Jim E. Greer

Student Modelling: The Key to individualized Knowledge-Based Instruction

1. Background.- 1. The State of Student Modelling.- 2. Artificial Intelligence Techniques for Student Modelling.- 2. Granularity-Based Reasoning and Belief Revision in Student Models.- 3. Student Modelling Through Qualitative Reasoning.- 4. Modeling the Student in Sherlock II.- 5. Using Machine Learning to Advise a Student Model.- 6. Building a Student Model for an Intelligent Tutoring System.- 3. Human Cognition and Student Modelling.- 7. Constraint-Based Student Modeling.- 8. Strengthening the Novice-Expert Shift Using the Self-Explanation Effect.- 9. Diagnosing and Evaluating the Acquisition Process of Problem Solving Schemata in the Domain of Functional Programming.- 4. Formalizing Student Modelling.- 10. Modelling a Students Inconsistent Beliefs and Attention.- 11. A Formal Approach To ILEs.- 12. Formal Approaches to Student Modelling.- 5. Epilogue.- 13. Re-Writing Cartesian Student Models.

Multi-Agent Multi-User Modeling in I-Help

This paper describesthe user modeling approach applied in I-Help, a distributed multi-agent based collaborative environment for peer help. There is a multitude of user modeling information in I-Help, developed by the various software agents populating the environment. These ‘user model fragments’ have been created in a variety of specific contexts to help achieve various goals. They are inherently inconsistent with one another and reflect not only characteristics of the users, but also certain social relationships among them. The paper explores some of the implications of multi-agent user modeling in distributed environments.

The Intelligent Helpdesk: Supporting Peer-Help in a University Course

Universities, experiencing growths in student enrollment and reductions in operating budgets, are faced with the problem of providing adequate help resources for students. Help resources are needed at an institution-wide and also at a course-specific level, due to the limited time of instructors to provide help and answer questions. The Intelligent IntraNet Peer Help Desk provides an integration and application of previously developed ARIES Lab tools for peer help to university teaching. One of its components, CPR, provides a subject-oriented discussion forum and FAQ-list providing students with electronic help. Another component, PHelpS, suggests an appropriate peer to provide human help. In both cases it is peer help, since the help originates from students themselves. The selection of the appropriate help resource (electronic or human) is based on modelling student knowledge and on a conceptual model of the subject material.

The State of Student Modelling

This review of the field of student modelling covers the basic concepts of overlays, bugs, and various more recent modelling representations. The function of student modelling is analyzed within the context of the original intelligent tutoring system architectures and more recent elaborations of that architecture such as intelligent learning environments and case-based tutorial systems. The authors explore issues surrounding cognitive modelling and model building. The contributions to student modelling from various research areas are outlined. It is concluded that student modelling is a vital research area underpinning future developments in intelligent learning environments and tutoring systems.

User Modelling in I-Help: What, Why, When and How

This paper describes user modelling in I-Help, a system to facilitate communication amongst learners. There are two I-Help components: Private and Public Discussions. In the Private Discussions learners take part in a one-on-one interaction with a partner (possibly a peer). The Public Discussions are open - everyone in the group has access to all discussion forums relevant to that group. The Public Discussions are most suited to discussion of issues where there might be a variety of valid viewpoints, or different solutions to a problem. It is also useful for straightforward questions and answers that have wide-spread applicability. The Private Discussions are better suited for more intensive interactions involving peer tutoring or in-depth discussions. Because there is only one helper in such situations, I-Help requires a method of selecting an appropriate helper for an individual. We describe the user modelling that takes place in each part of I-Help, in particular to effect this matchmaking for Private Discussions. This modelling takes advantage of a distributed multi-agent architecture, allowing currently relevant user model fragments in various locations to be integrated and computed at the time they are required.

Revising deductive knowledge and stereotypical knowledge in a student model

A user/student model must be revised when new information about the user/student is obtained. But a sophisticated user/student model is a complex structure that contains different types of knowledge. Different techniques may be needed for revising different types of knowledge. This paper presents a student model maintenance system (SMMS) which deals with revision of two important types of knowledge in student models: deductive knowledge and stereotypical knowledge. In the SMMS, deductive knowledge is represented by justified beliefs. Its revision is accomplished by a combination of techniques involving reason maintenance and formal diagnosis. Stereotypical knowledge is represented in the Default Package Network (DPN). The DPN is a knowledge partitioning hierarchy in which each node contains concepts in a sub-domain. Revision of stereotypical knowledge is realized by propagating new information through the DPN to change default packages (stereotypes) of the nodes in the DPN. A revision of deductive knowledge may trigger a revision of stereotypical knowledge, which results in a desirable student model in which the two types of knowledge exist harmoniously.

Inspecting and Visualizing Distributed Bayesian Student Models

Bayesian Belief Networks provide a principled, mathematically sound, and logically rational mechanism to represent student models. The belief net backbone structure proposed by Reye [14,15] offers a practical way to represent and update Bayesian student models describing both cognitive and social aspects of the learner. Considering students as active participants in the modelling process, this paper explores visualization and inspectability issues of Bayesian student modelling. This paper also presents ViSMod an integrated tool to visualize and inspect distributed Bayesian student models.

Granularity-Based Reasoning and Belief Revision in Student Models

In this chapter we discuss two important research topics surrounding student modelling: 1) how to represent knowledge about a student at various grain sizes and reason with this knowledge to enhance the capabilities of an intelligent tutoring system, and 2) how to maintain a consistent view of a student’s knowledge as the system-student interaction evolves. The ability to represent and reason about knowledge at various levels of detail is important for robust tutoring. A tutor can benefit from incorporating an explicit notion of granularity into its representation and can take advantage of granularity-based representations in reasoning about student behaviour. As the student’s understanding of concepts evolves and changes, the student model must track these changes. This leads to a difficult student model maintenance problem. Both of these topics are full of interesting subtleties and deep issues requiring years of research to be resolved (if they ever are), but a start has been made. In this chapter we characterize the main requirements for each topic, discuss some of our work that tackles these topics, and, finally, indicate important areas for future research.

Inspectable User Models for Just-In-Time Workplace Training

Workplace training is most effective when the training happens just in time as part of a worker’s regular job activities. We are developing a just-in-time training system called PHelpS (Peer Help System) which can select peer helpers with whom the worker can interact. User modelling is central in the PHelpS system. For each worker, a user model is kept containing several kinds of information about the worker, in particular a knowledge profile of how well they can carry out various specific tasks. These user models permit the system to select a knowledgeable, available, and appropriate set of helpers if a worker signals that he or she needs help in carrying out a particular task. Many interesting user modelling issues arise in this work, most importantly employing the same user model in multiple ways, making the user models inspectable by a variety of users, doing knowledge-based matching and retrieval, and maintaining the accuracy of the user model over time. There are several social issues that this research has also exposed.

Combining ITS and elearning technologies: opportunities and challenges

The development of Intelligent Tutoring Systems (ITS) and eLearning systems has been progressing largely independently over the past several years. Both types of systems have strengths and weaknesses – ITSs are typically domain specific and rely on concise knowledge modeling and learner modeling, while eLearning systems are deployable in a wide range of circumstances and focus on connecting learners both to content and to one another. This paper provides possibilities for convergence of these two areas, and describes two of our experiences in providing an ITS-style approach to eLearning systems.