Intelligent Tutoring & Other Educational Uses of AI

AITopics > Education


"If we understand the human mind, we begin to understand what we can do with educational technology."
- Herbert A. Simon

How do we learn? What makes a teacher effective? What unique qualities does a computer based system have to offer, and what are its limitations? What features should the best model for an intelligent tutoring system possess? AI scientists involved in education address these and other questions in three basic contexts:

teacher and student

"The first is to use AI and cognitive science techniques to model experts who problem solve in a domain, as well as tutors teaching and students learning in that domain. . . . The second research goal involves explaining learning and teaching as parts of the human information-processing system. . . . The third research goal is to demonstrate completeness and reliability in the engineering side of the discipline and to show that intelligent instructional systems can be used effectively in training and classroom situations."
(Shapiro, Stuart C., editor. 1992. Encyclopedia of Artificial Intelligence. 2nd ed. New York: John Wiley & Sons. Volume 1, page 434.)

Introductory Readings

Colleges see the future in technology - Online classes are only the start of how schools are looking at ways to use high-tech gear and change the way students learn. By Stuart Silverstein. Los Angeles Times (September 12, 2006). "Though long known for their adherence to tradition, colleges in California and elsewhere increasingly are embracing a variety of higher-tech approaches to teaching and learning. And new gizmos, including gear with cutting-edge videogame or artificial intelligence technology, are on the way to provide more individualized instruction. Some of the most futuristic devices -- if colleges are adventurous enough to try them -- could even monitor students' brainwaves to keep track of how they're learning. The trend toward electronic technology could be particularly dramatic in California, where demographic and economic forces are likely to promote ways to stretch the state's educational resources. ... [Carol] Twigg's outlook is based partly on her center's four-year effort with 30 colleges to redesign high-enrollment courses. The 30 projects involved such things as deemphasizing lectures and relying more on online tutorials and discussion forums, along with using computerized grading to give students speedier assessments of what they were learning well and what they were getting wrong. The result: Student learning rose in 25 of the 30 projects. ... At the same time, the cost of providing instruction was reduced an average 37%. ... Artificial intelligence experts are developing 'intelligent tutoring systems' that can help students working on problems in, say, algebra or physics where there are many ways to solve a problem -- as well as many ways to go wrong. ... The technologies have 'tremendous potential,' said Carnegie Mellon's [Joel M.] Smith. But if developers of the technologies don't keep in mind what psychologists and other experts have discovered about how people learn, he warned, 'there's a danger in going off the rails.'"

Top 10 Smart Technologies for Schools. Technology & Learning (November 2002). "What's a 'smart' technology? While one might argue that all technology -- from a toaster to a moon rover is smart, those we present in the following Top 10 list meet their own set of criteria. In contrast to the breakthroughs we profiled last year, which included such broad topics and trends as wireless and virtual learning, the technologies we've chosen to examine here perform more specific, identifiable functions. Fingerprint recognition and artificial intelligence can free educators and school staff of time-consuming tasks. Telementoring and virtual reality enable collaborations and instant expert guidance from any spot on earth. And voice-to-text technology and hybrid devices support young and challenged learners in formerly unheard of ways. In the hands of well-trained educators, these technologies can offer powerful new solutions for teaching children." One of the 10 articles offered is "Artificial Intelligence", by Kristen Kennedy: "They don't do windows -- but the next generation of AI applications can teach, tutor, and even grade essays."

Human Beings Not As Impressive As You Think. By Kevin Drum. Mother Jones (Apr. 29, 2012 ). "A recent study suggests that computers can score student essays about as well as human beings. Les Perelman, a director of writing at MIT, isn't impressed. While his research is limited, because E.T.S. is the only organization that has permitted him to test its product, he says the automated reader can be easily gamed, is vulnerable to test prep, sets a very limited and rigid standard for what good writing is, and will pressure teachers to dumb down writing instruction. The e-Rater’s biggest problem, he says, is that it can’t identify truth. ... In the end, this is my big difference with the AI naysayers: I'm just not as impressed by human intelligence as they are. All those human essay graders probably think they're making use of deep human values and intelligence as they score those essays, but in fact they're mostly just applying a few hundred (or maybe a few thousand) linguistic algorithms they've learned over the years and spitting out a number. "

Computers in Education: A Brief History. By Andrew Molnar. THE Journal (Technological Horizons in Education) June 1997.

Encyclopedia of Educational Technology. Bob Hoffman, General Editor; a publication of San Diego State UniversityDepartment of Educational Technology. "The Encyclopedia of Educational Technology (EET) is a collection of short multimedia articles on a variety of topics related to the fields of instructional design and education and training. The primary audiences for the EET are students and novice to intermediate practitioners in these fields, who need a brief overview as a starting point to further research on specific topics. Authors are graduate students, professors, and others who contribute voluntarily. Articles are short and use multimedia to enrich learning rather than merely decorate the pages."

Experts Use AI to Help GIs Learn Arabic. By Eric Mankin. USC News (June 21, 2004). " To teach soldiers basic Arabic quickly, USC computer scientists are developing a system that merges artificial intelligence with computer game techniques. The Rapid Tactical Language Training System, created by the USC Viterbi School of Engineering's Center for Research in Technology for Education (CARTE) and partners, tests soldier students with videogame missions in animated virtual environments where, to pass, the students must successfully phrase questions and understand answers in Arabic. ... 'Most adults find it extremely difficult to acquire even a rudimentary knowledge of a language, particularly in a short time,' said CARTE director W. Lewis Johnson. 'We’re trying to build an improved model of instruction, one that can be closely tailored to both the needs and the abilities of each individual student,' Johnson said." Read the story.

General Readings

International Artificial Intelligence in Education Society (IJAIED): Links to journal papers, AIED conference proceedings and reports on AIED conference workshops, reports on other events of interest to Society members; bibliographies of fields of AIED; etc.

  • Also available:
    • Proceedings of the A-IED 97 workshop: Intelligent Educational Systems on the World Wide Web
    • VEST-Lab. Benjamin Zayas, School of Cognitive and Computing Sciences, University of Sussex. "This paper [presented at AIED 2001] describes the design decisions made during the development of VEST-Lab, a desktop Virtual Reality (VR) system for safety training in chemistry laboratories. Realism afforded by VR representations and interactivity issues in desktop VR systems are discussed from the point of view of learner-centred design perspective."

Educational Data Mining: Papers from the 2005 AAAI Workshop, ed. Joseph E. Beck, 24-31. Technical Report WS-05-02. American Association for Artificial Intelligence, Menlo Park, California. "This workshop focuses on leveraging the ability of computer tutors to record their interactions with students to better understand how to teach students. Computer tutors are capable of recording both longitudinal data, as well as data at a fine-time scale, such as mouse clicks and response time data. Using these interactions as a source of data to be mined provides a new view into understanding student learning processes. The first objective of this workshop is to bring together researchers working at the intersection of AI and education to discuss how to better understand and learn from the data we are collecting. The second major objective of the workshop is to create a repository of data sets for educational data mining. The lack of shared data sets has hampered the field both by creating a high barrier to entry and by making it difficult for researchers to directly compare techniques and results."

The Promise of Artificial Intelligence. Carnegie Mellon Today ( Sept. 2004, Vol. 1, No. 3). "In a remote African hut, a mother clutches her collapsed child in dread. She’s seen the symptoms before. They’re almost always fatal. She describes them aloud in her native tongue. Across the room, a computer screen displays the face of her'pediatrician' He asks her a series of questions and then prescribes the cure. It’s water. Today, Africa has the highest mortality rate for children under five. Most times, death is from preventable problems like dehydration. But medical help is seldom close enough to matter. A proposal before the World Bank would change that by placing cognitive tutors in every hut of every village on the African continent. That’s the long-term dream of Carnegie Mellon’s Herbert A. Simon University Professor of Computer Science and Robotics Raj Reddy, who like the late visionary Herb Simon, believes education technology can cure many of the world’s ills. With the help of World Bank funding, Reddy hopes to deliver on this promise for $1 per dwelling. But, in the near term, cognitive tutors are helping underprivileged children learn math three times faster and perform much better on tests. Intelligent tutors are now in 4 percent of the nation’s schools and attracting wide interest among government agencies."

Smart Machines in Ed. book cover

Smart Machines in Education. Edited by Kenneth D. Forbus and Paul J. Feltovich. AAAI Press. "Multimedia, simulation, computer-mediated communication networks, and distance learning have all become part of the educational toolkit. The next major technology to change the face of education will be based on the widespread use of artificial intelligence (AI). Progress in AI has led to a deeper understanding of how to represent knowledge, to reason, and to describe procedural knowledge. Progress in cognitive science has led to a deeper understanding of how people think, solve problems, and learn. AI scientists use results from cognitive science to create software with more humanlike abilities, which can help students learn better.

This book looks at some of the results of this synergy among AI, cognitive science, and education."

"The International Journal of Artificial Intelligence in Education (IJAIED) is the official journal of the International AIED Society. IJAIED publishes papers and other items concerned with the application of artificial intelligence techniques and concepts to the design of systems to support learning."

Conceptualizing Intelligent Agents For Teaching and Learning. By Ali Jafar. Educause Quarterly. Volume 25, Number 3 (2002). "An intelligent agent is a set of independent software tools linked with other applications and databases running within one or several computer environments. The primary function of an intelligent agent is to help a user (client) better use, manage, and interact with a computer application such as a CMS [course management software] or campus portal system. Additionally, software agents, like human agents (for example, a secretary or an administrative assistant), can be authorized with the autonomy to make decisions and perform certain tasks. Agent-based technology systems are assumed to involve artificial intelligence (AI) and include a degree of autonomous problem-solving ability."

Enter the Computer Tutor - PCs Can Help Kids Pass No Child Left Behind Tests. By Angie C. Marek.'s E-Learning Guide (November 2005). "In a K-12 educational environment that is ever more focused on metrics-based achievement, computers -- with their massive processing power and number crunching abilities -- can become tireless teaching aides. 'These things can basically work for you as a testing administrator, a one-on-one tutor, or just a very well-organized storage space for massive amounts of data on your students,' says Jeff Cameron, the brain behind Brainchild. The versatility computers bring to the classroom is on full display in Wooster, Mass., where several classes of eighth graders are currently testing the Assistment program, a software tool designed by professors at Carnegie Mellon University. The program saves teachers time by administering and automatically scoring quizzes that are similar to Massachusetts's state achievement test. And when kids get tripped up on a problem, the PC breaks a larger problem down into step-by-step questions, trying to tease out exactly where the students got off track. ... And teachers don't necessarily feel supplanted by the high-tech gadgetry. Kate Ross, a seventh- and eighth-grade English teacher in Highland, Utah, says she sometimes thinks of the computer as a 'teaching assistant,'...."

Visual-Syntactic Text Formatting: A New Method to Enhance Online Reading. By Stan Walker,Phil Schloss,Charles R. Fletcher,Charles A. Vogel, and Randall C. Walker. Reading Online (May 2005). Abstract: "A new process, visual-syntactic text formatting (VSTF), transforms block-shaped text into cascading patterns that help readers identify grammatical structure. The new method integrates converging evidence from educational, visual, and cognitive research, and is made feasible through computer-executed algorithms and electronic displays. Among college readers, the VSTF method instantly increased reading comprehension and efficiency of reading online text, while reducing eyestrain. Among high school students, who read with the format over an entire academic year, the VSTF method increased both academic achievement and long-term reading proficiency by more than a full standard deviation over randomized controls."

  • Visit Walker Reading Technologies' Live Ink and check out their demo.

Related Resources

Advanced Learning Technologies (ALT): "Through the Advanced Learning Technologies (ALT) program, the CISE and EHR Directorates of NSF support research that (1) enables radical improvements in learning through innovative computer and information technologies, and (2) advances research in computer science, information technology, learning, and cognitive science through the unique challenges posed by learning environments and learning technology platforms. Integrative research approaches that build across disciplines and establish tight linkages among theory, experiment, and design are strongly encouraged. Technology goals may include systems for tutoring or assessment, modeling and sensing of cognitive or emotional states, context awareness, natural language interfaces, collaboration, knowledge management, and non-traditional goals that redefine the roles of technology in learning." Be sure to scroll down their page to the link to abstracts of awards made through ALT and related programs.

The International Conference on Artificial Intelligence in Education. "The International AIED Society organizes a biennial AIED conference, which is the main forum for reporting the best international research in the field of AI in Education. The conferences provide the opportunity for the exchange of information and ideas on related research, development and applications. Society members receive reduced registration rates for AIED conferences. AIED 97 was held in Kobe, Japan; AIED 99 was held in Le Mans, France; AIED 01 was held in San Antonio, USA; AIED 03 was held in Sydney, Australia. AIED 05 (the 12th International Conference on Artificial Intelligence in Education) was held in Amsterdam, The Netherlands."

  • Here are some of the topics for AIED2007:
    • " The technical program focuses on research linking theory and technology from artificial intelligence, cognitive science, and computer science with theory and practice from education and social science. Areas of interest include, but are not limited to:
    • Socially informed design: Social dimensions of learning. Social-historical-cultural contexts. Learning and identity. Motivation and engagement in learning. Informal learning environments.
    • Collaborative and group learning: Group learning environments. Networked learning communities. Analysis and modeling of group interactions. Design principles for collaborative learning environments. Communities of learners. Communities of practice.
    • Learning systems platforms and architectures: Web based learning platforms. Metadata standards for learning objects and materials. Document management for learning applications. Authoring tools and assessment tools.
    • Modeling and representation: Models of learners, facilitators, tasks and problem-solving processes. Knowledge representation and ontologies. Discourse representation and analysis. • Intelligent tutoring and scaffolding: Adaptive environments (web-based and others). Pedagogical agents. Cognitive diagnosis. Instructional planning. Motivational diagnosis and feedback. Data mining and machine learning.
    • Interaction design and novel interfaces: Ubiquitous computing/mixed reality learning environments. Virtual and 3D learning and training environments. Multi-modal interfaces for learning. Educational multimedia systems. • Special application fields: Language learning. Mathematics and science education. Industrial, medical and other applications.
    • Methods, tools and techniques for effective evaluation of cognitive, meta-cognitive and affective issues.
    • The design and modeling of learning contexts and their impact on learning. The development of systems that encompass multiple learning contexts, including mobile learning applications."
  • ... and there have been Workshops:
    • Educational Data Mining
    • SWEL'07: Ontologies and Semantic Web Services for Intelligent Distributed Educational Systems
    • Modeling and Scaffolding Affective Experiences to Impact Learning
    • Assessment of Group and Individual Learning through Intelligent Visualization (AGILeViz)
    • AIED Applications in Ill-Defined Domains
    • Emerging Technologies for Inquiry-Based Learning in Science
    • Metacognition and Self-Regulated Learning in Intelligent Tutoring Systems
    • Narrative Learning Environments

"ARIES, the Laboratory for Advanced Research in Intelligent Educational Systems [at the University of Saskatchewan], is a focal point for research projects in the areas of intelligent tutoring systems and adaptive learning environments. The mission of the ARIES Laboratory is to advance the development of Learning Technologies through the integration of Artificial Intelligence techniques and to advance Artificial Intelligence research through attempts to solve real-world education and training problems." Be sure to check out their many projects, both present and past.

Carnegie Learning. "Based on over 20 years of research at Carnegie Mellon University, our Cognitive Tutor® mathematics solutions integrate advanced cognitive research, interactive software sessions and engaging classroom lessons to immerse students in the experience of mathematics."

Center for Advanced Research in Technology for Education (CARTE), Information Sciences Institute at the University of Southern California. "A major current thrust at CARTE is 'socially aware learning environments,' which emulate aspects of human social interaction in engaging with learners and which support social learning processes. ... CARTE is a multidisciplinary center that includes artificial intelligence specialists, multimedia designers, cognitive psychologists, and educational researchers."

Learning Systems Institue (LSI) Florida State University. Laura Lang, Director. " Located on the campus of Florida State University in Tallahassee, Florida, the Learning Systems Institute is a multidisciplinary organization dedicated to bridging the gap between research and practice in education and training. The institute develops practical and workable solutions in learning and performance, based on solid research. "

"In the DEMO Lab [Dynamical & Evolutionary Machine Organization] at Brandeis University's Computer Science Department, we have directed a number of our research projects towards educational technologies. We synthesize several components of our work, uniting a unique combination of technologies and theories from which we are building the educational environment. We start with our experience in machine learning, where we have co-evolved software agents to play games. We add to this our success using the Internet to facilitate a new form of co-evolution between humans and software agents. Finally, we carry these concepts into the educational arena."

The Education Arcade (TEA): "The Education Arcade explores games that promote learning through authentic and engaging play. TEA’s research and development projects focus both on the learning that naturally occurs in popular commercial games, and on the design of games that more vigorously address the educational needs of players. Our mission is to demonstrate the social, cultural, and educational potentials of videogames by initiating new game development projects, coordinating interdisciplinary research efforts, and informing public conversations about the broader and sometimes unexpected uses of this emerging art form in education. Education Arcade projects have touched on mathematics, science, history, literacy, and language learning, and have been tailored to a wide range of ages. They have been designed for personal computers, handheld devices and on-line delivery."

EUROCALL Special Interest Group. "We feel that research in Artificial Intelligence, Computational Linguistics, Corpus-Driven and Corpus Linguistics, Formal Linguistics, Machine Aided Translation, Machine Translation, Natural Language Interfaces, Natural Language Processing, Theoretical Linguistics has produced results which have proven, are proving and will prove very useful in the field of Computer-Assisted Language Learning [CALL]. A SIG in Language Processing can therefore facilitate information exchange between individual EUROCALL members interested in or already working in one of the above fields."

"The Future of Learning Group [MIT Media Laboratory] explores how new technologies can enable new ways of thinking, learning, and designing. The group creates new 'tools to think with' and explores how these tools can help bring about change in real-world settings, such as schools, museums, and under-served communities."

International Conference for Interaction Design and Children. In cooperation with ACM SIGCHI and hosted by the University of Maryland and the Human-Computer Interaction Lab. "IDC 2004 will focus on addressing the challenges of interaction design for children and explore the opportunities and impact that new technologies can afford young people in their everyday lives. ... 1. Emerging technologies for children (e.g., educational simulations, online games, mobile communications devices, wireless embedded technologies, authoring/programming tools). 2. The impact these technologies can have on children's lives (e.g., in schools, at home, in public spaces). 3. New research methods which give children a voice in the design, development, and evaluation processes (e.g., participatory design methods, usability testing, etc.)."

Laboratoryfor Interactive Learning Technologies, Department of Informationand Computer Sciences, University of Hawaii at Manoa: "Researchin LILT addresses issues in artificial intelligence, cognitive science,and human-computer interaction, while applying methodologies fromthese fields to develop improved educational software. Several researchprojects are unified by the theme of representational affordancesfor collaborative learning."

"The Journal of Interactive Learning Research (JILR) publishes papers related to the underlying theory, design, implementation, effectiveness, and impact on education and training of the following interactive learning environments: authoring systems, cognitive tools for learning computer-assisted language learning, assessment systems, computer-based training, computer-mediated communications, computer-supported collaborative learning, distributed learning environments, electronic performance support systems, interactive learning environments, interactive multimedia systems, interactive simulations and games, intelligent agents on the Internet, intelligent tutoring systems, microworlds, virtual reality based learning systems."

The Teachable Agents Group at Vanderbilt University. "Our project's objective is to combine insights from recent work in computer science, psychology and education to create and study 'teachable agent' (TA) environments in mathematics and science that are motivating to students, intuitive to teachers and parents, and lead to high degrees of student learning. The hallmark of these environments is that students learn by instructing "teachable agents" who then venture forth in simulation-based exploratory environments and attempt to solve problems that require knowledge relevant to the disciplines of mathematics or science. If the agents have been taught properly they solve the problems; otherwise they need to be educated further. The simulation-based environments are carefully designed to focus attention on important concepts in science and mathematics, and to make explicit the errors that occur during problem solving. Students 'scout' the problem solving requirements of various environments before attempting to teach their agents. Additional help and coaching agents are available to point students in the right direction when they make errors or produce sub-optimal solutions. ... [Our project] will inform education by exploring design principles that are student, teacher and parent friendly, and that help all students achieve."

TECFA (Technologies de Formation et Apprentissage) Overview. "TECFA is a research and teaching unit created in 1989 by the Faculté de Psychologie et des Sciences de l'Education ('School of Psychology and Education') of the University of Geneva. It is active in the field of educational technology. TECFA's research covers a large area of interests, including: cognitive issues in learning technology, computer-supported collaborative learning, virtual learning environments, computer-mediated communication, information systems in education, and distance education."

Other References Offline

Hoffman, Robert R., Gavan Lintern, and Susan Eitelman. 2004. The Janus Principle. IEEE Intelligent Systems 19(2). "Because of traditions including the separation of disciplines (for example, instructional design in education versus control systems design in cognitive engineering), there is a widespread distinction between training systems (TSs) and performance support systems (PSSs) . ... Neither a TS nor a traditional PSS on its own can accomplish this fit between human, machine, and context. In his Unified Theories of Cognition, Allen Newell argued that people learn continuously, so in his SOAR cognitive architecture, he modeled learning as a result of problem solving, implying performance. Thus, learning and problem solving (or performance) are necessarily intertwined."

Smith, Matt, Alan Smaill, and Geraint A. Wiggins, editors. 1993. Proceedings of a Workshop held as part of AI-ED 93, World Conference on Artificial Intelligence in Education on Music Education: An Artificial Intelligence Approach. London, UK: Springer-Verlag. [Link is to the ACM Portal.]

Tags: Education
AAAI   Recent Changes   Edit   History   Print   Contact Us
Page last modified on April 29, 2012, at 12:23 PM