Empirical Analysis and Refinement of Expert System Knowledge Bases
Based on encouraging results in well circumscribed applications, researchers often propose ambitious projects to build expert systems for highly complex tasks, requiring many thousands of reasoning rules. The paradigm for designing these systems runs mainly along the lines of current expert system conventions: Knowledge engineers construct knowledge bases based on information gleaned from experts and other sources of information. In the larger, more complex applications, we can expect to have many knowledge engineers constructing a single knowledge base from multiple sources of information, with the involvement of numerous experts.
Unable to display preview. Download preview PDF.
- 2.Weiss S, Kulikowski C: A Practical Guide to Designing Expert Systems. Totowa, NJ: Rowman & Allanheld, 1984.Google Scholar
- 4.Vanker A, Stoecker W: An expert diagnostic program for dermatology. Comput Biol Med 17:241, 1984.Google Scholar
- 5.Ginsberg A, Weiss S, Politakis P: SEEK2: a generalized approach to automatic knowledge base refinement. p. 367. In: Proceedings of the Ninth International Joint Conference on Artificial Intelligence, Los Angeles, 1985.Google Scholar
- 6.Ginsberg A: Refinement of Expert System Knowledge Bases: A Metalinguistic Framework for Heuristic Analysis. PhD thesis, Department of Computer Science, Rutgers University, 1986.Google Scholar
- 7.Smith R, Winston H, Mitchell T, Buchanan B: Representation and use of explicit justification for knowledge base refinement. p. 673. In: Proceedings of the Ninth International Joint Conference on Artificial Intelligence, Los Angeles, 1985.Google Scholar