Fuzzy Clustering Analysis Based On Feature Weights With Cluster Center Separation

Fuzzy clustering analysis is a method of using fuzzy math solving fuzzy clustering problems. It is one of important researching fields and applying technologies. This article researches fuzzy clustering problems. Main contents are:1. Accounting the impact on clustering results of the importance of different characteristics, on the basis of fuzzy clustering model based on cluster centers separation, this article proposes Fuzzy Clustering Model based on Feature Weights with Cluster Center Separation, and provides the solutions to make objective function achieve the minimum.2. WFCM_CCS is easy to program. This article writes procedures with MATLAB to simulate, and the results show that importance of different characteristics indeed reflects clustering effects in WFCM_CCS model. Clustering accuracy is greatly affected by choosing different weights. If proper weights are selected, the clustering effects will be very good. This algorithm has better cluster performance and precedes FCM_CCS algorithm.In programing, this paper adopts special processing techniques. Changing trend of clustering results with the weights can be seen."Breaking layer" also appears in the clustering process.