Architecture of Knowledge Retrieval Based on Multi-Agent Systems

Abstract

Agent methodology represents a new way of analyzing, designing, and implementing complex software applications. Multi-agent systems (MAS) refer to all types of systems composed of multiple (semi-) autonomous components. To retrieval the knowledge in the fields of distributed fault detection or distributed fault tolerance, This paper have compared the characteristic of deliberative agent architectures with reactive agent architectures and hybrid agent architectures of MAS, proposed a hybrid intelligent agent model to solve the problem of fault learning. The MAS program language of Jade and Lygon are applied to accomplish the different kinds of knowledge learning.