Research on a personalized expert system explanation method based on fuzzy user model

Explanation is an important component of ES (expert systems). It is used to clarify the reasoning process of ES to users and then can improve the understand ability and acceptability of ES results. The users' level of knowledge is different, so a reasonable explanation should be adapted to users of different level. How to describe the difference and then generate a personalized explanation is the focus of this paper. We propose FUM (fuzzy user model) to be an effective method to resolve this problem. It can help the system to judge the users level of knowledge and to produce a suitable explanation. In this paper, we first give the definition of FUM, and then provide three algorithms such as FUM initialization algorithm (AIARS: Adaptive Inter-view Algorithm Based on Random Sequence), FUM updating algorithm (AUADH: Adaptive Update Algorithm Based on Dialogue History) and FUM-Knowledge matching algorithm (FMA). All of them are implemented in the Expert System for Earthquake Prediction (ESEP 3.0).