Dynamic Fuzzy Kernel Clustering Analysis of Enterprises Independent Innovation Capability Based on Artificial Immunity
暂无分享,去创建一个
In order to deal with the flaws of enterprise independent innovation capability clustering analysis method which can not computer complex distribution data, and sensitive to the initialization, and easy to fall into local optimum, and is artificially assigned the number of classes, a new method of dynamic fuzzy kernel clustering analysis based on artificial immunity is proposed in this paper. The above method uses for reference of the kernel function of the support vector machine theory, and defines the objective function of fuzzy clustering, and gives the equation of class centers and fuzzy dividing matrix, and optimize the process of computation with the artificial immunity idea which is based on clonal selection and affinity maturation in succession, and is combined with cluster validity judgment method that determine the optimal number of classes. The result of a case of enterprises independent innovation capability clustering analysis demonstrates the feasibility and effectiveness of the method.
[1] Guo Chen. Immune-based dynamic fuzzy clustering algorithm , 2007 .
[2] Ujjwal Maulik,et al. Performance Evaluation of Some Clustering Algorithms and Validity Indices , 2002, IEEE Trans. Pattern Anal. Mach. Intell..