Improved Fuzzy C Means Clustering Algorithm And Its Application

This paper first introduces Fuzzy C Means Clustering (FCM) algorithm, but because FCM algorithm does not consider that samples of various characteristics have impact of clustering results, on this basis,the paper also introduces weighted fuzzy C-means clustering (WFCM) algorithm. WFCM clustering algorithm is the same as that FCM clustering algorithm which only takes into account the distance of each sample to minimize within-class, and then they do not consider to enable the samples to maximize the distance between the classes. Therefore, for this problem, this paper further improves the FCM clustering algorithm based on WFCM clustering algorithm, the improved clustering algorithm considers maximizing the distance between the classes. Then the improved WFCM clustering algorithm-MWFCM clustering algorithm compares with FCM, WFCM clustering algorithm, The results show that in practice, MWFCM clustering algorithm has better effect than FCM, WFCM clustering algorithm. At last, it is applied to the assessment of scholarship.