Research On Cluster Algorithm Based On Chaos Immune Evolutionary Algorithm And Application

Recently data mining is becoming one of the most advanced and active research topics in the field of the information decision-making in the world. Clustering is the process of grouping together similar mufti-dimensional data vectors into a number of clusters or bins. Cluster analysis is an effective tool of data mining, is attracting wide attention. In the recent several decades, cluster analysis is made rapid development, and many new algorithms have been proposed for various applications. Cluster analysis have been applied to a wide range of problems, including pattern recognition, computer vision, fuzzy control, image segment, feature extraction, signal compression, etc.This paper is engaged in the hybrid algorithm of Artificial Immune System(AIS), Evolutionary Computation(EC),chaos optimization algorithm and Fuzzy c-means cluster algorithm(FCM). We present two novel fuzzy cluster algorithms, including Immune-evolution FCM algorithm(IEFCMA) and chaos Immune-evolution FCM algorithm(CIEFCMA),which used to avoid getting in local optimum, improve clustering efficiency and enhance convergence speed.Firstly, this paper introduces cluster, fuzzy cluster analysis and Fuzzy c-means cluster algorithm. Fuzzy c-means cluster algorithm is the most widespread in fuzzy cluster analysis. However its vital shortcoming is the sensibility to initial value, it is easy to run into a local optimum.Secondly, this paper introduces Artificial Immune System, Evolutionary Computation. This paper describes basic structure, theory and characteristics, and discusses the disadvantages and advantages of them. The paper introduces current status of the research on immune evolution algorithm.Thirdly, this paper introduces chaos theory, Logistic chaos optimization operator and Tent chaos optimization operator. Chaos has ergodicity, which can be used to an optimization mechanism. Chaos optimization algorithm can be used to avoid getting in local optimum relation and enhance search efficiency.Fourthly, to overcome the shortcomings of the existing Fuzzy clustering algorithms, this paper introduce two new algorithms, which combines the data searching advantages of Artificial Immune System, Evolutionary Computation and chaotic optimization algorithm with Fuzzy c-means cluster algorithm. These algorithms greatly enhance the local searching efficiency and the global searching performance. These algorithms introduce a kind of cluster-center-based floating point encoding mode, clone selection operator, immune memory, oblivion, the control of antibody concentration, evolution operator, chaotic optimization operator, evolutionary operator. And the simulation shows that the proposed new ideas and new methods on fuzzy cluster analysis are effective, these algorithms not only can be used to avoid precocity, which usually occurs in the common cluster algorithm, but also has a more efficient convergence rate and accuracy.