Clustering And Classification Of Data And Text Using Such Technologies As Genetic Algorithm

Recently data mining and text mining are important research areas in information technology. Applying genetic algorithm, one of soft computing technologies, to data mining and text mining has a great theoretical significance and practical value. Several methods of data mining and text mining have been studied in this paper, which mainly includes: attribute reduction methods, clustering methods. The main works are shown as follows.This paper presents a new clustering method based on Genetic Algorithm and K-medoids algorithm. The method can better solve not only the problem of local optimization but also the problem of isolated points. At the same time, the new method may expedite the convergence of GA and save the time cost because of introducing the k-medoids algorithm in GA.In this paper ,the attributes of text are reduced by using Pattern Aggregation and Genetic algorithm. The dimensions of text can be reduced greatly by using Pattern Aggregation which can reduce the dimentions to several hundreds dimensions from several thousands. In this base, this paper contiues to reduce the dimensions by using genetic algorithm.In this paper, the attributes of text are reduced by using Latent Semantic Analysis and Genetic algorithm. The dimensions of Vector Space model can be reduced greatly by Singular Value Decomposition of Latent Semantic Analysis. In this base, this paper contiues to reduce the dimensions by using genetic algorithm.A new clustering method is proposed by using Social Evolutionary Programming. K-means clustering algorithm usually stops at place of local optimization and finding the global...