Optimization of the clusters number of an improved fuzzy C-means clustering algorithm

Cluster analysis is an unsupervised most important research topics in the field of pattern recognition. Fuzzy clustering from the sample to the category of uncertainty description, it is possible to more objectively reflect the real world. Traditional fuzzy clustering algorithm can not achieve the optimal allocation of the number of clusters is calculated automatically. In this paper, by adopting the idea of hierarchical clustering, one can automatically and efficiently determine the optimal number of clusters of new adaptive fuzzy c-means clustering algorithm-A-FCM algorithm. Numerical experiments show that the other through a variety of validity function to determine the number of clusters of adaptive fuzzy clustering algorithm, the better the performance of the method.