Date Added: Aug 2010
Document clustering has been extensively investigated as a methodology for improving document retrieval process. In Traditional clustering algorithm each documents belongs to exactly one cluster & hence cannot detect the multiple themes of a document where as soft clustering algorithm each document can belong to multiple clusters. This paper gives a comparative study of hard clustering & soft clustering algorithm. The World Wide Web is a large repository of information. Over the last the decade the authors have witness an explosive growth in the information available on it.