直觉模糊C-均值聚类算法研究

Research on intuitionistic fuzzy C-means clustering algorithm

鉴于直觉模糊集理论作为模糊理论的推广已得到广泛的应用,研究了将模糊C-均值聚类推广为直觉模糊C-均值聚类(IFCM)的途径和方法,分析了现有的几种IFCM算法,并提出了一种基于直觉模糊集的模糊C-均值聚类算法.该算法首先定义了直觉模糊集之间的距离;然后构造了聚类的目标函数;最后给出了聚类算法步骤.将算法用于目标识别,实验结果表明了算法的有效性.

The intuitionistic fuzzy set（IFS） theory, which is generalized fuzzy set theory, is used in a wide rang of applications. Therefor the approaches of making the fuzzy C-means（FCM） clustering algorithm expand into an intuitionistic fuzzy C-means（IFCM） clustering algorithm are investigated, and several existing IFCM algorithms are analyzed. A fuzzy C- means clustering algorithm based on IFSs is proposed, which firstly defines the distance of IFSs, then constructs the objective function of clustering, finally gives the clustering procedure. The proposed algorithm is applied to target recognition, and the experiment results show the feasibility of the algorithm.