Study Of Auto-Adaption Fuzzy C-Means Clustering Algorithm

With the development of information and databases technology, the volume in data collection and storage in the recent decades grows explosively. In order to explore useful knowledge and make use of data resource, we put forward data mining. At present, the clustering analysis derived from statistic has become an active area in the research on data mining. And fuzzy C-means clustering method based aim function becomes the focus of clustering analysis.In the dissertation, the technology of data mining is introduced in detail, including the concepts and functionalities of data mining, the sorts of data mining system, the main applications and trends in data mining. Subsequently, clustering analysis in data mining is disserted, involving the concepts of clustering used in data mining, data structure and data type of clustering, and the universal process of clustering in practical application. The main clustering algorithms and sorts in data mining are introduced in detail. Both C-means clustering algorithm and fuzzy C-means clustering algorithm are described respectively, and the process of fuzzy C-means clustering algorithm is researched and analyzed in detail. FCM clustering algorithm is improved by using phased clustering idea. The whole process of clustering is divided in two phases, and the number of clustering and initialization of clustering center are dealt self-adaptively, and better original parameters can be gained automatically. We adapt new amendatory method to quicken constringency speed and enhance clustering effect, for the calculation of membership matrix and clustering center, and a new algorithm of fuzzy clustering is put forward. At last, new algorithm advanced is analyzed by experiments, and gets better results.