FCM Algorithm Based on the Optimization Parameters of Objective Function Point

Fuzzy c-means clustering (FCM) algorithm is an important and basic tool of classification and analysis for no supervision data, which has been used extensively in pattern recognition, data analysis, image processing and fuzzy modeling. Although the FCM algorithm is an unsupervised machine self-learning algorithm, however, there are two parameters must be given appropriate assignment before conducting cluster analysis, that is, the fuzzy weighted index m and the number of clustering c, otherwise, the analysis of FCM algorithm will be affected, and the reasonable explanation of clustering analysis will also be influenced directly. In order to give the two important parameters of clustering analysis reasonable assignment, this paper uses the objective function point method to optimize the parameter m, and thus proceed by solving the optimal m to determine the number of optimum clusters c.