AI research had a profound effect on our way of life by the beginning of the 21st century. However, AI technologies have been successfully applied in many industrial settings, medicine and health care, and video games. Programming techniques developed in AI research were incorporated into more widespread programming practices, such as high-level programming languages and time-sharing operating systems. While AI did not succeed in constructing a computer which displays the general mental capabilities of a typical human, such as the HAL computer in Arthur C. Clarke and Stanley Kubrick’s film 2001: A Space Odyssey, it has produced programs that perform some apparently intelligent tasks, often at a much greater level of skill and reliability than humans. More than this, AI has provided a powerful and defining image of what computer technology might someday be capable of achieving.

3. Computer and Video Games

Interactive computer and video games were first developed in laboratories as the late-night amusements of computer programmers or independent projects of television engineers. Their formats include computer software; networked, multiplayer games on time-shared systems or servers; arcade consoles; home consoles connected to television sets; and handheld game machines. The first experimental projects grew out of early work in computer graphics, artificial intelligence, television technology, hardware and software interface development, computer-aided education, and microelectronics. Important examples were Willy Higinbotham’s oscilloscope-based ‘‘Tennis for Two’’ at the Brookhaven National Laboratory (1958); ‘‘Spacewar!,’’ by Steve Russell, Alan Kotok, J. Martin Graetz and others at the Massachusetts Institute of Technology (1962); Ralph Baer’s television-based tennis game for Sanders Associates (1966); several networked games from the PLATO (Programmed Logic for Automatic Teaching Operations) Project at the University of Illinois during the early 1970s; and ‘‘Adventure,’’ by Will Crowther of Bolt, Beranek & Newman (1972), extended by Don Woods at Stanford University’s Artificial Intelligence Laboratory (1976). The main lines of development during the 1970s and early 1980s were home video consoles, coin-operated arcade games, and computer software.

4. Computer Displays

The display is an essential part of any general-purpose computer. Its function is to act as an output device to communicate data to humans using the highest bandwidth input system that humans possess—the eyes. Much of the development of computer displays has been about trying to get closer to the limits of human visual perception in terms of color and spatial resolution. Mainframe and minicomputers used ‘‘terminals’’ to display the output. These were fed data from the host computer and processed the data to create screen images using a graphics processor. The display was typically integrated with a keyboard system and some communication hardware as a terminal or video display unit (VDU) following the basic model used for teletypes. Personal computers (PCs) in the late 1970s and early 1980s changed this model by integrating the graphics controller into the computer chassis itself. Early PC displays typically displayed only monochrome text and communicated in character codes such as ASCII. Line-scanning frequencies were typically from 15 to 20 kilohertz—similar to television. CRT displays rapidly developed after the introduction of video graphics array (VGA) technology (640 by 480 pixels in16 colors) in the mid-1980s and scan frequencies rose to 60 kilohertz or more for mainstream displays; 100 kilohertz or more for high-end displays. These displays were capable of displaying formats up to 2048 by 1536 pixels with high color depths. Because the human eye is very quick to respond to visual stimulation, developments in display technology have tended to track the development of semiconductor technology that allows the rapid manipulation of the stored image.

5. Computer Memory for Personal Computers

During the second half of the twentieth century, the two primary methods used for the long-term storage of digital information were magnetic and optical recording. These methods were selected primarily on the basis of cost. Compared to core or transistorized random-access memory (RAM), storage costs for magnetic and optical media were several orders of magnitude cheaper per bit of information and were not volatile; that is, the information did not vanish when electrical power was turned off. However, access to information stored on magnetic and optical recorders was much slower compared to RAM memory. As a result, computer designers used a mix of both types of memory to accomplish computational tasks. Designers of magnetic and optical storage systems have sought meanwhile to increase the speed of access to stored information to increase the overall performance of computer systems, since most digital information is stored magnetically or optically for reasons of cost.

6. Computer Modeling

Computer simulation models have transformed the natural, engineering, and social sciences, becoming crucial tools for disciplines as diverse as ecology, epidemiology, economics, urban planning, aerospace engineering, meteorology, and military operations. Computer models help researchers study systems of extreme complexity, predict the behavior of natural phenomena, and examine the effects of human interventions in natural processes. Engineers use models to design everything from jets and nuclear-waste repositories to diapers and golf clubs. Models enable astrophysicists to simulate supernovas, biochemists to replicate protein folding, geologists to predict volcanic eruptions, and physiologists to identify populations at risk of lead poisoning. Clearly, computer models provide a powerful means of solving problems, both theoretical and applied.

7. Computer Networks

Computers and computer networks have changed the way we do almost everything—the way we teach, learn, do research, access or share information, communicate with each other, and even the way we entertain ourselves. A computer network, in simple terms, consists of two or more computing devices (often called nodes) interconnected by means of some medium capable of transmitting data that allows the computers to communicate with each other in order to provide a variety of services to users.

8. Computer Science

Computer science occupies a unique position among the scientific and technical disciplines. It revolves around a specific artifact—the electronic digital computer—that touches upon a broad and diverse set of fields in its design, operation, and application. As a result, computer science represents a synthesis and extension of many different areas of mathematics, science, engineering, and business.

9. Computer-Aided Control Technology

The story of computer-aided control technology is inextricably entwined with the modern history of automation. Automation in the first half of the twentieth century involved (often analog) processes for continuous automatic measurement and control of hardware by hydraulic, mechanical, or electromechanical means. These processes facilitated the development and refinement of battlefield fire-control systems, feedback amplifiers for use in telephony, electrical grid simulators, numerically controlled milling machines, and dozens of other innovations.

10. Computer-Aided Design and Manufacture

Computer-aided design and manufacture, known by the acronym CAD/CAM, is a process for manufacturing mechanical components, wherein computers are used to link the information needed in and produced by the design process to the information needed to control the machine tools that produce the parts. However, CAD/CAM actually constitutes two separate technologies that developed along similar, but unrelated, lines until they were combined in the 1970s.

11. Computer-User Interface

A computer interface is the point of contact between a person and an electronic computer. Today’s interfaces include a keyboard, mouse, and display screen. Computer user interfaces developed through three distinct stages, which can be identified as batch processing, interactive computing, and the graphical user interface (GUI). Today’s graphical interfaces support additional multimedia features, such as streaming audio and video. In GUI design, every new software feature introduces more icons into the process of computer– user interaction. Presently, the large vocabulary of icons used in GUI design is difficult for users to remember, which creates a complexity problem. As GUIs become more complex, interface designers are adding voice recognition and intelligent agent technologies to make computer user interfaces even easier to operate.

12. Early Computer Memory

Mechanisms to store information were present in early mechanical calculating machines, going back to Charles Babbage’s analytical engine proposed in the 1830s. It introduced the concept of the ‘‘store’’ and, if ever built, would have held 1000 numbers of up to 50 decimal digits. However, the move toward base-2 or binary computing in the 1930s brought about a new paradigm in technology—the digital computer, whose most elementary component was an on–off switch. Information on a digital system is represented using a combination of on and off signals, stored as binary digits (shortened to bits): zeros and ones. Text characters, symbols, or numerical values can all be coded as bits, so that information stored in digital memory is just zeros and ones, regardless of the storage medium. The history of computer memory is closely linked to the history of computers but a distinction should be made between primary (or main) and secondary memory. Computers only need operate on one segment of data at a time, and with memory being a scarce resource, the rest of the data set could be stored in less expensive and more abundant secondary memory.

13. Early Digital Computers

Digital computers were a marked departure from the electrical and mechanical calculating and computing machines in wide use from the early twentieth century. The innovation was of information being represented using only two states (on or off), which came to be known as ‘‘digital.’’ Binary (base 2) arithmetic and logic provided the tools for these machines to perform useful functions. George Boole’s binary system of algebra allowed any mathematical equation to be represented by simply true or false logic statements. By using only two states, engineering was also greatly simplified, and universality and accuracy increased. Further developments from the early purpose-built machines, to ones that were programmable accompanied by many key technological developments, resulted in the well-known success and proliferation of the digital computer.

14. Electronic Control Technology

The advancement of electrical engineering in the twentieth century made a fundamental change in control technology. New electronic devices including vacuum tubes (valves) and transistors were used to replace electromechanical elements in conventional controllers and to develop new types of controllers. In these practices, engineers discovered basic principles of control theory that could be further applied to design electronic control systems.

15. Encryption and Code Breaking

The word cryptography comes from the Greek words for ‘‘hidden’’ (kryptos) and ‘‘to write’’ (graphein)—literally, the science of ‘‘hidden writing.’’ In the twentieth century, cryptography became fundamental to information technology (IT) security generally. Before the invention of the digital computer at mid-century, national governments across the world relied on mechanical and electromechanical cryptanalytic devices to protect their own national secrets and communications, as well as to expose enemy secrets. Code breaking played an important role in both World Wars I and II, and the successful exploits of Polish and British cryptographers and signals intelligence experts in breaking the code of the German Enigma ciphering machine (which had a range of possible transformations between a message and its code of approximately 150 trillion (or 150 million million million) are well documented.

16. Error Checking and Correction

In telecommunications, whether transmission of data or voice signals is over copper, fiber-optic, or wireless links, information coded in the signal transmitted must be decoded by the receiver from a background of noise. Signal errors can be introduced, for example from physical defects in the transmission medium (semiconductor crystal defects, dust or scratches on magnetic memory, bubbles in optical fibers), from electromagnetic interference (natural or manmade) or cosmic rays, or from cross-talk (unwanted coupling) between channels. In digital signal transmission, data is transmitted as ‘‘bits’’ (ones or zeros, corresponding to on or off in electronic circuits). Random bit errors occur singly and in no relation to each other. Burst error is a large, sustained error or loss of data, perhaps caused by transmission problems in the connecting cables, or sudden noise. Analog to digital conversion can also introduce sampling errors.

17. Global Positioning System (GPS)

The NAVSTAR (NAVigation System Timing And Ranging) Global Positioning System (GPS) provides an unlimited number of military and civilian users worldwide with continuous, highly accurate data on their position in four dimensions— latitude, longitude, altitude, and time— through all weather conditions. It includes space, control, and user segments (Figure 6). A constellation of 24 satellites in 10,900 nautical miles, nearly circular orbits—six orbital planes, equally spaced 60 degrees apart, inclined approximately 55 degrees relative to the equator, and each with four equidistant satellites—transmits microwave signals in two different L-band frequencies. From any point on earth, between five and eight satellites are ‘‘visible’’ to the user. Synchronized, extremely precise atomic clocks—rubidium and cesium— aboard the satellites render the constellation semiautonomous by alleviating the need to continuously control the satellites from the ground. The control segment consists of a master facility at Schriever Air Force Base, Colorado, and a global network of automated stations. It passively tracks the entire constellation and, via an S-band uplink, periodically sends updated orbital and clock data to each satellite to ensure that navigation signals received by users remain accurate. Finally, GPS users—on land, at sea, in the air or space—rely on commercially produced receivers to convert satellite signals into position, time, and velocity estimates.

18. Gyrocompass and Inertial Guidance

Before the twentieth century, navigation at sea employed two complementary methods, astronomical and dead reckoning. The former involved direct measurements of celestial phenomena to ascertain position, while the latter required continuous monitoring of a ship’s course, speed, and distance run. New navigational technology was required not only for iron ships in which traditional compasses required correction, but for aircraft and submarines in which magnetic compasses cannot be used. Owing to their rapid motion, aircraft presented challenges for near instantaneous navigation data collection and reduction. Electronics furnished the exploitation of radio and the adaptation of a gyroscope to direction finding through the invention of the nonmagnetic gyrocompass.

Although the Cold War arms race after World War II led to the development of inertial navigation, German manufacture of the V-2 rocket under the direction of Wernher von Braun during the war involved a proto-inertial system, a two-gimballed gyro with an integrator to determine speed. Inertial guidance combines a gyrocompass with accelerometers installed along orthogonal axes, devices that record all accelerations of the vehicle in which inertial guidance has been installed. With this system, if the initial position of the vehicle is known, then the vehicle’s position at any moment is known because integrators record all directions and accelerations and calculate speeds and distance run. Inertial guidance devices can subtract accelerations due to gravity or other motions of the vehicle. Because inertial guidance does not depend on an outside reference, it is the ultimate dead reckoning system, ideal for the nuclear submarines for which they were invented and for ballistic missiles. Their self-contained nature makes them resistant to electronic countermeasures. Inertial systems were first installed in commercial aircraft during the 1960s. The expense of manufacturing inertial guidance mechanisms (and their necessary management by computer) has limited their application largely to military and some commercial purposes. Inertial systems accumulate errors, so their use at sea (except for submarines) has been as an adjunct to other navigational methods, unlike aircraft applications. Only the development of the global positioning system (GPS) at the end of the century promised to render all previous navigational technologies obsolete. Nevertheless, a range of technologies, some dating to the beginning of the century, remain in use in a variety of commercial and leisure applications.

19. Hybrid Computers

Following the emergence of the analog–digital demarcation in the late 1940s—and the ensuing battle between a speedy analog versus the accurate digital—the term ‘‘hybrid computer’’ surfaced in the early 1960s. The assumptions held by the adherents of the digital computer—regarding the dynamic mechanization of computational labor to accompany the equally dynamic increase in computational work—was becoming a universal ideology. From this perspective, the digital computer justly appeared to be technically superior. In introducing the digital computer to social realities, however, extensive interaction with the experienced analog computer adherents proved indispensable, especially given that the digital proponents’ expectation of progress by employing the available and inexpensive hardware was stymied by the lack of inexpensive software. From this perspective—as historiographically unwanted it may be by those who agree with the essentialist conception of the analog–digital demarcation—the history of the hybrid computer suggests that the computer as we now know it was brought about by linking the analog and the digital, not by separating them. Placing the ideal analog and the ideal digital at the two poles, all computing techniques that combined some features of both fell beneath ‘‘hybrid computation’’; the designators ‘‘balanced’’ or ‘‘true’’ were preserved for those built with appreciable amounts of both. True hybrids fell into the middle spectrum that included: pure analog computers, analog computers using digital-type numerical analysis techniques, analog computers programmed with the aid of digital computers, analog computers using digital control and logic, analog computers using digital subunits, analog computers using digital computers as peripheral equipment, balanced hybrid computer systems, digital computers using analog subroutines, digital computers with analog arithmetic elements, digital computers designed to permit analog-type programming, digital computers with analog-oriented compilers and interpreters, and pure digital computers.

20. Information Theory

Information theory, also known originally as the mathematical theory of communication, was first explicitly formulated during the mid-twentieth century. Almost immediately it became a foundation; first, for the more systematic design and utilization of numerous telecommunication and information technologies; and second, for resolving a paradox in thermodynamics. Finally, information theory has contributed to new interpretations of a wide range of biological and cultural phenomena, from organic physiology and genetics to cognitive behavior, human language, economics, and political decision making. Reflecting the symbiosis between theory and practice typical of twentieth century technology, technical issues in early telegraphy and telephony gave rise to a proto-information theory developed by Harry Nyquist at Bell Labs in 1924 and Ralph Hartley, also at Bell Labs, in 1928. This theory in turn contributed to advances in telecommunications, which stimulated the development of information theory per se by Claude Shannon and Warren Weaver, in their book The Mathematical Theory of Communication published in 1949. As articulated by Claude Shannon, a Bell Labs researcher, the technical concept of information is defined by the probability of a specific message or signal being picked out from a number of possibilities and transmitted from A to B. Information in this sense is mathematically quantifiable. The amount of information, I, conveyed by signal, S, is inversely related to its probability, P. That is, the more improbable a message, the more information it contains. To facilitate the mathematical analysis of messages, the measure is conveniently defined as I ¼ log2 1/P(S), and is named a binary digit or ‘‘bit’’ for short. Thus in the simplest case of a two-state signal (1 or 0, corresponding to on or off in electronic circuits), with equal probability for each state, the transmission of either state as the code for a message would convey one bit of information. The theory of information opened up by this conceptual analysis has become the basis for constructing and analyzing digital computational devices and a whole range of information technologies (i.e., technologies including telecommunications and data processing), from telephones to computer networks.

21. Internet

The Internet is a global computer network of networks whose origins are found in U.S. military efforts. In response to Sputnik and the emerging space race, the Advanced Research Projects Agency (ARPA) was formed in 1958 as an agency of the Pentagon. The researchers at ARPA were given a generous mandate to develop innovative technologies such as communications.

In 1962, psychologist J.C.R. Licklider from the Massachusetts Institute of Technology’s Lincoln Laboratory joined ARPA to take charge of the Information Processing Techniques Office (IPTO). In 1963 Licklider wrote a memo proposing an interactive network allowing people to communicate via computer. This project did not materialize. In 1966, Bob Taylor, then head of the IPTO, noted that he needed three different computer terminals to connect to three different machines in different locations around the nation. Taylor also recognized that universities working with IPTO needed more computing resources. Instead of the government buying machines for each university, why not share machines? Taylor revitalized Licklider’s idea, securing $1 million in funding, and hired 29-yearold Larry Roberts to direct the creation of ARPAnet.

In 1974, Robert Kahn and Vincent Cerf proposed the first internet-working protocol, a way for datagrams (packets) to be communicated between disparate networks, and they called it an ‘‘internet.’’ Their efforts created transmission control protocol/internet protocol (TCP/IP). In 1982, TCP/IP replaced NCP on ARPAnet. Other networks adopted TCP/IP and it became the dominant standard for all networking by the late 1990s.

In 1981 the U.S. National Science Foundation (NSF) created Computer Science Network (CSNET) to provide universities that did not have access to ARPAnet with their own network. In 1986, the NSF sponsored the NSFNET ‘‘backbone’’ to connect five supercomputing centers. The backbone also connected ARPAnet and CSNET together, and the idea of a network of networks became firmly entrenched. The open technical architecture of the Internet allowed numerous innovations to be grafted easily onto the whole. When ARPAnet was dismantled in 1990, the Internet was thriving at universities and technology- oriented companies. The NSF backbone was dismantled in 1995 when the NSF realized that commercial entities could keep the Internet running and growing on their own, without government subsidy. Commercial network providers worked through the Commercial Internet Exchange to manage network traffic.

22. Mainframe Computers

The term ‘‘computer’’ currently refers to a general-purpose, digital, electronic, stored-program calculating machine. The term ‘‘mainframe’’ refers to a large, expensive, multiuser computer, able to handle a wide range of applications. The term was derived from the main frame or cabinet in which the central processing unit (CPU) and main memory of a computer were kept separate from those cabinets that held peripheral devices used for input and output.

Computers are generally classified as supercomputers, mainframes, minicomputers, or microcomputers. This classification is based on factors such as processing capability, cost, and applications, with supercomputers the fastest and most expensive. All computers were called mainframes until the 1960s, including the first supercomputer, the naval ordnance research calculator (NORC), offered by International Business Machines (IBM) in 1954. In 1960, Digital Equipment Corporation (DEC) shipped the PDP-1, a computer that was much smaller and cheaper than a mainframe.

Mainframes once each filled a large room, cost millions of dollars, and needed a full maintenance staff, partly in order to repair the damage caused by the heat generated by their vacuum tubes. These machines were characterized by proprietary operating systems and connections through dumb terminals that had no local processing capabilities. As personal computers developed and began to approach mainframes in speed and processing power, however, mainframes have evolved to support a client/server relationship, and to interconnect with open standard-based systems. They have become particularly useful for systems that require reliability, security, and centralized control. Their ability to process large amounts of data quickly make them particularly valuable for storage area networks (SANs). Mainframes today contain multiple CPUs, providing additional speed through multiprocessing operations. They support many hundreds of simultaneously executing programs, as well as numerous input and output processors for multiplexing devices, such as video display terminals and disk drives. Many legacy systems, large applications that have been developed, tested, and used over time, are still running on mainframes.

23. Mineral Prospecting

Twentieth century mineral prospecting draws upon the accumulated knowledge of previous exploration and mining activities, advancing technology, expanding knowledge of geologic processes and deposit models, and mining and processing capabilities to determine where and how to look for minerals of interest. Geologic models have been developed for a wide variety of deposit types; the prospector compares geologic characteristics of potential exploration areas with those of deposit models to determine which areas have similar characteristics and are suitable prospecting locations. Mineral prospecting programs are often team efforts, integrating general and site-specific knowledge of geochemistry, geology, geophysics, and remote sensing to ‘‘discover’’ hidden mineral deposits and ‘‘measure’’ their economic potential with increasing accuracy and reduced environmental disturbance. Once a likely target zone has been identified, multiple exploration tools are used in a coordinated program to characterize the deposit and its economic potential.

24. Packet Switching

Historically the first communications networks were telegraphic—the electrical telegraph replacing the mechanical semaphore stations in the mid-nineteenth century. Telegraph networks were largely eclipsed by the advent of the voice (telephone) network, which first appeared in the late nineteenth century, and provided the immediacy of voice conversation. The Public Switched Telephone Network allows a subscriber to dial a connection to another subscriber, with the connection being a series of telephone lines connected together through switches at the telephone exchanges along the route. This technique is known as circuit switching, as a circuit is set up between the subscribers, and is held until the call is cleared.

One of the disadvantages of circuit switching is the fact that the capacity of the link is often significantly underused due to silences in the conversation, but the spare capacity cannot be shared with other traffic. Another disadvantage is the time it takes to establish the connection before the conversation can begin. One could liken this to sending a railway engine from London to Edinburgh to set the points before returning to pick up the carriages. What is required is a compromise between the immediacy of conversation on an established circuit-switched connection, with the ad hoc delivery of a store-and-forward message system. This is what packet switching is designed to provide.

25. Personal Computers

A personal computer, or PC, is designed for personal use. Its central processing unit (CPU) runs single-user systems and application software, processes input from the user, sending output to a variety of peripheral devices. Programs and data are stored in memory and attached storage devices. Personal computers are generally single-user desktop machines, but the term has been applied to any computer that ‘‘stands alone’’ for a single user, including portable computers.

The technology that enabled the construction of personal computers was the microprocessor, a programmable integrated circuit (or ‘‘chip’’) that acts as the CPU. Intel introduced the first microprocessor in 1971, the 4-bit 4004, which it called a ‘‘microprogrammable computer on a chip.’’ The 4004 was originally developed as a general-purpose chip for a programmable calculator, but Intel introduced it as part of Intel’s Microcomputer System 4-bit, or MCS-4, which also included read-only memory (ROM) and random-access memory (RAM) memory chips and a shift register chip. In August 1972, Intel followed with the 8-bit 8008, then the more powerful 8080 in June 1974. Following Intel’s lead, computers based on the 8080 were usually called microcomputers.

The success of the minicomputer during the 1960s prepared computer engineers and users for ‘‘single person, single CPU’’ computers. Digital Equipment Corporation’s (DEC) widely used PDP-10, for example, was smaller, cheaper, and more accessible than large mainframe computers. Timeshared computers operating under operating systems such as TOPS-10 on the PDP-10— co-developed by the Massachusetts Institute of Technology (MIT) and DEC in 1972—created the illusion of individual control of computing power by providing rapid access to personal programs and files. By the early 1970s, the accessibility of minicomputers, advances in microelectronics, and component miniaturization created expectations of affordable personal computers.

26. Printers

Printers generally can be categorized as either impact or nonimpact. Like typewriters, impact printers generate output by striking the page with a solid substance. Impact printers include daisy wheel and dot matrix printers. The daisy wheel printer, which was introduced in 1972 by Diablo Systems, operates by spinning the daisy wheel to the correct character whereupon a hammer strikes it, forcing the character through an inked ribbon and onto the paper. Dot matrix printers operate by using a series of small pins to strike a matrix or grid ribbon coated with ink. The strike of the pin forces the ink to transfer to the paper at the point of impact. Unlike daisy wheel printers, dot matrix printers can generate italic and other character types through producing different pin patterns. Nonimpact printers generate images by spraying or fusing ink to paper or other output media. This category includes inkjet printers, laser printers, and thermal printers. Whether they are inkjet or laser, impact or nonimpact, all modern printers incorporate features of dot matrix technology in their design: they operate by generating dots onto paper or other physical media.

27. Processors for Computers

A processor is the part of the computer system that manipulates the data. The first computer processors of the late 1940s and early 1950s performed three main functions and had three main components. They worked in a cycle to gather, decode, and execute instructions. They were made up of the arithmetic and logic unit, the control unit, and some extra storage components or registers. Today, most processors contain these components and perform these same functions, but since the 1960s they have developed different forms, capabilities, and organization. As with computers in general, increasing speed and decreasing size has marked their development.

28. Radionavigation

Astronomical and dead-reckoning techniques furnished the methods of navigating ships until the twentieth century, when exploitation of radio waves, coupled with electronics, met the needs of aircraft with their fast speeds, but also transformed all navigational techniques. The application of radio to dead reckoning has allowed vessels to determine their positions in all-weather by direction finding (known as radio direction finding, or RDF) or by hyperbolic systems. Another use of radio, radar (radio direction and rangefinding), enables vessels to determine their distance to, or their bearing from, objects of known position. Radionavigation complements traditional navigational methods by employing three frames of reference. First, radio enables a vessel to navigate by lines of bearing to shore transmitters (the most common use of radio). This is directly analogous to the use of lighthouses for bearings. Second, shore stations may take radio bearings of craft and relay to them computed positions. Third, radio beacons provide aircraft or ships with signals that function as true compasses.

29. Software Application Programs

At the beginning of the computer age around the late 1940s, inventors of the intelligent machine were not thinking about applications software, or any software other than that needed to run the bare machine to do mathematical calculating. It was only when Maurice Wilkes’ young protégé David Williams crafted a tidy set of initial orders for the EDSAC, an early programmable digital computer, that users could string together standard subroutines to a program and have the execution jump between them. This was the beginning of software as we know it—something that runs on a machine other than an operating system to make it do anything desired. ‘‘Applications’’ are software other than system programs that run the actual hardware. Manufacturers always had this software, and as the 1950s progressed they would ‘‘bundle’’ applications with hardware to make expensive computers more attractive. Some programming departments were even placed in the marketing departments.

30. Software Engineering

Software engineering aims to develop the programs that allow digital computers to do useful work in a systematic, disciplined manner that produces high-quality software on time and on budget. As computers have spread throughout industrialized societies, software has become a multibillion dollar industry. Both the users and developers of software depend a great deal on the effectiveness of the development process.

Software is a concept that didn’t even pertain to the first electronic digital computers. They were ‘‘programmed’’ through switches and patch cables that physically altered the electrical pathways of the machine. It was not until the Manchester Mark I, the first operational stored-program electronic digital computer, was developed in 1948 at the University of Manchester in England that configuring the machine to solve a specific problem became a matter of software rather than hardware. Subsequently, instructions were stored in memory along with data.

31. Supercomputers

Supercomputers are high-performance computing devices that are generally used for numerical calculation, for the study of physical systems either through numerical simulation or the processing of scientific data. Initially, they were large, expensive, mainframe computers, which were usually owned by government research labs. By the end of the twentieth century, they were more often networks of inexpensive small computers. The common element of all of these machines was their ability to perform high-speed floating-point arithmetic— binary arithmetic that approximates decimal numbers with a fixed number of bits—the basis of numerical computation.

With the advent of inexpensive supercomputers, these machines moved beyond the large government labs and into smaller research and engineering facilities. Some were used for the study of social science. A few were employed by business concerns, such as stock brokerages or graphic designers.

32. Systems Programs

The operating systems used in all computers today are a result of the development and organization of early systems programs designed to control and regulate the operations of computer hardware. The early computing machines such as the ENIAC of 1945 were ‘‘programmed’’ manually with connecting cables and setting switches for each new calculation. With the advent of the stored program computer of the late 1940s (the Manchester Mark I, EDVAC, EDSAC (electronic delay storage automatic calculator), the first system programs such as assemblers and compilers were developed and installed. These programs performed oft repeated and basic operations for computer use including converting programs into machine code, storing and retrieving files, managing computer resources and peripherals, and aiding in the compilation of new programs. With the advent of programming languages, and the dissemination of more computers in research centers, universities, and businesses during the late 1950s and 1960s, a large group of users began developing programs, improving usability, and organizing system programs into operating systems.

The 1970s and 1980s saw a turn away from some of the complications of system software, an interweaving of features from different operating systems, and the development of systems programs for the personal computer. In the early 1970s, two programmers from Bell Laboratories, Ken Thompson and Dennis Ritchie, developed a smaller, simpler operating system called UNIX. Unlike past system software, UNIX was portable and could be run on different computer systems. Due in part to low licensing fees and simplicity of design, UNIX increased in popularity throughout the 1970s. At the Xerox Palo Alto Research Center, research during the 1970s led to the development of system software for the Apple Macintosh computer that included a GUI (graphical user interface). This type of system software filtered the user’s interaction with the computer through the use of graphics or icons representing computer processes. In 1985, a year after the release of the Apple Macintosh computer, a GUI was overlaid on Microsoft’s then dominant operating system, MS-DOS, to produce Microsoft Windows. The Microsoft Windows series of operating systems became and remains the dominant operating system on personal computers.

33. World Wide Web

The World Wide Web (Web) is a ‘‘finite but unbounded’’ collection of media-rich digital resources that are connected through high-speed digital networks. It relies upon an Internet protocol suite that supports cross-platform transmission and makes available a wide variety of media types (i.e., multimedia). The cross-platform delivery environment represents an important departure from more traditional network communications protocols such as e-mail, telnet, and file transfer protocols (FTP) because it is content-centric. It is also to be distinguished from earlier document acquisition systems such as Gopher, which was designed in 1991, originally as a mainframe program but quickly implemented over networks, and wide area information systems (WAIS), also released in 1991. WAIS accommodated a narrower range of media formats and failed to include hyperlinks within their navigation protocols. Following the success of Gopher on the Internet, the Web quickly extended and enriched the metaphor of integrated browsing and navigation. This made it possible to navigate and peruse a wide variety of media types effortlessly on the Web, which in turn led to the Web’s hegemony as an Internet protocol.

History of Computer Technology

Computer Technology

The modern computer—the (electronic) digital computer in which the stored program concept is realized and hence self-modifying programs are possible—was only invented in the 1940s. Nevertheless, the history of computing (interpreted as the usage of modern computers) is only understandable against the background of the many forms of information processing as well as mechanical computing devices that solved mathematical problems in the first half of the twentieth century. The part these several predecessors played in the invention and early history of the computer may be interpreted from two different perspectives: on the one hand it can be argued that these machines prepared the way for the modern digital computer, on the other hand it can be argued that the computer, which was invented as a mathematical instrument, was reconstructed to be a data-processing machine, a control mechanism, and a communication tool.

The invention and early history of the digital computer has its roots in two different kinds of developments: first, information processing in business and government bureaucracies; and second, the use and the search for mathematical instruments and methods that could solve mathematical problems arising in the sciences and in engineering.

Origins in Mechanical Office Equipment

The development of information processing in business and government bureaucracies had its origins in the late nineteenth century, which was not just an era of industrialization and mass production but also a time of continuous growth in administrative work. The economic precondition for this development was the creation of a global economy, which caused growth in production of goods and trade. This brought with it an immense increase in correspondence, as well as monitoring and accounting activities—corporate bureaucracies began to collect and process data in increasing quantities. Almost at the same time, government organizations became more and more interested in collating data on population and demographic changes (e.g., expanding tax revenues, social security, and wide-ranging planning and monitoring functions) and analyzing this data statistically.

Bureaucracies in the U.S. and in Europe reacted in a different way to these changes. While in Europe for the most part neither office machines nor telephones entered offices until 1900, in the U.S. in the last quarter of the nineteenth century the information-handling techniques in bureaucracies were radically changed because of the introduction of mechanical devices for writing, copying, and counting data. The rise of big business in the U.S. had caused a growing demand for management control tools, which was fulfilled by a new ideology of systematic management together with the products of the rising office machines industry. Because of a later start in industrialization, the government and businesses in the U.S. were not forced to reorganize their bureaucracies when they introduced office machines. This, together with an ideological preference for modern office equipment, was the cause of a market for office machines and of a far-reaching mechanization of office work in the U.S. In the 1880s typewriters and cash registers became very widespread, followed by adding machines and book-keeping machines in the 1890s. From 1880 onward, the makers of office machines in the U.S. underwent a period of enormous growth, and in 1920 the office machine industry annually generated about $200 million in revenue. In Europe, by comparison, mechanization of office work emerged about two decades later than in the U.S.—both Germany and Britain adopted the American system of office organization and extensive use of office machines for the most part no earlier than the 1920s.

During the same period the rise of a new office machine technology began. Punched card systems, initially invented by Herman Hollerith to analyze the U.S. census in 1890, were introduced. By 1911 Hollerith’s company had only about 100 customers, but after it had been merged in the same year with two other companies to become the Computing- Tabulating-Recording Company (CTR), it began a tremendous ascent to become the world leader in the office machine industry. CTR’s general manager, Thomas J. Watson, understood the extraordinary potential of these punched-card accounting devices, which enabled their users to process enormous amounts of data largely automatically, in a rapid way and at an adequate level of cost and effort. Due to Watson’s insights and his extraordinary management abilities, the company (which had since been renamed to International Business Machines (IBM)) became the fourth largest office machine supplier in the world by 1928—topped only by Remington Rand, National Cash Register (NCR), and the Burroughs Adding Machine Company.

Origin of Calculating Devices and Analog Instruments

Compared with the fundamental changes in the world of corporate and government bureaucracies caused by office machinery during the late nineteenth and early twentieth century, calculating machines and instruments seemed to have only a minor influence in the world of science and engineering. Scientists and engineers had always been confronted with mathematical problems and had over the centuries developed techniques such as mathematical tables. However, many new mathematical instruments emerged in the nineteenth century and increasingly began to change the world of science and engineering. Apart from the slide rule, which came into popular use in Europe from the early nineteenth century onwards (and became the symbol of the engineer for decades), calculating machines and instruments were only produced on a large scale in the middle of the nineteenth century.

In the 1850s the production of calculating machines as well as that of planimeters (used to measure the area of closed curves, a typical problem in land surveying) started on different scales. Worldwide, less than 2,000 calculating machines were produced before 1880, but more than 10,000 planimeters were produced by the early 1880s. Also, various types of specialized mathematical analog instruments were produced on a very small scale in the late nineteenth century; among them were integraphs for the graphical solution of special types of differential equations, harmonic analyzers for the determination of Fourier coefficients of a periodic function, and tide predictors that could calculate the time and height of the ebb and flood tides.

Nonetheless, in 1900 only geodesists and astronomers (as well as part of the engineering community) made extensive use of mathematical instruments. In addition, the establishment of applied mathematics as a new discipline took place at German universities on a small scale and the use of apparatus and machines as well as graphical and numerical methods began to flourish during this time. After World War I, the development of engineering sciences and of technical physics gave a tremendous boost to applied mathematics in Germany and Britain. In general, scientists and engineers became more aware of the capabilities of calculating machines and a change of the calculating culture—from the use of tables to the use of calculating machines—took place.

One particular problem that was increasingly encountered by mechanical and electrical engineers in the 1920s was the solution of several types of differential equations, which were not solvable by analytic solutions. As one important result of this development, a new type of analog instrument— the so called ‘‘differential analyzer’’—was invented in 1931 by the engineer Vannevar Bush at the Massachusetts Institute of Technology (MIT). In contrast to its predecessors—several types of integraphs—this machine (which was later called an analog computer) could be used not only to solve a special class of differential equation, but a more general class of differential equations associated with engineering problems. Before the digital computer was invented in the 1940s there was an intensive use of analog instruments (similar to Bush’s differential analyzer) and a number of machines were constructed in the U.S. and in Europe after the model of Bush’s machine before and during World War II. Analog instruments also became increasingly important in several fields such as the firing control of artillery on warships or the control of rockets. It is worth mentioning here that only for a limited class of scientific and engineering problems was it possible to construct an analog computer— weather forecasting and the problem of shock waves produced by an atomic bomb, for example, required the solution of partial differential equations, for which a digital computer was needed.

The Invention of the Computer

The invention of the electronic digital stored-program computer is directly connected with the development of numerical calculation tools for the solution of mathematical problems in the sciences and in engineering. The ideas that led to the invention of the computer were developed simultaneously by scientists and engineers in Germany, Britain, and the U.S. in the 1930s and 1940s. The first freely programmable program-controlled automatic calculator was developed by the civil engineering student Konrad Zuse in Germany. Zuse started development work on program-controlled computing machines in the 1930s, when he had to deal with extensive calculations in static, and in 1941 his Z3, which was based on electromechanical relay technology, became operational.

Several similar developments in the U.S. were in progress at the same time. In 1937 Howard Aiken, a physics student at Harvard University, approached IBM to build a program-controlled calculator— later called the ‘‘Harvard Mark I.’’ On the basis of a concept Aiken had developed because of his experiences with the numerical solution of partial differential equations, the machine was built and became operational in 1944. At almost the same time a series of important relay computers was built at the Bell Laboratories in New York following a suggestion by George R. Stibitz. All these developments in the U.S. were spurred by the outbreak of World War II. The first large-scale programmable electronic computer called the Colossus was built in complete secrecy in 1943 to 1944 at Bletchley Park in Britain in order to help break the German Enigma machine ciphers.

However, it was neither these relay calculators nor the Colossus that were decisive for the development of the universal computer, but the ENIAC (electronic numerical integrator and computer), which was developed at the Moore School of Engineering at the University of Pennsylvania. Extensive ballistic calculations were carried out there for the U.S. Army during World War II with the aid of the Bush ‘‘differential analyzer’’ and more than a hundred women (‘‘computors’’) working on mechanical desk calculators. Observing that capacity was barely sufficient to compute the artillery firing tables, the physicist John W. Mauchly and the electronic engineer John Presper Eckert started developing the ENIAC, a digital version of the differential analyzer, in 1943 with funding from the U.S. Army.

In 1944 the mathematician John von Neumann turned his attention to the ENIAC because of his mathematical work on the Manhattan Project (on the implosion of the hydrogen bomb). While the ENIAC was being built, Neumann and the ENIAC team drew up plans for a successor to the ENIAC in order to improve the shortcomings of the ENIAC concept, such as the very small memory and the time-consuming reprogramming (actually rewiring) required to change the setup for a new calculation. In these meetings the idea of a stored-program, universal machine evolved. Memory was to be used to store the program in addition to data. This would enable the machine to execute conditional branches and change the flow of the program. The concept of a computer in the modern sense of the word was born and in 1945 von Neumann wrote the important ‘‘First Draft of a Report on the EDVAC,’’ which described the stored-program, universal computer. The logical structure that was presented in this draft report is now referred to as the ‘‘von Neumann architecture.’’ This EDVAC report was originally intended for internal use but once made freely available it became the ‘‘bible’’ for computer pioneers throughout the world in the 1940s and 1950s. The first computer featuring the von Neumann architecture operated at Cambridge University in the U.K.; in June 1949 the EDSAC (electronic delay storage automatic computer) computer built by Maurice Wilkes—designed according to the EDVAC principles—became operational.

The Computer as a Scientific Instrument

As soon as the computer was invented, a growing demand for computers by scientists and engineers evolved, and numerous American and European universities started their own computer projects in the 1940s and 1950s. After the technical difficulties of building an electronic computer were solved, scientists grasped the opportunity to use the new scientific instrument for their research. For example, at the University of Gottingen in Germany, the early computers were used for the initial value problems of partial differential equations associated with hydrodynamic problems from atomic physics and aerodynamics. Another striking example was the application of von Neumann’s computer at the Institute for Advanced Study (IAS) in Princeton to numerical weather forecasts in 1950. As a result, numerical weather forecasts could be made on a regular basis from the mid-1950s onwards.

Mathematical methods have always been of a certain importance for science and engineering sciences, but only the use of the electronic digital computer (as an enabling technology) made it possible to broaden the application of mathematical methods to such a degree that research in science, medicine, and engineering without computer- based mathematical methods has become virtually inconceivable at the end of the twentieth century. A number of additional computer-based techniques, such as scientific visualization, medical imaging, computerized tomography, pattern recognition, image processing, and statistical applications, have become of the utmost significance for science, medicine, engineering, and social sciences. In addition, the computer changed the way engineers construct technical artifacts fundamentally because of the use of computer-based methods such as computer-aided design (CAD), computer-aided manufacture (CAM), computer-aided engineering, control applications, and finite-element methods. However, the most striking example seems to be the development of scientific computing and computer modeling, which became accepted as a third mode of scientific research that complements experimentation and theoretical analysis. Scientific computing and computer modeling are based on supercomputers as the enabling technology, which became important tools for modern science routinely used to simulate physical and chemical phenomena. These high-speed computers became equated with the machines developed by Seymour Cray, who built the fastest computers in the world for many years. The supercomputers he launched such as the legendary CRAY I from 1976 were the basis for computer modeling of real world systems, and helped, for example, the defense industry in the U.S. to build weapons systems and the oil industry to create geological models that show potential oil deposits.

Growth of Digital Computers in Business and Information Processing

When the digital computer was invented as a mathematical instrument in the 1940s, it could not have been foreseen that this new artifact would ever be of a certain importance in the business world. About 50 firms entered the computer business worldwide in the late 1940s and the early 1950s, and the computer was reconstructed to be a type of electronic data-processing machine that took the place of punched-card technology as well as other office machine technology. It is interesting to consider that there were mainly three types of companies building computers in the 1950s and 1960s: newly created computer firms (such as the company founded by the ENIAC inventors Eckert and Mauchly), electronics and control equipments firms (such as RCA and General Electric), and office appliance companies (such as Burroughs and NCR). Despite the fact that the first digital computers were put on the market by a German and a British company, U.S. firms dominated the world market from the 1950s onward, as these firms had the biggest market as well as financial support from the government.

Generally speaking, the Cold War exerted an enormous influence on the development of computer technology. Until the early 1960s the U.S. military and the defense industry were the central drivers of the digital computer expansion, serving as the main market for computer technology and shaping and speeding up the formation of the rising computer industry. Because of the U.S. military’s role as the ‘‘tester’’ for prototype hard- and software, it had a direct and lasting influence on technological developments; in addition, it has to be noted that the spread of computer technology was partly hindered by military secrecy. Even after the emergence of a large civilian computer market in the 1960s, the U.S. military maintained its influence by investing a great deal in computer in hard- and software and in computer research projects.

From the middle of the 1950s onwards the world computer market was dominated by IBM, which accounted for more than 70 percent of the computer industry revenues until the mid-1970s. The reasons for IBM’s overwhelming success were diverse, but the company had a unique combination of technical and organizational capabilities at its disposal that prepared it perfectly for the mainframe computer market. In addition, IBM benefited from enormous government contracts, which helped to develop excellence in computer technology and design. However, the greatest advantage of IBM was by no doubt its marketing organization and its reputation as a service-oriented firm, which was used to working closely with customers to adapt machinery to address specific problems, and this key difference between IBM and its competitors persisted right into the computer age.

During the late 1950s and early 1960s, the computer market—consisting of IBM and seven other companies called the ‘‘seven dwarves’’—was dominated by IBM, with its 650 and 1401 computers. By 1960 the market for computers was still small. Only about 7,000 computers had been delivered by the computer industry, and at this time even IBM was primarily a punched-card machine supplier, which was still the major source of its income. Only in 1960 did a boom in demand for computers start, and by 1970 the number of computers installed worldwide had increased to more than 100,000. The computer industry was on the track to become one of the world’s major industries, and was totally dominated by IBM.

The outstanding computer system of this period was IBM’s System/360. It was announced in 1964 as a compatible family of the same computer architecture, and employed interchangeable peripheral devices in order to solve IBM’s problems with a hotchpotch of incompatible product lines (which had evoked large problems in the development and maintenance of a great deal of different hardware and software products). Despite the fact that neither the technology used nor the systems programming were of a high-tech technology at the time, the System/360 established a new standard for mainframe computers for decades. Various computer firms in the U.S., Europe, Japan and even Russia, concentrated on copying components, peripherals for System/360 or tried to build System/360-compatible computers.

The growth of the computer market during the 1960s was accompanied by market shakeouts: two of the ‘‘seven dwarves’’ left the computer business after the first computer recession in the early 1970s, and afterwards the computer market was controlled by IBM and BUNCH (Burroughs, UNIVAC, NCR, Control Data, and Honeywell). At the same time, an internationalization of the computer market took place—U.S. companies controlled the world market for computers— which caused considerable fears over loss of national independence in European and Japanese national governments, and these subsequently stirred up national computing programs. While the European attempts to create national champions as well as the more general attempt to create a European-wide market for mainframe computers failed in the end, Japan’s attempt to found a national computer industry has been successful: Until today Japan is the only nation able to compete with the U.S. in a wide array of high-tech computer-related products.

Real-Time and Time-Sharing

Until the 1960s almost all computers in government and business were running batch-processing applications (i.e., the computers were only used in the same way as the punched-card accounting machines they had replaced). In the early 1950s, however, the computer industry introduced a new mode of computing named ‘‘real-time’’ in the business sector for the first time, which was originally developed for military purposes in MIT’s Whirlwind project. This project was initially started in World War II with the aim of designing an aircraft simulator by analog methods, and later became a part of a research and development program for the gigantic, computerized anti-aircraft defense system SAGE (semi-automatic ground environment) built up by IBM in the 1950s.

The demand for this new mode of computing was created by cultural and structural changes in economy. The increasing number of financial transactions in banks and insurance companies as well as increasing airline traveling activities made necessary new computer-based information systems that led finally to new forms of business evolution through information technology.

The case of the first computerized airline reservation system SABRE, developed for American Airlines by IBM in the 1950s and finally implemented in the early 1960s, serves to thoroughly illustrate these structural and structural changes in economy. Until the early 1950s, airline reservations had been made manually without any problems, but by 1953 this system was in crisis because increased air traffic and growing flight plan complexity had made reservation costs insupportable. SABRE became a complete success, demonstrating the potential of centralized real-time computing systems connected via a network. The system enabled flight agents throughout the U.S., who were equipped with desktop terminals, to gain a direct, real-time access to the central reservation system based on central IBM mainframe computers, while the airline was able to assign appropriate resources in response. Therefore, an effective combination of advantages was offered by SABRE—a better utilization of resources and a much higher customer convenience.

Very soon this new mode of computing spread around the business and government world and became commonplace throughout the service and distribution sectors of the economy; for example, bank tellers and insurance account representatives increasingly worked at terminals. On the one hand structural information problems led managers to go this way, and on the other hand the increasing use of computers as information handling machines in government and business had brought about the idea of computer-based accessible data retrieval. In the end, more and more IBM customers wanted to link dozens of operators directly to central computers by using terminal keyboards and display screens.

In the late 1950s and early 1960s—at the same time that IBM and American Airlines had begun the development of the SABRE airline reservation system—a group of brilliant computer scientists had a new idea for computer usage named ‘‘time sharing.’’ Instead of dedicating a multi-terminal system solely to a single application, they had the computer utility vision of organizing a mainframe computer so that several users could interact with it simultaneously. This vision was to change the nature of computing profoundly, because computing was no longer provided to naive users by programmers and systems analysts, and by the late 1960s time-sharing computers became widespread in the U.S.

Particularly important for this development had been the work of J.C.R. Licklider of the Advanced Research Project Agency (ARPA) of the U.S. Department of Defense. In 1960 Licklider had published a now-classic paper ‘‘Man–Computer Symbiosis’’ proposing the use of computers to augment human intellect and creating the vision of interactive computing. Licklider was very successful in translating his idea of a network allowing people on different computers to communicate into action, and convinced ARPA to start an enormous research program in 1962. Its budget surpassed that of all other sources of U.S. public research funding for computers combined. The ARPA research programs resulted in a series of fundamental moves forward in computer technology in areas such as computer graphics, artificial intelligence, and operating systems. For example, even the most influential current operating system, the general-purpose time-sharing system Unix, developed in the early 1970s at the Bell Laboratories, was a spin-off of an ambitious operating system project, Multics, funded by ARPA. The designers of Unix successfully attempted to keep away from complexity by using a clear, minimalist design approach to software design, and created a multitasking, multiuser operating system, which became the standard operating system in the 1980s.

Electronic Component Revolution

While the nature of business computing was changed by the new paradigms such as real time and time sharing, advances in solid-state components increasingly became a driving force for fundamental changes in the computer industry, and led to a dynamic interplay between new computer designs and new programming techniques that resulted in a remarkable series of technical developments. The technical progress of the mainframe computer had always run parallel to conversions in the electronics components. During the period from 1945 to 1965, two fundamental transformations in the electronics industry took place that were marked by the invention of the transistor in 1947 and the integrated circuit in 1957 to 1958. While the first generation of computers—lasting until about 1960—was characterized by vacuum tubes (valves) for switching elements, the second generation used the much smaller and more reliable transistors, which could be produced at a lower price. A new phase was inaugurated when an entire integrated circuit on a chip of silicon was produced in 1961, and when the first integrated circuits were produced for the military in 1962. A remarkable pace of progress in semiconductor innovations, known as the ‘‘revolution in miniature,’’ began to speed up the computer industry. The third generation of computers characterized by the use of integrated circuits began with the announcement of the IBM System/360 in 1964 (although this computer system did not use true integrated circuits). The most important effect of the introduction of integrated circuits was not to strengthen the leading mainframe computer systems, but to destroy Grosch’s Law, which stated that computing power increases as the square of its costs. In fact, the cost of computer power dramatically reduced during the next ten years.

This became clear with the introduction of the first computer to use integrated circuits on a full scale in 1965: the Digital Equipment Corporation (DEC) offered its PDP-8 computer for just $18,000, creating a new class of computers called minicomputers—small in size and low in cost—as well as opening up the market to new customers. Minicomputers were mainly used in areas other than general-purpose computing such as industrial applications and interactive graphics systems. The PDP-8 became the first widely successful minicomputer with over 50,000 items sold, demonstrating that there was a market for smaller computers. This success of DEC (by 1970 it had become the world’s third largest computer manufacturer) was supported by dramatic advances in solid-state technology. During the 1960s the number of transistors on a chip doubled every two years, and as a result minicomputers became continuously more powerful and more inexpensive at an inconceivable speed.

Personal Computing

The most striking aspect of the consequences of the exponential increase of the number of transistors on a chip during the 1960s—as stated by ‘‘Moore’s Law’’: the number of transistors on a chip doubled every two years—was not the lowering of the costs of mainframe computer and minicomputer processing and storage, but the introduction of the first consumer products based on chip technology such as hand-held calculators and digital watches in about 1970. More specifically, the market acts in these industries were changed overnight by the shift from mechanical to chip technology, which led to an enormous deterioration in prices as well as a dramatic industry shakeout. These episodes only marked the beginning of wide-ranging changes in economy and society during the last quarter of the twentieth century leading to a new situation where chips played an essential role in almost every part of business and modern life.

The case of the invention of the personal computer serves to illustrate that it was not sufficient to develop the microprocessor as the enabling technology in order to create a new invention, but how much new technologies can be socially constructed by cultural factors and commercial interests. When the microprocessor, a single-chip integrated circuit implementation of a CPU, was launched by the semiconductor company Intel in 1971, there was no hindrance to producing a reasonably priced microcomputer, but it took six years until the consumer product PC emerged. None of the traditional mainframe and minicomputer companies were involved in creating the early personal computer. Instead, a group of computer hobbyists as well as the ‘‘computer liberation’’ movement in the U.S. became the driving force behind the invention of the PC. These two groups were desperately keen on a low-priced type of minicomputer for use at home for leisure activities such as computer games; or rather they had the counterculture vision of an unreservedly available and personal access to an inexpensive computer utility provided with rich information. When in 1975 the Altair 8800, an Intel 8080 microprocessor-based computer, was offered as an electronic hobbyist kit for less than $400, these two groups began to realize their vision of a ‘‘personal computer.’’ Very soon dozens of computer clubs and computer magazines were founded around the U.S., and these computer enthusiasts created the personal computer by combining the Altair with keyboards, disk drives, and monitors as well as by developing standard software for it. Consequently, in only two years, a more or less useless hobbyist kit had been changed into a computer that could easily be transformed in a consumer product.

The computer hobbyist period ended in 1977, when the first standard machines for an emerging consumer product mass market were sold. These included products such as the Commodore Pet and the Apple II, which included its own monitor, disk drive, and keyboard, and was provided with several basic software packages. Over next three years, spreadsheet, word processing, and database software were developed, and an immense market for games software evolved. As a result, personal computers became more and more a consumer product for ordinary people, and Apple’s revenues shot to more than $500 million in 1982. By 1980, the personal computer had transformed into a business machine, and IBM decided to develop its own personal computer, which was introduced as the IBM PC in 1981. It became an overwhelming success and set a new industry standard.

Apple tried to compete by launching their new Macintosh computer in 1984 provided with a revolutionary graphical user interface (GUI), which set a new standard for a user-friendly human–computer interaction. It was based on technology created by computer scientists at the Xerox Palo Alto Research Center in California, who had picked up on ideas about human– computer interaction developed at the Stanford Research Institute and at the University of Utah. Despite the fact that the Macintosh’s GUI was far superior to the MS-DOS operating system of the IBM-compatible PCs, Apple failed to win the business market and remained a niche player with a market share of about 10 percent. The PC main branch was determined by the companies IBM had chosen as its original suppliers in 1981 for the design of the microprocessor (Intel) and the operating system (Microsoft). While IBM failed to seize power in the operating system software market for PCs in a software war with Microsoft, Microsoft achieved dominance not only of the key market for PC operating systems, but also the key market of office applications during the first half of the 1990s.

In the early 1990s computing again underwent further fundamental changes with the appearance of the Internet, and for the most computer users, networking became an integral part of what it means to have a computer. Furthermore, the rise of the Internet indicated the impending arrival of a new ‘‘information infrastructure’’ as well as of a ‘‘digital convergence,’’ as the coupling of computers and communications networks was often called.

In addition, the 1990s were a period of an information technology boom, which was mainly based on the Internet hype. For many years previously, it seemed to a great deal of managers and journalists that the Internet would become not just an indispensable business tool, but also a miracle cure for economic growth and prosperity. In addition, computer scientists and sociologists started a discussion predicting the beginning of a new ‘‘information age’’ based on the Internet as a ‘‘technological revolution’’ and reshaping the ‘‘material basis’’ of industrial societies.

The Internet was the outcome of an unusual collaboration of a military–industrial–academic complex that promoted the development of this extraordinary innovation. It grew out of a military network called the ARPAnet, a project established and funded by ARPA in the 1960s. The ARPAnet was initially devoted to support of data communications for defense research projects and was only used by a small number of researchers in the 1970s. Its further development was primarily promoted by unintentional forms of network usage. The users of the ARPAnet became very much attracted by the opportunity for communicating through electronic mail, which rapidly surpassed all other forms of network activities. Another unplanned spin-off of the ARPAnet was the Usenet (Unix User Network), which started in 1979 as a link between two universities and enabled its users to subscribe to newsgroups. Electronic mail became a driving force for the creation of a large number of new proprietary networks funded by the existing computer services industry or by organizations such as the NSF (NSFnet). Because networks users’ desire for email to be able to cross network boundaries, an ARPA project on ‘‘internetworking’’ became the origin for the ‘‘Internet’’—a network of networks linked by several layers of protocols such as TCP/IP (transmission control protocol/internet protocol), which quickly developed into the actual standard.

Only after the government funding had solved many of the most essential technical issues and had shaped a number of the most characteristic features of the Internet, did private sector entrepreneurs start Internet-related ventures and quickly developed user-oriented enhancements. Nevertheless, the Internet did not make a promising start and it took more than ten years before significant numbers of networks were connected. In 1980, the Internet had less than two hundred hosts, and during the next four years the number of hosts went up only to 1000. Only when the Internet reached the educational and business community of PC users in the late 1980s, did it start to become an important economic and social phenomenon. The number of hosts began an explosive growth in the late 1980s—by 1988 there were over 50,000 hosts. An important and unforeseen side effect of this development became the creation of the Internet into a new electronic publishing medium. The electronic publishing development that excited most interest in the Internet was the World Wide Web, originally developed at the CERN High Energy Physics Laboratory in Geneva in 1989. Soon there were millions of documents on the Internet, and private PC users became excited by the joys of surfing the Internet. A number of firms such as AOL soon provided low-cost network access and a range of consumer-oriented information services. The Internet boom was also helped by the Clinton–Gore presidential election campaign on the ‘‘information superhighway’’ and by the amazing news reporting on the national information infrastructure in the early 1990s. Nevertheless, for many observers it was astounding how fast the number of hosts on the Internet increased during the next few years—from more than 1 million in 1992 to 72 million in 1999.

The overwhelming success of the PC and of the Internet tends to hide the fact that its arrival marked only a branching in computer history and not a sequence. (Take, for example, the case of mainframe computers, which still continue to run, being of great importance to government facilities and the private sector (such as banks and insurance companies), or the case of supercomputers, being of the utmost significance for modern science and engineering.) Furthermore, it should be noted that only a small part of the computer applications performed today is easily observable—98 percent of programmable CPUs are used in embedded systems such as automobiles, medical devices, washing machines and mobile telephones.

Browse other Technology Research Paper Topics .


research paper on computers


Researchers show classical computers can keep up with, and surpass, their quantum counterparts

Researchers adopt innovative method to boost speed and accuracy of traditional computing.

Quantum computing has been hailed as a technology that can outperform classical computing in both speed and memory usage, potentially opening the way to making predictions of physical phenomena not previously possible.

Many see quantum computing's advent as marking a paradigm shift from classical, or conventional, computing. Conventional computers process information in the form of digital bits (0s and 1s), while quantum computers deploy quantum bits (qubits) to store quantum information in values between 0 and 1. Under certain conditions this ability to process and store information in qubits can be used to design quantum algorithms that drastically outperform their classical counterparts. Notably, quantum's ability to store information in values between 0 and 1 makes it difficult for classical computers to perfectly emulate quantum ones.

However, quantum computers are finicky and have a tendency to lose information. Moreover, even if information loss can be avoided, it is difficult to translate it into classical information -- which is necessary to yield a useful computation.

Classical computers suffer from neither of those two problems. Moreover, cleverly devised classical algorithms can further exploit the twin challenges of information loss and translation to mimic a quantum computer with far fewer resources than previously thought -- as recently reported in a research paper in the journal PRX Quantum .

The scientists' results show that classical computing can be reconfigured to perform faster and more accurate calculations than state-of-the-art quantum computers.

This breakthrough was achieved with an algorithm that keeps only part of the information stored in the quantum state -- and just enough to be able to accurately compute the final outcome.

"This work shows that there are many potential routes to improving computations, encompassing both classical and quantum approaches," explains Dries Sels, an assistant professor in New York University's Department of Physics and one of the paper's authors. "Moreover, our work highlights how difficult it is to achieve quantum advantage with an error-prone quantum computer."

In seeking ways to optimize classical computing, Sels and his colleagues at the Simons Foundation focused on a type of tensor network that faithfully represents the interactions between the qubits. Those types of networks have been notoriously hard to deal with, but recent advances in the field now allow these networks to be optimized with tools borrowed from statistical inference.

The authors compare the work of the algorithm to the compression of an image into a JPEG file, which allows large images to be stored using less space by eliminating information with barely perceivable loss in the quality of the image.

"Choosing different structures for the tensor network corresponds to choosing different forms of compression, like different formats for your image," says the Flatiron Institute's Joseph Tindall, who led the project. "We are successfully developing tools for working with a wide range of different tensor networks. This work reflects that, and we are confident that we will soon be raising the bar for quantum computing even further."

The work was supported by the Flatiron Institute and a grant from the Air Force Office of Scientific Research (FA9550-21-1-0236).

  • Quantum Computers
  • Computers and Internet
  • Information Technology
  • Spintronics Research
  • Distributed Computing
  • Computer Science
  • Quantum computer
  • Quantum entanglement
  • Grid computing
  • Computing power everywhere
  • Quantum tunnelling
  • Introduction to quantum mechanics
  • Quantum mechanics

Story Source:

Materials provided by New York University . Note: Content may be edited for style and length.

Journal Reference :

  • Joseph Tindall, Matthew Fishman, E. Miles Stoudenmire, Dries Sels. Efficient Tensor Network Simulation of IBM’s Eagle Kicked Ising Experiment . PRX Quantum , 2024; 5 (1) DOI: 10.1103/PRXQuantum.5.010308

Cite This Page :

  • From CO2 to Biodegradable Plastics, Efficiently
  • Oldest Fossil Reptile from Alps Is Forgery
  • Salt Substitutes for Blood Pressure Control
  • Icy Dwarf Planets With Geothermal Activity
  • Reforestation Threatens Vast Tropical Grasslands
  • New Direction for Sustainable Batteries
  • A 'Quantum Leap' at Room Temperature
  • Gravastar Inside Another Gravastar?
  • Root Microbes: Secret to a Tastier Cup of Tea
  • Ancient Retroviruses and Our Big Brains

Help | Advanced Search

Computer Science > Artificial Intelligence

Title: an interactive agent foundation model.

Abstract: The development of artificial intelligence systems is transitioning from creating static, task-specific models to dynamic, agent-based systems capable of performing well in a wide range of applications. We propose an Interactive Agent Foundation Model that uses a novel multi-task agent training paradigm for training AI agents across a wide range of domains, datasets, and tasks. Our training paradigm unifies diverse pre-training strategies, including visual masked auto-encoders, language modeling, and next-action prediction, enabling a versatile and adaptable AI framework. We demonstrate the performance of our framework across three separate domains -- Robotics, Gaming AI, and Healthcare. Our model demonstrates its ability to generate meaningful and contextually relevant outputs in each area. The strength of our approach lies in its generality, leveraging a variety of data sources such as robotics sequences, gameplay data, large-scale video datasets, and textual information for effective multimodal and multi-task learning. Our approach provides a promising avenue for developing generalist, action-taking, multimodal systems.

Submission history

Access paper:.

  • Download PDF
  • Other Formats

license icon

References & Citations

  • Google Scholar
  • Semantic Scholar

BibTeX formatted citation

BibSonomy logo

Bibliographic and Citation Tools

Code, data and media associated with this article, recommenders and search tools.

  • Institution

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs .

research paper on computers


  • I-JEPA: The first AI model based on Yann LeCun’s vision for more human-like AI
  • Celebrating 10 years of FAIR: A decade of advancing the state-of-the-art through open research
  • Turing Award presented to Yann LeCun, Geoffrey Hinton, and Yoshua Bengio
  • Today, we’re publicly releasing the Video Joint Embedding Predictive Architecture (V-JEPA) model, a crucial step in advancing machine intelligence with a more grounded understanding of the world.
  • This early example of a physical world model excels at detecting and understanding highly detailed interactions between objects.
  • In the spirit of responsible open science, we’re releasing this model under a Creative Commons NonCommercial license for researchers to further explore.

As humans, much of what we learn about the world around us—particularly in our early stages of life—is gleaned through observation. Take Newton’s third law of motion: Even an infant (or a cat) can intuit, after knocking several items off a table and observing the results, that what goes up must come down. You don’t need hours of instruction or to read thousands of books to arrive at that result. Your internal world model—a contextual understanding based on a mental model of the world—predicts these consequences for you, and it’s highly efficient.

“V-JEPA is a step toward a more grounded understanding of the world so machines can achieve more generalized reasoning and planning,” says Meta’s VP & Chief AI Scientist Yann LeCun, who proposed the original Joint Embedding Predictive Architectures (JEPA) in 2022. “Our goal is to build advanced machine intelligence that can learn more like humans do, forming internal models of the world around them to learn, adapt, and forge plans efficiently in the service of completing complex tasks.”

Video JEPA in focus

V-JEPA is a non-generative model that learns by predicting missing or masked parts of a video in an abstract representation space. This is similar to how our Image Joint Embedding Predictive Architecture (I-JEPA) compares abstract representations of images (rather than comparing the pixels themselves). Unlike generative approaches that try to fill in every missing pixel, V-JEPA has the flexibility to discard unpredictable information, which leads to improved training and sample efficiency by a factor between 1.5x and 6x.

Because it takes a self-supervised learning approach, V-JEPA is pre-trained entirely with unlabeled data. Labels are only used to adapt the model to a particular task after pre-training. This type of architecture proves more efficient than previous models, both in terms of the number of labeled examples needed and the total amount of effort put into learning even the unlabeled data. With V-JEPA, we’ve seen efficiency boosts on both of these fronts.

With V-JEPA, we mask out a large portion of a video so the model is only shown a little bit of the context. We then ask the predictor to fill in the blanks of what’s missing—not in terms of the actual pixels, but rather as a more abstract description in this representation space.

research paper on computers

Masking methodology

V-JEPA wasn’t trained to understand one specific type of action. Instead it used self-supervised training on a range of videos and learned a number of things about how the world works. The team also carefully considered the masking strategy—if you don’t block out large regions of the video and instead randomly sample patches here and there, it makes the task too easy and your model doesn’t learn anything particularly complicated about the world.

It’s also important to note that, in most videos, things evolve somewhat slowly over time. If you mask a portion of the video but only for a specific instant in time and the model can see what came immediately before and/or immediately after, it also makes things too easy and the model almost certainly won’t learn anything interesting. As such, the team used an approach where it masked portions of the video in both space and time, which forces the model to learn and develop an understanding of the scene.

Efficient predictions

Making these predictions in the abstract representation space is important because it allows the model to focus on the higher-level conceptual information of what the video contains without worrying about the kind of details that are most often unimportant for downstream tasks. After all, if a video shows a tree, you’re likely not concerned about the minute movements of each individual leaf.

One of the reasons why we’re excited about this direction is that V-JEPA is the first model for video that’s good at “frozen evaluations,” which means we do all of our self-supervised pre-training on the encoder and the predictor, and then we don’t touch those parts of the model anymore. When we want to adapt them to learn a new skill, we just train a small lightweight specialized layer or a small network on top of that, which is very efficient and quick.

research paper on computers

Previous work had to do full fine-tuning, which means that after pre-training your model, when you want the model to get really good at fine-grained action recognition while you’re adapting your model to take on that task, you have to update the parameters or the weights in all of your model. And then that model overall becomes specialized at doing that one task and it’s not going to be good for anything else anymore. If you want to teach the model a different task, you have to use different data, and you have to specialize the entire model for this other task. With V-JEPA, as we’ve demonstrated in this work, we can pre-train the model once without any labeled data, fix that, and then reuse those same parts of the model for several different tasks, like action classification, recognition of fine-grained object interactions, and activity localization.

research paper on computers

Avenues for future research...

While the “V” in V-JEPA stands for “video,” it only accounts for the visual content of videos thus far. A more multimodal approach is an obvious next step, so we’re thinking carefully about incorporating audio along with the visuals.

As a proof of concept, the current V-JEPA model excels at fine-grained object interactions and distinguishing detailed object-to-object interactions that happen over time. For example, if the model needs to be able to distinguish between someone putting down a pen, picking up a pen, and pretending to put down a pen but not actually doing it, V-JEPA is quite good compared to previous methods for that high-grade action recognition task. However, those things work on relatively short time scales. If you show V-JEPA a video clip of a few seconds, maybe up to 10 seconds, it’s great for that. So another important step for us is thinking about planning and the model’s ability to make predictions over a longer time horizon.

...and the path toward AMI

To date, our work with V-JEPA has been primarily about perception—understanding the contents of various video streams in order to obtain some context about the world immediately surrounding us. The predictor in this Joint Embedding Predictive Architecture serves as an early physical world model: You don’t have to see everything that’s happening in the frame, and it can tell you conceptually what’s happening there. As a next step, we want to show how we can use this kind of a predictor or world model for planning or sequential decision-making.

We know that it’s possible to train JEPA models on video data without requiring strong supervision and that they can watch videos in the way an infant might—just observing the world passively, learning a lot of interesting things about how to understand the context of those videos in such a way that, with a small amount of labeled data, you can quickly acquire a new task and ability to recognize different actions.

V-JEPA is a research model, and we’re exploring a number of future applications. For example, we expect that the context V-JEPA provides could be useful for our embodied AI work as well as our work to build a contextual AI assistant for future AR glasses. We firmly believe in the value of responsible open science, and that’s why we’re releasing the V-JEPA model under the CC BY-NC license so other researchers can extend this work.

Our latest updates delivered to your inbox

Subscribe to our newsletter to keep up with Meta AI news, events, research breakthroughs, and more.

Join us in the pursuit of what’s possible with AI.

research paper on computers

Latest Work

Our Actions

Meta © 2024

OpenAI’s Sora video-generating model can render video games, too

research paper on computers

OpenAI’s new — and first! — video-generating model, Sora , can pull off some genuinely impressive cinematographic feats. But the model’s even more capable than OpenAI initially made it out to be, at least judging by a technical paper published this evening.

The paper, titled “Video generation models as world simulators,” co-authored by a host of OpenAI researchers, peels back the curtains on key aspects of Sora’s architecture — for instance revealing that Sora can generate videos of an arbitrary resolution and aspect ratio (up to 1080p). Per the paper, Sora’s able to perform a range of image and video editing tasks, from creating looping videos to extending videos forwards or backwards in time to changing the background in an existing video.

But most intriguing to this writer is Sora’s ability to “simulate digital worlds,” as the OpenAI co-authors put it. In an experiment, OpenAI fed Sora prompts containing the word “Minecraft” and had it render a convincingly Minecraft-like HUD and game — and the game’s dynamics, including physics — while simultaneously controlling the player character.

OpenAI Sora can simulate Minecraft I guess. Maybe next generation game console will be "Sora box" and games are distributed as 2-3 paragraphs of text. — Andrew White (@andrewwhite01) February 16, 2024

So how’s Sora able to do this? Well, as observed by senior Nvidia researcher Jim Fan ( via Quartz ), Sora’s more of a “data-driven physics engine” than a creative too. It’s not just generating a single photo or video, but determining the physics of each object in an environment — and rendering a photo or video (or interactive 3D world, as the case may be) based on these calculations.

“These capabilities suggest that continued scaling of video models is a promising path towards the development of highly-capable simulators of the physical and digital world, and the objects, animals and people that live within them,” the OpenAI co-authors write.

Now, Sora’s usual limitations apply in the video game domain. The model can’t accurately approximate the physics of basic interactions like glass shattering. And even with interactions it  can model, Sora’s often inconsistent — for example rendering a person eating a burger but failing to render bite marks.

Still, if I’m reading the paper correctly, it seems Sora could pave the way for more realistic — perhaps even photorealistic — procedurally generated games from text descriptions alone. That’s in equal parts exciting and terrifying (consider the deepfake implications, for one) — which is probably why OpenAI’s choosing to gate Sora behind a very limited access program for now.

Here’s hoping we learn more sooner rather than later.

OpenAI’s newest model Sora can generate videos — and they look decent
  • United States
  • Netherlands
  • New Zealand
  • United Kingdom

How to use Google’s genAI-powered note-taking app

Google’s experimental notebooklm lacks many of the features of more established notes apps, but its generative ai core gives it analysis and summarization superpowers..

By Howard Wen

Computerworld |

google notebooklm splash screen

Create your first notebook

Add your sources, create and manage notes, chat with notebooklm about your sources, get a source summary and kick off a related chat, have notebooklm synthesize your notes, share and collaborate on a notebook, use notebooklm... with another notes app.

Among the experimental tools that Google offers is NotebookLM , a bare-bones notes app that you use in your web browser. What makes it interesting is that it uses Google’s generative AI chatbot (variously known as Bard or Duet AI and now rebranded as Gemini) to analyze sources of text that you “feed” it, and then generate notes based on the information in them.

Here are some scenarios where you might want to use NotebookLM:

  • To turn the contents of a long document into a brief summary.
  • To extract insights from several documents that cover a certain topic. For example, you can have NotebookLM analyze documents that pertain to a project that your business is working on.
  • To bring together and summarize information that you find while doing online research, such as snippets of text from web pages.

This guide will take you through setting up and using NotebookLM. You’ll need to have a Google user account to sign up for Google Workspace Labs (if you’re not already a member) and to sign in to NotebookLM. You must be at least 18 years old and live in the US to use NotebookLM.

When using NotebookLM, keep in mind that generative AI tools are in their infancy and sometimes get facts wrong. Be sure to fact-check all responses in this or any other genAI tool. Also note that this is an early beta product, and its performance can sometimes be glitchy.

In NotebookLM , a “notebook” contains one or more notes. You can create several notebooks. So each notebook represents a project that you’re working on, containing notes that are specific to that project.

On the start page of your NotebookLM account, click New Notebook . On the small panel that opens, type a name for the new notebook.

Starting a new notebook in NotebookLM. (Click image to enlarge it.)

When you click Save or press the Enter key, NotebookLM will switch to a new page showing the workspace of your new notebook.

A new blank notebook. (Click image to enlarge it.)

The sources of text that you add for NotebookLM to analyze will appear as cards on the panel that runs down the left side of the workspace.

On the Sources panel, click ADD SOURCE or the + sign to the right of “Sources.” A small panel will open that presents three ways that you can add a text source:

  • Drive: You select a Google Docs document that’s stored in your Google Drive.
  • PDF: You upload a PDF that’s stored on your PC. (Note: you cannot add a PDF that’s stored in your Google Drive.)
  • Copied text: A blank source card opens over the workspace. You can paste text into it that you’ve copied to your PC clipboard, such as text from an email or messaging chat. You can also copy and paste in a web link for NotebookLM to analyze. And you can optionally type in additional text.

Designating a new source. (Click image to enlarge it.)

After you make your choice and go through the steps that the NotebookLM interface guides you through, your source will appear as a thumbnail on the source panel.

You can have up to 20 sources in a notebook, and each source must contain fewer than 200,000 words. NotebookLM doesn’t currently support images, media files, or complex tables or charts as sources.

Google says that the data you upload is not used to train NotebookLM and will stay private, but the company also warns, “Avoid uploading documents with any personal or sensitive information” and “Avoid uploading documents you don’t have the applicable rights to.”

A notebook can have up to 1,000 notes. To create a new note manually, click the Add note button that’s at the upper right of the Notes panel. A card for a new note will appear on the Notes panel.

A card will appear on the main workspace for each note you create. (Click image to enlarge it.)

Click this card and it’ll open as a panel over the workspace. You can type text into this panel, as well as paste text saved on your PC clipboard.

Adding text to a note. (Click image to enlarge it.)

When you’re finished, click the double-arrow icon at the lower right of the panel. Your note will once again appear as a card on the Notes panel. Click it to reopen it so you can read or edit it.

To delete a note, move the pointer over its card. Click the square that appears on its upper-right corner. This will add a blue checkmark inside the square. Then click the Delete notes button at the top of the Notes panel. To delete multiple notes at once, select their individual note cards (or click the Select all button) and click Delete notes .

Deleting notes from a notebook. (Click image to enlarge it.)

With those basics out of the way, we can get to the true power of NotebookLM: its built-in chat with genAI capabilities.

Below the Notes area is a chat panel, and this is the real point of using NotebookLM. Type inside the entry box and click the arrow to the right to make a request to the AI about your text sources. The best way to word your requests is in the form of a question.

The AI also provides three suggested queries above the entry box; these are based on information in your sources. Click a suggestion and it’ll be posted in the chat window.

Type in a question or choose one of the suggested queries. (Click image to enlarge it.)

By default, NotebookLM analyzes all the sources in a notebook when formulating a response. You can click the blue checkbox on any source’s thumbnail to deselect it, and the AI will ignore that source. To remove a source entirely, move the pointer over its thumbnail, click the three-dot icon, and select Remove source .

When you enter a request, the AI will take a few seconds to process it and post a response.

NotebookLM’s response to a query. (Click image to enlarge it.)

There are a few ways to interact with response card:

Pin: Click the pin icon at the upper right to turn the AI’s response into a note card that will appear on the Notes panel. You can then click the card to open and read the note, but you cannot edit it.

Copy: Click this icon at the lower right to copy the text of the AI’s response to your PC clipboard. You can then paste the response into a document or a note you create manually, which you will be able to edit.

Thumbs Up/Down: Click either of these icons at the lower right to rate how good you think the AI’s response is. This helps to train the AI to give you better results for future requests.

Citations: At the lower left is a button telling you the number of citations — text segments from one or more sources — used to create the response. When you click the button, a number appears for each citation. Moving the pointer over a citation’s number opens a panel that shows you the source text.

Citations show specific chunks of text that Duet AI drew from to generate its response. (Click image to enlarge it.)

When you click a source’s thumbnail on the Sources panel, the source card will expand to fill the left half of the workspace and NotebookLM will generate a source guide — a summary that describes the contents of the source.

Duet AI creates a summary of the source and a list of key topics. (Click image to enlarge it.)

Note: if you created a copied text source card that includes a web link, NotebookLM will try to generate a summary of the content on that web page.

To the right of the summary is a list of key topics. Clicking any of these sends a request about the topic that the AI will respond to in the chat window.

Once you have a number of notes collected in a notebook, you can ask NotebookLM to do something new with them, such as summarize them all or create an outline from them. In the main Notes workspace, select the cards for the notes you want to draw from, and NotebookLM shows several suggested actions above the chat window, such as Summarize , Suggest related ideas , and Create outline .

NotebookLM offers suggestions for synthesizing selected notes. (Click image to enlarge it.)

Click one of the suggestions — or type your own directive into the chat window, such as “Create a bulleted list of key points” — and the AI will generate an appropriate response. Some responses you generate this way are automatically saved as a new note; in other cases you’ll need to pin or copy them if you want to save them.

You can share a notebook with others, either restricting their permissions so that they can only view its contents or allowing them to edit it, such as by adding, editing, or removing notes and sources. Everyone you share a notebook with can interact with the AI and copy its responses to their clipboard, but only those with Editor status can add a response as a note.

To share a notebook, click the angled-line icon that’s toward the upper right of your notebook’s workspace. A Share panel opens over the workspace.

Sharing a notebook. (Click image to enlarge it.)

Inside the entry box, type the name or email address of someone who’s in your Google contacts. (Note: they must already be in your Google contacts. You cannot type in any email address here.) When their name appears, click it or press the Enter key to add them to the “People who have access” list.

By default, a person you add will be granted Viewer access to your notebook. If you want them to collaborate with you on the notebook, click Viewer and change it to Editor .

When you are done adding people to share your notebook with and setting their access levels, click the Send button. They’ll be notified by email that you’re sharing this notebook with them.

To re-emphasize, NotebookLM in its current form is bare bones. It lacks several features of a typical notes app. There are no task lists, nor can you add images to your notes. And even if you add a web address to a note, it won’t turn into a clickable link.

This raises the question: will more features be added to NotebookLM? Or could its AI tool for generating notes — which is currently the sole reason for using it — be rolled into Google’s existing notes app, Google Keep ?

While Google sorts this out, remember that you can always copy the text of a response that the AI generates to your clipboard. Then you can paste it into a note in Google Keep or another notes app. In this way, NotebookLM works well as a companion tool.

  • How to use the new AI writing tool in Google Docs and Gmail
  • How to use the new genAI template tool in Google Sheets
  • Generative AI
  • Productivity Software
  • Google Workspace

Howard Wen is a longtime contributor to Computerworld . He specializes in explainer guides, how-tos, and reviews of office applications and productivity tools.

Copyright © 2024 IDG Communications, Inc.

Google Workspace power tips

research paper on computers


  1. (PDF) A Study on Artificial Intelligence Technologies and its

    research paper on computers

  2. 😊 Research paper about computer. Research Paper About Computer. 2019-02-20

    research paper on computers

  3. 😍 Computer virus research paper. Research Papers on Computer Viruses

    research paper on computers

  4. ⇉Computer Research Paper Essay Example

    research paper on computers

  5. How to read a computer science research paper?

    research paper on computers

  6. 😂 Distributed computing research papers. Distributed Computing Research

    research paper on computers


  1. Check out the details of my research paper 😇

  2. Step-by-step approach to starting and completing a good research paper

  3. How to Write a Scientific Research Paper

  4. Difference between review paper and research paper #shorts #short #trending #india

  5. Tech Science Press Reference Style

  6. BEST AI TOOLS FOR RESEARCH PAPER WRITING, Assignment, Article review and literature 2023 in Amharic


  1. Computer Science

    Covers all theoretical and applied aspects at the intersection of computer science and game theory, including work in mechanism design, learning in games (which may overlap with Learning), foundations of agent modeling in games (which may overlap with Multiagent systems), coordination, specification and formal methods for non-cooperative computa...

  2. Computer science

    Computer science is the study and development of the protocols required for automated processing and manipulation of data. This includes, for example, creating algorithms for efficiently...

  3. computer science Latest Research Papers

    computer science Latest Research Papers | ScienceGate computer science Recently Published Documents TOTAL DOCUMENTS 14255 (FIVE YEARS 3547) H-INDEX 73 (FIVE YEARS 9) Latest Documents Most Cited Documents Contributed Authors Related Sources Related Keywords Hiring CS Graduates: What We Learned from Employers ACM Transactions on Computing Education

  4. PDF Technology and Education: Computers, Software, and the Internet

    1 A growing number of are experimenting with oneschools -to-one laptop or tablet programs that provide a computer to each student and often allow the student to take the computer home Warschauer 2006; Maine Education Policy Research Institute( 2007; Texas Center for Educational Research 2009). 2

  5. 533984 PDFs

    Explore the latest publications in Computer Science, and find Computer Science experts. Filters Sorted by most recent Walert: Putting Conversational Search Knowledge into Action by Building and...

  6. Computer Science and Engineering

    Research Full-text available Sep 2022 Boopathi .D Dear Researcher, Kindly find the link for Second International Conference on Computer Science, Engineering and Applications (ICCSEA)...

  7. 40 years of quantum computing

    40 years of quantum computing Nature Reviews Physics 4 , 1 ( 2022) Cite this article 21k Accesses 8 Citations 51 Altmetric Metrics This year we celebrate four decades of quantum computing by...

  8. Computers

    Computers is an international, scientific, peer-reviewed, open access journal of computer science, including computer and network architecture and computer-human interaction as its main foci, published monthly online by MDPI. Open Access — free for readers, with article processing charges (APC) paid by authors or their institutions.

  9. PDF MIT Open Access Articles Quantum computing

    lenges in physics and computer science, ranging from hard‑ ware architectures and data management to application soft‑ ware and algorithms, which requires fundamental research in all these areas and beyond (Almudever et al., 2017). To inform information systems (IS) research, this Fun‑ damental provides the fundamental concepts of quantum

  10. Publications

    Distributed Systems and Parallel Computing 340. Economics and Electronic Commerce 340. Education Innovation 68. General Science 328. Hardware and Architecture 145. Health & Bioscience 358. Human-Computer Interaction and Visualization 803. Information Retrieval and the Web 414. Machine Intelligence 3780.

  11. Top Ten Computer Science Education Research Papers of the Last 50 Years

    1. " Identifying student misconceptions of programming " (2010) Lisa C. Kaczmarczyk, Elizabeth R. Petrick, University of California, San Diego; Philip East, University of Northern Iowa; Geoffrey L. Herman, University of Illinois at Urbana-Champaign Computing educators are often baffled by the misconceptions that their CS1 students hold.

  12. A comprehensive review study of cyber-attacks and cyber security

    Research paper. A comprehensive review study of cyber-attacks and cyber security; Emerging trends and recent developments. Author ... Information operations integrated use of the main capabilities of electronic warfare, psychological, computer network, military trickery and security operations in coordination with special support and relevant ...

  13. Computers & Education

    Computers & Education aims to increase knowledge and understanding of ways in which digital technology can enhance education, through the publication of high-quality research, which extends theory and practice. The Editors welcome research papers on the pedagogical uses of digital technology, where the focus is broad enough to be of interest to a wider education community.

  14. Quantum computers

    Initialization is an important challenge for nuclear magnetic resonance quantum computers. The first proposals employed pseudo-pure-state techniques, which isolate the signal of an initialized ...

  15. Computers & Graphics

    1. Research and applications of interactive computer graphics. … View full aims & scope $2690 Article publishing charge for open access 2 days Time to first decision 78 days Review time 105 days Submission to acceptance View all insights Editor-in-Chief View full editorial board Joaquim Jorge, PhD

  16. Computer Science Research Papers

    Computer Science Research Papers - Computer Science Using the Pedagogy of Technology for Student Success through Media Centers, Flipping, Multiple Intelligences theory and other such "Rott" in Teacher Education Media Centers offer an opportunity to diversify optional delivery methods.

  17. Influence of computers in students' academic achievement

    We propose a model on the influence of computer attitudes, computer learning environments, computer learning motivations, computer confidence, computer use, computer self-efficacy, loneliness, mothers' education, parents' marital status and family size on academic achievement (AA).

  18. Analysis of Students' learning of computer programming in a computer

    The purpose of this paper is to present our approach to investigating students' learning of computer programming, to gain a more detailed understanding of this interaction. We collect empirical data in contexts where students work with computer programming exercises in a computer laboratory, using the programming language Java, and we then ...

  19. Brain-Computer Interface: Advancement and Challenges

    Brain-Computer Interface (BCI) is an advanced and multidisciplinary active research domain based on neuroscience, signal processing, biomedical sensors, hardware, etc. Since the last decades, several groundbreaking research has been conducted in this domain. ... The future of brain-computer interfacing (keynote paper); Proceedings of the 2016 ...

  20. Quantum Computing Review: A Decade of Research

    Quantum computing (QC) has the potential to be the next abstruse technology, with a wide range of possible applications and ramifications for organizations and markets. QC provides an exponential speedup by employing quantum mechanics principles, including superposition and entanglement. The potential advantages offered by the revolutionary paradigm have propelled scientific productions ...

  21. Computer Technology Research Paper Topics

    This list of computer technology research paper topics provides the list of 33 potential topics for research papers and an overview article on the history of computer technology. 1. Analog Computers

  22. Researchers show classical computers can keep up with, and surpass

    The work was supported by the Flatiron Institute and a grant from the Air Force Office of Scientific Research (FA9550-21-1-0236). RELATED TOPICS Computers & Math

  23. [2402.05929] An Interactive Agent Foundation Model

    The development of artificial intelligence systems is transitioning from creating static, task-specific models to dynamic, agent-based systems capable of performing well in a wide range of applications. We propose an Interactive Agent Foundation Model that uses a novel multi-task agent training paradigm for training AI agents across a wide range of domains, datasets, and tasks. Our training ...

  24. Computers & Operations Research

    Computers & Operations Research open access 4.6 Impact Factor Guide for authors About the journal Aims & Scope Now incorporating Surveys in Operations Research & Management Science Operations research and computers interact in many scientific fields of vital importance to our society.

  25. V-JEPA: The next step toward advanced machine intelligence

    V-JEPA is a non-generative model that learns by predicting missing or masked parts of a video in an abstract representation space. This is similar to how our Image Joint Embedding Predictive Architecture (I-JEPA) compares abstract representations of images (rather than comparing the pixels themselves). Unlike generative approaches that try to fill in every missing pixel, V-JEPA has the ...

  26. OpenAI's Sora video-generating model can render video games, too

    The paper, titled "Video generation models as world simulators," co-authored by a host of […] OpenAI's new — and first! — video-generating model, Sora, can pull off some genuinely ...

  27. COSE

    About the journal. Computers & Security is one of the most respected journals in IT security, being recognized worldwide as THE primary source of reference for IT security research and applications expertise. Computers & Security provides the IT security community with a unique blend of leading edge research and sound practical management advice.

  28. How to use Google's genAI-powered note-taking app

    To bring together and summarize information that you find while doing online research, such as snippets of text from web pages. This guide will take you through setting up and using NotebookLM.