LAUSR

Abstract Conflict occurs naturally in all walks of life at any time. When agents face a conflict, they usually have multiple alternatives and hesitate to make a decision. This paper studies conflict analysis in the case of hesitant fuzzy setting, i.e. the attitude of an agent on an issue has multiple choices. We first propose a conflict analysis model based on hesitant fuzzy information systems. Subsequently, we present the concepts of conflict, neutrality and alliance sets on one issue and multiple issues respectively, and introduce a method to compute the three sets via Bayesian decision theory. Moreover, the degree of conflict, neutrality and alliance among agents with respect to an issue are defined respectively. According to the conflict and alliance degrees, the set of issues is divided into three parts: the main reason, secondary reason and irrelevant reason. Finally, two methods of computing coalitions are discussed. One is based on Bayesian decision theory. The other is to find a complete subgraph of the coalition graph.