FAISC: A Fuzzy Artificial Immune System Clustering Algorithm

Clustering is an unsupervised knowledge discovery process that groups a set of data such that the intra-cluster similarity is maximized and the inter-cluster similarity is minimized. Existing clustering algorithms, such as k-means and PAM, are designed to find clusters that fit static models. In this paper, we describe a clustering algorithm called fuzzy artificial immune system clustering (FAISC), which is based on artificial immune network and fuzzy system. Experiments on benchmark datasets show very good qualitative results obtained better than k-means which is one of the widest used clustering algorithm.