一种基于新的直觉模糊相似度的聚类方法  

Clustering Method Based on New Intuitionistic Fuzzy Similarity Degree

作　　者：袁泽华 潘小东[1] 

机构地区：[1]西南交通大学数学学院,四川成都610031

出　　处：《洛阳理工学院学报（自然科学版）》2017年第4期85-89,共5页Journal of Luoyang Institute of Science and Technology：Natural Science Edition

摘　　要：针对现有直觉模糊聚类方法中大多数未考虑指标权重,且计算结果为实数的问题,提出了一种基于新直觉模糊相似度的聚类方法。运用直觉模糊熵得到指标权重,构造了一种考虑指标权重的直觉模糊相似度公式,得到直觉模糊相似矩阵,构造了满意度阈值。决策者根据自身满意度偏好选择合适满意度阈值,将直觉模糊相似矩阵转化为实数矩阵,然后利用平方自合成方法得到模糊等价矩阵,选择适当的置信值进行聚类,最后通过实例验证所提方法的有效性和可行性。Aiming at the problem, which most of the existing intuitionistic fuzzy clustering methods do not take the weight of the index into account and the calculation result is real number, a clustering method based on the new intuitionistic fuzzy similarity degree is pro- posed. By using the intuitionistic fuzzy entropy, the index weight is obtained, and an intuitionistic fuzzy similarity degree formula for attribute weight is constructed as well as an approach is developed to construct an intuitionistic fuzzy similarity matrix. A satisfaction threshold is designed and the decision maker selects the appropriate satisfaction threshold according to their own preference, and trans- forms the intuitionistic fuzzy similarity matrix into real matrix, then, the fuzzy equivalence matrix is obtained by using the square self synthesis method, selects the appropriate confidence value to cluster. Finally, an example shows the validity and feasibility of this method.

关 键 词：直觉模糊相似度 聚类方法 直觉模糊熵 满意度阈值 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]