Abstract: Self-adaptive strategy with the traditional fuzzy C-means clustering algorithm forms a new fuzzy clustering algorithm.Without prejudice to the speed of convergence，it can resolve the problems of local optimal and sensitivity to initial values.With the two data sets in the database of UCI machine learning for the study，the experimental results indicate that it does not lose the precision to the adaptive immune clustering algorithm.The number of clusters is accurate and its faster convergence is more important in the nowadays of high-speed network data changing.