DocumentCode :
1766637
Title :
A Novel Evolutionary Kernel Intuitionistic Fuzzy C -means Clustering Algorithm
Author :
Kuo-Ping Lin
Author_Institution :
Dept. of Inf. Manage., Lunghwa Univ. of Sci. & Technol., Taoyuan, Taiwan
Volume :
22
Issue :
5
fYear :
2014
fDate :
Oct. 2014
Firstpage :
1074
Lastpage :
1087
Abstract :
This study proposes a novel evolutionary kernel intuitionistic fuzzy c-means clustering algorithm (EKIFCM) that combines Atanassov´s intuitionistic fuzzy sets (IFSs) with kernel-based fuzzy c-means (KFCM), and genetic algorithms (GA) are optimally used simultaneously to select the parameters of the EKIFCM. The EKIFCM can obtain the advantages of intuitionistic fuzzy sets, kernel functions, and GA in actual clustering problems. Experiments on 2-D synthetic datasets and machine learning repository (http://archive.ics.uci.edu/beta/) datasets show that the proposed EKIFCM is more efficient than conventional algorithms such as the k-means (KM), FCM, Gustafson-Kessel (GK) clustering algorithm, Gath-Geva (GG) clustering algorithm, Chaira´s intuitionistic fuzzy c-means (IFCM), and kernel-based fuzzy c-means with Gaussian kernel functions [KFCM(G)] in standard measurement indexes.
Keywords :
fuzzy set theory; genetic algorithms; learning (artificial intelligence); pattern clustering; 2D synthetic datasets; Atanassov intuitionistic fuzzy sets; Chaira intuitionistic fuzzy c-means; EKIFCM algorithm; GA; GG clustering algorithm; GK clustering algorithm; Gath-Geva clustering algorithm; Gustafson-Kessel clustering algorithm; IFS; evolutionary kernel intuitionistic fuzzy c-means clustering algorithm; genetic algorithms; kernel functions; machine learning repository dataset; measurement index; Algorithm design and analysis; Clustering algorithms; Fuzzy sets; Image segmentation; Kernel; Linear programming; Prototypes; Evolutionary kernel intuitionistic fuzzy $c$-means (EKIFCM); fuzzy $c$-means (FCM); genetic algorithm (GA); intuitionistic fuzzy sets; kernel function;
fLanguage :
English
Journal_Title :
Fuzzy Systems, IEEE Transactions on
Publisher :
ieee
ISSN :
1063-6706
Type :
jour
DOI :
10.1109/TFUZZ.2013.2280141
Filename :
6587744
Link To Document :
بازگشت