Skip to main content
SpringerLink
Log in

Using statistics to make expert systems “user-acquainted”

  • Published:
Annals of Mathematics and Artificial Intelligence Aims and scope Submit manuscript

Abstract

The user-acquainted feature collects user-specific data regarding the types of advice that have been sought over time and uses this historical information to update the probabilities to affect the firing of its rules and the ordering of the recommendations of the expert system. These updates are done on auser-specific basis so that the expert can more closely emulate a true expert by providing more informed advice. One example of a place where evidence suggests that such a feature would be useful is in the area of debugging of computer programs, especially in support of novice programmers who tend, as individuals, to commit similar classes of errors over time, but who, as a group, commit very different types of errors. We conjecture that the user-acquainted feature, which can keep track of the tendencies of the users and take them into account in the evaluation of diagnostics, will be more effectiveand efficient in determining the fault. In this paper, we discuss the statistical analyses necessary to implement this feature in an expert system for debugging errors in SAS.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. An AI productivity roundtable,3rd Artificial Intelligence Satellite Symp., sponsored by Texas Instruments (April 8, 1987).

  2. J.R. Anderson and B.J. Reiser, The LISP tutor, Byte 10(4) (April, 1985).

  3. M.T., Harandi, A knowledge-based programming support tool,Proc. Conf. on Trends and Applications, Automating Intelligent Behaviour, IEEE (1983) pp. 233–239.

  4. J.R. Hartley and D.H. Sleeman, Towards intelligent teaching systems, Int. J. Man-Machine Studies (1973).

  5. W.L. Johnson and E. Soloway, PROUST, Byte 10(4) (April, 1985).

  6. F. Pipitone, The FIS electronics troubleshooting system, Computer: IEEE Computer Society (July, 1986) 68–76.

  7. W.B. Rausch-Hindin,Artificial Intelligence In Business, Science and Industry, vol. 2 (Prentice-Hall, Englewood Cliffs, NJ, 1985).

    Google Scholar 

  8. C. Rich and H.E. Shrobe, Design of a programmer's apprentice, in:Artificial Intelligence: An MIT Perspective, eds. P.H.W. Henry and R.H. Brown (The MIT Press, Cambridge, MA, 1979).

    Google Scholar 

  9. M. Schindler, Artificial intelligence begins to pay off with expert systems for engineering, Electronic Design (August, 1984) 106–146.

  10. R.L. Sedlmeyer, W.B. Thompson and P.E. Johnson, Diagnostic reasoning in software fault localization,IJCAI-83, Karlsruhe (1983) pp. 29–31.

  11. S.E. Dreyfus, Formal models vs human situational understanding: Inherent limitations on the modeling of business expertise, Office: Technology and People 1 (August, 1982) 133–165.

    Google Scholar 

  12. L.F. Pau, Survey of expert systems for fault detection, test generation and maintenance, Expert Systems 3(2) (April, 1986) 100–111.

    Google Scholar 

  13. D.R. Miller, A continuity theorem and some counterexamples for the theory of maintained systems, Technical Report no. 66, Department of Statistics, University of Missouri — Columbia (December, 1975).

    Google Scholar 

  14. V.L. Sauter and L.A. Madeo, The need for user-acquainted expert systems for fault detection in the business environment, submitted to Information Systems Research (September, 1988).

  15. K.B. McKeithen, J.S. Reitman, H.H. Reuter and S.C. Hirtle, Knowledge organization and skill differences in computer programmers, Cognitive Psychol. 13 (1981) 307–325.

    Google Scholar 

  16. I. Vessey, Expertise in debugging computer programs: A process analysis, Int. J. Man-Machine Studies 23 (1985) 459–494.

    Google Scholar 

  17. M.J. Schervish, Comments on some axioms for combining expert judgments, Management Sci. 32(3) (March, 1986) 306–311.

    Google Scholar 

  18. A.B. Shahidul-Hussain, On the correctness of some sequential classification schemes in pattern recognition, IEEE Trans. Computers 21(3) (March, 1972) 318–320.

    Google Scholar 

  19. R.L. Winkler, Expert resolution, Management Sci. 32(3) (March, 1986) 298–303.

    Google Scholar 

  20. D. St. Clair, W.E. Bond and B.B. Flachsbart, Using output to evaluate and refine rules in rule-based expert systems,Proc. 3rd Annual Conf. on Artificial Intelligence for Space Applications, Huntsville, AL (November, 1987).

  21. J. Pearl, On evidential reasoning in a hierarchy of hypotheses, Artificial Intelligence 28(1) (1986) 9–15.

    Google Scholar 

  22. W.A. Gale, Knowledge-based acquisition for a statistical consulting system, Int. J. Man-Machine Studies 26(1) (January 1987) 55–64.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. School of Business Administration, University of Missouri-St. Louis, 63121-4499, St. Louis, MO, USA

    Vicki L. Sauter & Laurence A. Madeo

Authors
  1. Vicki L. Sauter
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Laurence A. Madeo
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

About this article

Cite this article

Sauter, V.L., Madeo, L.A. Using statistics to make expert systems “user-acquainted”. Ann Math Artif Intell 2, 309–326 (1990). https://doi.org/10.1007/BF01531014

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1007/BF01531014

Keywords

Search

Navigation