Clustering Analysis Of Multidimensional Data Based On Rough Set

With the development of information technology, to produce a variety of information resources in the network. With the promotion of microblog, microblog users are icreasing with the annual rate of hundreds of millions. That is a challenge to cluster the microblog users with Data Mining,because each user contains dozens of properties of information.This paper mainly studies on the clustering division of multidimensional data set.First, preprocessing the data with Rough set theory to reduce the data dimension, eliminate repeatitive data, and get a new data set.Then calculating the attribute core subset by using knowledge decision system;Secondly, reducting the attributes of Date set through GA and the attribute core subset, to get the important attributes,that is the min reduction. Finally, optimizing Genetic algorithm by improving fitness function according to the attribute reduction and the clustering distance characteristics, to get the clustering results.From the result of processing Sina microblogging users,this method of this paper can easily access the cluster centers of the data sets and data clustering. Simultaneously, increasing a way for processing multidimensional data.