Adaptive learning expert systems

Abstract

The purpose of this research is to improve the performance of an expert system through the use of a neural network, thus allowing the expert system to learn from experience. Even though the knowledge representation schemes used by expert systems allow them to succeed and proliferate, these schemes cause them to be brittle. Human experts usually use more knowledge to reason than expert systems do and often use experience in quantitative reasoning whereas expert systems cannot. Our study shows that a neural network can learn from an expert system's experience and guide the expert system when the expert system does not have enough knowledge to reason.