A Belief Revision Mechanism with Trust Reasoning based on Extended Reciprocal Logic for Multi-Agent Systems

Abstract

When an agent receives messages from other agents, it does belief revision. A belief revision includes, i) a trust reasoning process, i.e., it obtains new belief related to the messages, and deduces implicitly unknown beliefs from the obtained belief; ii) in the case of contradiction in the belief set, it resolves the contradiction. So, trust reasoning, and belief revision must be included in the decision-making process of an intelligent agent in multi-agent systems. Although a belief revision mechanism with trust reasoning is demanded to construct multi-agent systems, there is no such belief revision mechanism. We, therefore, present a belief revision mechanism with trust reasoning based on extended reciprocal logic for multi-agent systems.