一种基于模糊超图的冲突分析模型

A conflict analysis model based on fuzzy hypergraph

针对多方冲突分析采用加权求和方式得到的冲突解决方案存在局限性的问题,提出对所有属性集整体考虑的一种基于模糊超图的冲突分析模型。首先引入模糊横贯和最小模糊横贯,将对象子集以不同规则划分,以此定义一致性测量函数,得到冲突集、中立集和同盟集。其次为了简化最大同盟的计算,建立对象子集树,提出最大同盟剪枝算法。最后通过实例和对比实验对模型和算法进行验证,结果表明,该模型可以准确描述多对象在整体属性下的关系,能够分析出每个对象子集的状态,并缓解了最大同盟剪枝算法的冗余问题,提高了算法性能。

For the limitation that traditional multi-party conflict analysis often uses weighted summation to get the conflict solution,a fuzzy hypergraph-based conflict analysis model is proposed to consider all attribute sets as a whole.Firstly,fuzzy transversal and minimum fuzzy transversal are introduced to divide the object subsets with different rules,so as to define the consistency measurement function to obtain the conflict set,the neutral set and the alliance set.Secondly,in order to simplify the computation of maximal coalitions,the object subset tree is established and the maximal coalitions-based object subset tree pruning algorithm is proposed.Finally,the model and algorithm are validated through examples and comparative experiments.The results show that the model can accurately describe the relationship of multiple objects under the overall attributes,and can analyze the state of each object subset,as well as alleviate the redundancy problem of the maximal coalitions-based object subset tree pruning algorithm,which improves its performance.