Using expert tutor knowledge to design a Self-Improving intelligent tutoring system

  • Eric Gutstein
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 608)


What knowledge does an intelligent tutoring system need to learn from its experiences with students and improve its tutoring, and what are the necessary learning mechanisms? I address these in discussing (1) SIFT, a Self-Improving Fractions Tutor and (2) my study of an expert tutor on whose knowledge SIFT is based. SIFT is a production system with a tutor and a learning module which learns from its interactions with the students who use it. The students who use it are models of problem solvers, and the input transcripts are simulations of interactions. After augmenting its knowledge, SIFT evaluates its modifications and updates its rule probabilities using the Dempster-Shafer theory of evidence—a domain-independent modification. Thus its choice of which rule to fire is determined by the empirical effects of the changes it makes to its tutorial knowledge.


Intelligent Tutoring System Student Modeling Rule Probability Human Tutor Compound Problem 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    J. R. Anderson, C. F. Boyle, G. Yost: The geometry tutor. Proceedings of the Ninth IJCAI Conference, Los Angeles, CA, 1985Google Scholar
  2. 2.
    Y. Anzai, H. A. Simon: The theory of learning by doing. Psy. Review, 86 (2), 1979Google Scholar
  3. 3.
    J. S. Brown, R. R. Burton: Diagnostic models for procedural bugs in basic mathematical skills. Cognitive Science 2, 155–192, 1979Google Scholar
  4. 4.
    R. R. Burton: Diagnosing bugs in a simple procedural skill. In D. Sleeman & J. S. Brown (Eds.), Intelligent tutoring systems, London: Academic Press, 1982Google Scholar
  5. 5.
    J. G. Carbonell, Y. Gil: Learning by experimentation. Proc. of the 4th International Machine Learning Workshop, University of California, Irvine, 1987Google Scholar
  6. 6.
    T. P. Carpenter, J. M. Moser: The acquisition of addition and subtraction concepts. In R. Lesh & M. Landau (eds.), Acquisition of mathematics concepts and processes, New York: Academic Press, 1983Google Scholar
  7. 7.
    P. Dillenbourg: Self-improving tutoring systems. International Journal of Educational Research, 12, 851–862, 1988Google Scholar
  8. 8.
    P. Dillenbourg, P. Goodyear: Towards reflective tutoring systems: self-representation and self-improvement. In D. Bierman, J. Breuker, & J. Sandberg (Eds.), Proc. of the 4th International Conference on AI and Education, Amsterdam: IOS, 1989Google Scholar
  9. 9.
    E. Fennema, T. P. Carpenter, P. Peterson: Learning mathematics with understanding. In J. Brophy (Ed.), Advances in research on teaching, Vol. 1, Greenwich, CN: JAI Press, 1989Google Scholar
  10. 10.
    J. Gordon, E. H. Shortliffe: The Dempster-Shafer theory of evidence. In B. Buchanan & E. Shortliffe (Eds.), Rule-based expert systems: The MYCIN experiments, Reading, MA: Addison-Wesley, 1984Google Scholar
  11. 11.
    E. Gutstein: Learning from students to improve an intelligent tutoring system. Machine Learning Research Group Working Paper 92–1, University of Wisconsin-Madison, 1992Google Scholar
  12. 12.
    E. Gutstein: SIFT: A self-improving fractions tutor. PhD thesis, in preparation, University of Wisconsin-Madison, 1992Google Scholar
  13. 13.
    R. Kimble: A self-improving tutor for symbolic integration. In D. Sleeman & J. S. Brown (Eds.), Intelligent tutoring systems, London: Academic Press, 1982Google Scholar
  14. 14.
    P. Langley: Data-driven discovery of physical laws. Cognitive Science, 5(1), 1981Google Scholar
  15. 15.
    D. B. Lenat: The ubiquity of discovery. Artificial Intelligence, 9, 257–285, 1977Google Scholar
  16. 16.
    M. W. Lewis, D. McArthur, C. Stasz, M. Zmuidzinas: Discovery-based tutoring in mathematics. Proc. of the AAAI Symposium on Knowledge-Based Environments for Learning and Teaching, Stanford, CA, 1990Google Scholar
  17. 17.
    N. K. Mack: Learning fractions with understanding. PhD thesis, University of Wisconsin-Madison, 1987Google Scholar
  18. 18.
    N. K. Mack: Learning fractions with understanding: Building on informal knowledge. Journal of Research in Mathematics Education, 21 (1), 1990Google Scholar
  19. 19.
    D. McArthur, C. Stasz, M. Zmuidzinas: Tutoring techniques in algebra. Cognition and Instruction, 7(3), 1990Google Scholar
  20. 20.
    T. O'Shea: A self-improving quadratics tutor. In D. Sleeman & J. S. Brown (Eds.), Intelligent Tutoring Systems, London: Academic Press, 1982Google Scholar
  21. 21.
    R. T. Putnam: Structuring and adjusting content for students: A study of live and simulated tutoring of addition. American Educational Research Journal, 24(1), 1987Google Scholar
  22. 22.
    J. R. Quinlan: Induction of decision trees, Machine Learning, 1 (1), 1986Google Scholar
  23. 23.
    S. A. Rajamoney, G. A. DeJong: The classification, detection, and handling of imperfect theory problems. Proc. of the 10th IJCAI Conf., Milan, Italy, 1987Google Scholar
  24. 24.
    W. Shen: Learning from the environment based on percepts and actions. PhD thesis, Carnegie-Mellon University, 1989Google Scholar
  25. 25.
    J. Self: Bypassing the intractable problem of student modeling. Proc.of the ITS Conference, Montreal, PQ, 1988Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1992

Authors and Affiliations

  • Eric Gutstein
    • 1
  1. 1.Computer Science DepartmentUniversity of Wisconsin-MadisonMadison

Personalised recommendations