Bayesian Analysis in a Knowledge-Intensive CBR System

Abstract

This study presents a case-based reasoning system that makes use of general domain knowledge - referred to as a knowledge-intensive CBR system. The system applies a Bayesian analysis aimed at increasing the accuracy of the similarity assessment. The idea is to employ the Bayesian posterior distribution for each case symptom to modify the case descriptions and the dependencies in the model. The system is evaluated against a simplified version of a corresponding system named TrollCreek and the results of running two examples from two different application domains, i.e., a "food domain" and a "drilling process domain" are compared with a human expert prediction. The obtained results reveal the capability of Bayesian analysis to increase the accuracy of the similarity assessment.