Study of Clustering Algorithm based on Fuzzy C-Means and Immunological Partheno Genetic

Clustering algorithm is very important for data mining. Fuzzy c-means clustering algorithm is one of the earliest goal-function clustering algorithms, which has achieved much attention. This paper analyzes the lack of fuzzy C-means (FCM) algorithm and genetic clustering algorithm. Propose a hybrid clustering algorithm based on immune single genetic and fuzzy C-means. This algorithm uses the fuzzy clustering of Immune Partheno-Genetic to guide the number and the choice of the clustering centers. And then utilize FCM to make the clustering (IPGA-FCM). This algorithm not only overcomes the local optimal problem of FCM, the choice of the initial value is inappropriate, but also overcomes the contradictions between the search speed and clustering accuracy of the general genetic clustering algorithm. Then it applies the novel clustering algorithm to Chinese document clustering. The clustering algorithm is superior to other ordinary clustering algorithm and the result can embody the wide diversity and large amount of Chinese document. Experiments show the algorithm is effective.