Accession number:20090211849736
Title:Fuzzy c-mean algorithm based on complete mahalanobis distances and separable criterion
Authors:Liu, Hsiang-Chuan (1); Wu, Der-Bang (2); Yih, Jeng-Ming (2);
Liu, Shin-Wu (3)
Author affiliation:(1) Asia University, Taiwan; (2) Taichung University, Taiwan; (3) Rutgers University, United States
Corresponding author:Liu, H.-C.
Source title:Proceedings - 5th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2008
Abbreviated source title:Proc. - Int. Conf. Fuzzy Syst. Knowl. Discov., FSKD
Volume:1
Monograph title:Proceedings - 5th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2008
Issue date:2008
Publication year:2008 Pages:87-91
Article number:4665945 Language:English
ISBN-13:9780769533056
Document type:Conference article (CA)
Conference name:5th International Conference on Fuzzy Systems and Knowledge Discovery, FSKD 2008
Conference date:October 18, 2008 - October 20, 2008 Conference location:Jinan, Shandong, China
Conference code:74641
Publisher:Inst. of Elec. and Elec. Eng. Computer Society, 445 Hoes Lane - P.O.Box 1331, Piscataway, NJ 08855-1331, United States Abstract:The well known fuzzy partition clustering algorithms are most based on Euclidean distance function, which can only be used to detect spherical structural clusters. GK clustering algorithm and GG clustering algorithm, were developed to detect non-spherical structural clusters, but both of them fail to consider the relationships between cluster centers in the objective function, needing additional prior information. In our previous studies, we developed two
improved algorithms, FCM-M and FCM-CM based on unsupervised Mahalanobis distance without any additional prior information. And FCM-CM is better than FCM-M, since the former has the more
information about the overall covariance matrix than the later. In this paper, an improved new unsupervised algorithm, "fuzzy c-mean based on complete Mahalanobis distance and separable criterion without any prior information (FCM-CMS)", is proposed. In our new algorithm, not only the local and overall covariance matrices of all clusters but also an additional separable criterion were considered.
It can get more information and higher accuracy by considering the additional separable criterion than FCM-CMx. A real data set was applied to prove that the performance of the FCM-CMS algorithm is better than those of the traditional FCM algorithm and our previous FCM-M. © 2008 IEEE.
Number of references:9
Main heading:Clustering algorithms
Controlled terms:Covariance matrix - Diesel engines - Fuzzy clustering - Fuzzy logic - Fuzzy rules - Fuzzy systems - Knowledge based systems - Solenoids - Spheres
Uncontrolled terms:Cluster centers - Euclidean distances - Fcm algorithms - Fuzzy partitions - Improved algorithms - Mahalanobis distances - New algorithms - Objective functions - Prior informations - Real data sets - Structural clusters - Unsupervised algorithms
Classification code:961 Systems Science - 723.4.1 Expert Systems - 731.1 Control Systems - 903.1 Information Sources and Analysis - 921 Mathematics - 921.4 Combinatorial Mathematics, Includes Graph Theory, Set Theory - 922 Statistical Methods - 723.4 Artificial Intelligence - 721.1 Computer Theory, Includes Formal Logic,
Automata Theory, Switching Theory, Programming Theory - 721 Computer Circuits and Logic Elements - 704.1 Electric Components - 631 Fluid Flow - 612.2 Diesel Engines - 723 Computer Software, Data Handling and Applications
DOI:10.1109/FSKD.2008.34 Database:Compendex
Compilation and indexing terms, Copyright 2009 Elsevier Inc.
