Fuzzy C-means algorithm based on pso and mahalanobis distance

(1)

Accession number:20100312651188

Title: Fuzzy C-means algorithm based on pso and mahalanobis distance

Authors:Liu, Hsiang-Chuan (1); Yih, Jeng-Ming (2); Lin, Wen-Chih (3);

Liu, Tung-Sheng (4)

Author affiliation:(1) Department of Bioinformatics, Asia University, 500 Lioufeng Rd., Wufeng, Taichung, 41354, Taiwan; (2) Institute of Educational Measuremnent and Statistics, Departmnent of

Mathemnatics Education, National Taichung University, No. 140, Ming-sheng Road, Taichung, 40306, Taiwan; (3) Departmnent of Computer Science and Information Engineering, Asia University, 500 Lioufeng Rd., Wufeng, Taichung, 41354, Taiwan; (4) Science Center of Military Police Comnmnand, Taiwan

Corresponding author:Liu, H.-C.

([email protected])

Source title: International Journal of Innovative Computing, Information and Control

Abbreviated source title:Int. J. Innov. Comput. Inf. Control Volume:5

Issue:12

Issue date:December 2009 Publication year:2009 Pages:5033-5040 Language:English ISSN:13494198

Document type:Conference article (CA)

Publisher:IJICIC Editorial Office, 9-1-1 Toroku, Kamamoto, 862 8652, Japan

Abstract:Some of the wall-known fuzzy clustering algorithms are based on Euclidean distance function, which, can only be used to detect spherical structural clusters. Gustafson-Kessel (GK) clustering algorithm and Gath-Geva (GG) clustering algorithm were developed to detect non-spherical structural clusters. Both of GG and GK

algorithms suffer from the singularity problem of covariance matrix and the effect of initial status. In this paper, a new Fuzzy C-Means algorithm based on Particle Swarm Optimization and Mahalanobis is distance without prior information (PSO-FCM-M) is proposed, to

(2)

improve those limitations of GG and GK algorithms. And we point out that the PSO-FCM algorithm is a special case of PSO-FCM-M algorithm. The experimental results of two real data sets show that the performance of our proposed PSO-FCM-M algorithm is better than those of the FCM, GG, GK algorithms. © 2009 ISSN.

Number of references:21

Main heading:Clustering algorithms

Controlled terms: Copying - Covariance matrix - Fuzzy clustering - Fuzzy rules - Fuzzy systems - Particle swarm optimization (PSO) Uncontrolled terms: Euclidean distance - FCM algorithm - Fuzzy C- means algorithms - G-K algorithm - Gustafson-Kessel - M-algorithms - Mahalanobis - Mahalanobis distances - Non-Spherical - Prior

information - Real data sets - Singularity problems

Classification code:922 Statistical Methods - 921.5 Optimization Techniques - 921.4 Combinatorial Mathematics, Includes Graph Theory, Set Theory - 921 Mathematics - 903.2 Information

Dissemination - 961 Systems Science - 903.1 Information Sources and Analysis - 731.1 Control Systems - 723.4 Artificial Intelligence - 723 Computer Software, Data Handling and Applications - 721 Computer Circuits and Logic Elements - 745.2 Reproduction, Copying

Database:Compendex