Clustering analysis is an unsupervised mode of pattern recognition and is one of primary techniques in the filed of data mining.A clustering analysis method based on a modified shuffled frog leaping algorithm（MSFLA）is proposed.The new approach integrates the advantages of the MSFLA and the K-means algorithms

which introduces the K-means operation and utilizes the MSFLA for optimization

improves the locally searching capability and convergence speed of the clustering algorithm based on MSFLA.Simulations are performed to compare the performance of the clustering algorithm based on modified MSFLA and other clustering algorithm

which validates the effectiveness of the proposed algorithm.