EXPLANATION IN A DECISION-THEORETIC CONSULTING SYSTEM - AN AXIOMATIC APPROACH

Abstract

Many end users consider the ability to explain reasoning processes the most important feature of a knowledge-based advice system. Unlike the existing rule-based system, it is difficult to explain the generated advice in a decision-theoretic consulting system. Nevertheless, the existing rule-based systems often fail to treat explicitly the important uncertainties and tradeoffs that underlie symbolic reasoning. Thus, this research attempts to produce qualitative text explanation of treatment decisions based on explicit normative decision models. Especially, the sequential application of the five axioms for reasoning about preference makes it possible to explain the reason why a decision is generated. To implement the idea, a decision-theoretic explanation system (DETES) was designed and a prototype was developed on the microcomputer.