Text clustering ensemble based on genetic algorithms

Text feature is usually expressed as a matrix of huge dimensionality in text mining, and common clustering algorithm are not stable and cannot obtain clustering solution efficiently. Latent Semantic Analysis can reduce dimensionality effectively, and emerges the semantic relations between texts and terms. Clustering ensemble can get better clustering solution than single clustering method. A text clustering ensemble based on genetic algorithms is presented, which combines Latent Semantic Analysis and Clustering ensemble based on genetic algorithms. Experiments have demonstrated that text clustering ensemble based on genetic algorithms can effectively improve the clustering performance.