Analysis Method And Application Research Of Structural Clustering

In this paper, structural clustering and clustering fusion are focused on as main objects based on the theory of granular computing, and the analysis method and application research of structural clustering based on fuzzy proximity relations is discussed. This paper is organized as follows:In chapter one, the developments of granular computing, clustering analysis, information fusion and species distribution prediction are briefly outlined both at home and abroad. Besides, the main work and innovations of this paper are given.In chapter two, the representation and improving algorithm of structural clustering are firstly given. Besides, the optimization principle on optimal clustering is determined. Finally, the method of clustering fusion is provided.In chapter three, based on the theory and methods of structural clustering and clustering fusion based on fuzzy proximity relations, the clustering structures of the climatic data in the distribution areas of the main forestry species in Northeast China from1981to1990are analyzed. Then the inner information of the climatic data is extracted. Furthermore, the conclusion on the growth period of forestry species is obtained.In chapter four, the climatic factor indicators are extracted based on the conclusion of chapter three. Then the random mathematical model of tree species on the climate change is built by using the rigorous theory and methods of statistical analysis and data processing, and the prediction algorithm is given. Based on these, the impacting analysis of the climate change on single species is studied.In chapter five, Gauss competitive exclusion principle is introduced, and the impacting analysis of the climate change on multiple species competition is studied based on chapter three and chapter four.In chapter six, the whole work is summarized and reported. Besides, the ideas and thoughts of further research are pointed out.