A Fuzzy Cluster Algorithm Based on Mutative Scale Chaos Optimization

Authors:
Chaoshun Li;Jianzhong Zhou;Qingqing Li;Xiuqiao Xiang
Affiliations:
College of Hydroelectric Digitization Engineering, Huazhong University of Science and Technology, Wuhan, China 430074;College of Hydroelectric Digitization Engineering, Huazhong University of Science and Technology, Wuhan, China 430074;College of Hydroelectric Digitization Engineering, Huazhong University of Science and Technology, Wuhan, China 430074;College of Hydroelectric Digitization Engineering, Huazhong University of Science and Technology, Wuhan, China 430074
Venue:
ISNN '08 Proceedings of the 5th international symposium on Neural Networks: Advances in Neural Networks, Part II
Year:
2008

Citing 4
Cited 2

A Validity Measure for Fuzzy Clustering

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Kernel-Based Fuzzy Clustering Algorithm

ICICIC '06 Proceedings of the First International Conference on Innovative Computing, Information and Control - Volume 1
Image segmentation by clustering of spatial patterns

Pattern Recognition Letters
A study of some fuzzy cluster validity indices, genetic clustering and application to pixel classification

Fuzzy Sets and Systems

Genetic algorithm-tuned entropy-based fuzzy C-means algorithm for obtaining distinct and compact clusters

Fuzzy Optimization and Decision Making
A novel chaotic particle swarm optimization based fuzzy clustering algorithm

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

The traditional Fuzzy C-Means (FCM) algorithm has some disadvantages in optimization method, which makes the algorithm liable to fall into local optimum, thus failing to get the optimal clustering results. According to the defect of FCM algorithm, a new Fuzzy Clustering algorithm based on Chaos Optimization (FCCO) is proposed in this paper, which combines mutative scale chaos optimization strategy and gradient method together. Moreover, a fuzzy cluster validity index (PBMF) is introduced to make the FCCO algorithm capable of clustering automatically. Three other fuzzy cluster validity indices, namely XB, PC and PE, are utilized to compare the performances of FCCO, FCM and another algorithm, when applied to artificial and real data sets classification. Experiment results show FCCO algorithm is more likely to obtain the global optimum and achieve better performances on validity indices than other algorithms.