An efficient data structure for document clustering using k-means algorithm

Abstract

In this paper, we proposed an efficient data structure called "Sparse Matrices" for representing documents. The document database can be represented by using sparse matrices rather than dense matrices. The matrix can be given as an input for k-means algorithm. Using sparse matrices not only will reduce the size of the database as well as it found efficient in running the program. The experimental results have shown that sparse matrices gives good results compared to dense matrices. © 2012 Springer-Verlag GmbH Berlin Heidelberg.