凝聚中心犹豫度恒定的模糊层次聚类算法

Fuzzy Hierarchical Clustering Algorithm with Constant Hesitation of Agglomeration Center

传统的模糊方法已无法解决数据本身不确定性的问题,犹豫模糊集方法却行之有效.原有的犹豫模糊层次聚类算法没有考虑犹豫模糊集对权值的影响,缺乏合理的权重计算方法,并且算法的时间复杂度和空间复杂度都为指数级.为了更有效地解决聚类分析问题,本文提出一种凝聚中心犹豫度恒定的模糊层次聚类算法(FHCA),首先设计了一种基于数据集本身信息的权重公式,可以得到更加合理的权重分配.此外还提出了新的簇中心的计算公式,不仅使聚类过程中,簇中心的犹豫度具有不变性,还将原有算法的时间复杂度以及空间复杂度从指数级降至线性级,并且聚类的质量不劣于原有的聚类算法.

Traditional fuzzy methods are unable to solve the uncertainty of the data,but the hesitant fuzzy set methods can w ork efficiently.Existing hesitant fuzzy hierarchical clustering algorithms lack a reasonable weight calculation method,and the time complexity and space complexity of the algorithms are both exponential.In order to address these issues,this paper proposes a fuzzy hierarchical clustering algorithm with constant hesitation of agglomeration center(FHCA).Firstly,a weight formula based on the information of the data set is designed to obtain a more reasonable w eight distribution.In addition,a new formula for calculating the cluster center is proposed,which not only makes the cluster center’s hesitancy invariant during the clustering process,but also reduces the time complexity and space complexity of the original algorithm from exponential to linear,and the quality of clustering is not inferior to the original clustering algorithm.