COMPARATIVE STUDY ON TEXT DOCUMENT CLUSTERING ALGORITHMS BASED ON LATENT SEMANTIC INDEXING

In this paper, a comparative analysis of text document clustering algorithms based on latent semantic indexing dimension reduction technique is done. In recent days, the huge amount of textual information is available in electronic form. In order to bring out the interesting patterns from very large text databases, several heuristic algorithms have been developed and still it seems to be quite challenging. Text document clustering is the fastest growing research area for grouping enormous text documents in such a way that documents within a cluster have high intra-similarity and low inter-similarity to other clusters. One of the major issues in document clustering is high dimensionality. Latent Semantic Indexing (LSI) is used with clustering algorithms to analyze the performance of text document clustering algorithms. The experimental results on the dataset constructed from Reuters21578 collections show that dimensionality reduction can improve clustering performance with respect to the computation time and average fitness of the clustered documents.