Knowledge Acquisition in the Development of a Large Expert System

Abstract

This article discusses several effective techniques for expert system knowledge acquisition based on the techniques that were successfilly used to develop the Cen- tral Office Maintenance Printout Analysis and Suggestion System (COMPASS) Knowledge acquisition is not a science, and expert system developers and experts must tailor their methodologies to fit their situation and the people involved. Devel- opers of future expert systems should find a description of proven knowledge-acquisi- tion techniques and an account of the experience of the COMPASS project in applying these techniques to be useful in developing their own knowledge-acquisi- tion procedures nowledge acquisition is the process by which expert sys- tem developers find the knowl- edge that domain experts use to perform the task of interest. This knowledge is then implemented to form an expert system. The essential part of an expert system is its knowl- edge, and therefore, knowledge acqui- sition is probably the most important task in the development of an expert system. In this article, several effective techniques for expert system knowl- edge acquisition are discussed based on the techniques that were success- fully used at GTE Laboratories to develop the COMPASS expert system. Knowledge acquisition for expert sys- tem development is still a new field and not (yet?) a science. Therefore, expert system developers and the experts they work with must tailor their knowledge-acquisition method- ologies to fit their own particular situ- ation and the people involved. As expert system developers define their own knowledge-acquisition proce- dures, they should find a description of proven knowledge-acquisition tech- niques and an account of the experi- ence of the COMPASS developers in applying these techniques to be use- ful. The next section of this article is a discussion of the COMPASS project. The major portion of the article fol- lows, with over SO points on knowl- edge acquisition that were found to be important during the work on COM- PASS. Initial points cover the knowl- edge-acquisition considerations in selecting an expert and an appropriate domain for the expert system. The remaining points highlight techniques for getting started in knowledge acqui- sition, documenting the knowledge, and finally, actually acquiring and recording the knowledge. Each point is followed by a general discussion and then by a description of how the point specifically applied to the COMPASS project.