Analytical and Numerical Evaluation of the Suppressed Fuzzy C-Means Algorithm

被引：0

作者：

Szilagyi, Laszlo ^{[1
,2
]}

Szilagyi, Sandor M. ^{[1
]}

Benyo, Zoltan ^{[2
]}

机构：

[1] Sapientia Hungarian Sci Univ Transylvania, Fac Tech & Human Sci, Targu Mures, Romania

[2] Budapest Univ Technol & Econom, Dept Control Engn & Informat Technol, Budapest 1117, Hungary

来源：

MODELING DECISIONS FOR ARTIFICIAL INTELLIGENCE, PROCEEDINGS | 2008年 / 5285卷

关键词：

fuzzy c-means algorithm; suppressed fuzzy c-means algorithm; competitive clustering; alternating optimization;

D O I：

暂无

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Suppressed fuzzy c-means (s-FCM) clustering was introduced in [Fan, J. L., Zhen, W. Z., Xie, W. X.: Suppressed fuzzy c-means clustering algorithm. Patt. Recogn. Lett.. 24, 1607-1612 (2003)] with the intention of combining the higher speed of hard c-means (HCM) clustering with the better classification properties of fuzzy c-means (FCM) algorithm. They modified the FCM iteration to create a competition among clusters: lower degrees of memberships were diminished according to a previously set suppression rate, while the largest fuzzy membership grew by swallowing all the suppressed parts of the small Ones. Suppressing the FCM algorithm was found successful in the, terms of accuracy and working time, but the authors failed to answer a series of important questions. hi this paper we clarify the view upon the optimality and the competitive behavior of s-FCM via analytical computations and numerical analysis.

引用

页码：146 / +

页数：2

共 19 条

[1]

Anderson E., 1935, Bulletin of the American IRIS Society, V59, P2

[2]

[Anonymous], Pattern Recognition With Fuzzy Objective Function Algorithms

[3]

Arthur D., 2007, P 18 ANN ACMS S, P1027

[4]

Asuncion A., 2007, UCI MACHINE LEARNING

[5] Classification of Time Series Using Singular Values and Wavelet Subband Analysis with ANN and SVM Classifiers [J].

Benyo, Balazs ;

Somogyi, Peter ;

Palancz, Bela .

JOURNAL OF ADVANCED COMPUTATIONAL INTELLIGENCE AND INTELLIGENT INFORMATICS, 2006, 10 (04) :498-503

[6] EFFICIENT IMPLEMENTATION OF THE FUZZY C-MEANS CLUSTERING ALGORITHMS [J].

CANNON, RL ;

DAVE, JV ;

BEZDEK, JC .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1986, 8 (02) :248-255

[7] Suppressed fuzzy C-means clustering algorithm [J].

Fan, JL ;

Zhen, WZ ;

Xie, WX .

PATTERN RECOGNITION LETTERS, 2003, 24 (9-10) :1607-1612

[8] Generalized fuzzy c-means clustering strategies using Lp norm distances [J].

Hathaway, RJ ;

Bezdek, JC ;

Hu, YK .

IEEE TRANSACTIONS ON FUZZY SYSTEMS, 2000, 8 (05) :576-582

[9] OPTIMIZATION OF CLUSTERING CRITERIA BY REFORMULATION [J].

HATHAWAY, RJ ;

BEZDEK, JC .

IEEE TRANSACTIONS ON FUZZY SYSTEMS, 1995, 3 (02) :241-245

[10] Parameter selection for suppressed fuzzy c-means with an application to MRI segmentation [J].

Hung, WL ;

Yang, MS ;

Chen, DH .

PATTERN RECOGNITION LETTERS, 2006, 27 (05) :424-438

← 1 2 →