Title :
Memetic Algorithm based fuzzy clustering
Author :
Do, Anh-Duc ; Cho, Siu-Yeung
Author_Institution :
Nanyang Technol. Univ., Singapore
Abstract :
This paper presents a Memetic Algorithm (MA) based Fuzzy C-Means (FCM) clustering algorithm. Traditional FCM algorithm suffers from the problem of local optimal, whereas the proposed MA-based FCM algorithm is able to overcome this problem and produce good performance in various ways. Experimental results showed that the proposed clustering algorithm outperforms traditional fuzzy clustering algorithms significantly on a wide variety of datasets with overlapping class boundaries and spread data distributions.
Keywords :
fuzzy set theory; pattern clustering; fuzzy c-means clustering algorithm; memetic algorithm; overlapping class boundaries; spread data distributions; Clustering algorithms; Evolutionary computation; Genetic algorithms; Iterative algorithms; Minimization methods; Partitioning algorithms; Robustness; Search methods; Stochastic processes;
Conference_Titel :
Evolutionary Computation, 2007. CEC 2007. IEEE Congress on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-1339-3
Electronic_ISBN :
978-1-4244-1340-9
DOI :
10.1109/CEC.2007.4424771
