Feature Weighted and Improved Partition Fuzzy C-Means Cluster Algorithm

This paper proposes a new algorithm based on the FCM, which is called feature weighting and optimal fuzzy partition of fuzzy c-means algorithm (WGIFP - FCM). This algorithm combines the optimization partition method and the feature weight idea, obtains feature weight vector formula and updates it in the entire clustering process.Simulation experiments have proved that the proposed algorithm has gotten better clustering results than GIFP - FCM, WFCM and FCM while calculating the weighting vector scientifically, and so it will be available in practice.