A flexible explanation system for case-based reasoning

Abstract

Artificial Intelligence systems have reached a level of sophistication and accuracy which means that they can now be confidently used as decision aids in real world situations. However, although these systems are accurate, users are still reluctant to use them. Unlike domain experts such systems are unable to present convincing explanations in support of the recommendations that they make. This is a major shortcoming, as without this reassurance, people are naturally reluctant to accept unsupported recommendations.