Inducing Rules for Expert System Development: An Example Using Default and Bankruptcy Data

William F. Messier, James V. Hansen

Management Science · 1988 · 292 citations · 11 references

Concepts

TL;DR

Knowledge acquisition challenges are increasingly important as interest in knowledge‑based consulting systems grows. The study evaluates an inductive algorithm’s ability to uncover predictive knowledge structures in financial data. Performance was measured by comparing the algorithm’s outputs to discriminant analysis, individual, and group judgments. The algorithm identified overlapping predictive attributes and consistently outperformed discriminant analysis, individual, and group judgments across all test cases.

Abstract

With rapidly growing interest in the development of knowledge-based computer consulting systems for various problem domains, the difficulties associated with knowledge acquisition have special importance. This paper reports on the results of experiments designed to assess the effectiveness of an inductive algorithm in discovering predictive knowledge structures in financial data. The quality of the results are evaluated by comparing them to results generated by discriminant analysis, individual judgments, and group judgments. A partial intersection of predictive attributes occurs. More importantly, for all cases tested, the inductively produced knowledge structures perform better than the competing models.

References

11