A fuzzy clustering method using Genetic Algorithm and Fuzzy Subtractive Clustering

 Clustering is a challenging problem in data mining, requiring both accurate determination of the number of clusters and correct clustering of the data. Fuzzy C-means (FCM) is a popular algorithm using the partitioning approach to solve this problem. A drawback to FCM is that it requires the number of clusters to be set a priori. In this study, we combine FCM with Genetic Algorithm (GA), Subtractive Clustering (SC) and Bayesian cluster validation for a novel clustering method, fzGASCE that both determines the correct number of clusters and efficiently constructs these clusters from a given dataset. We show that fzGASCE outperforms existing methods using similar approaches on both artificial and real datasets. Availability: The test datasets and the method software are available online at http://ouray.ucdenver.edu/~tnle/fzgasce.