A neural network based expert system model

Abstract

The architecture of an expert system model using artificial neural networks is proposed. The proposed model effectively supports the necessary components of an expert system such as user interface facility knowledge base, inference engine, and explanation system. The expert system model (ESM) consists of several orders of simple neural networks, each realizing a simple task. These simple neural networks are organized vertically, thereby achieving a second level of parallelism. A novel way to handle both forward and backward chaining reasoning mechanisms is presented. A secondary network model monitors the reasoning patterns of the primary model.<>