Comparison of Algorithms for Document Clustering

Clustering is "the method of organizing objects into groups whose members are related in some way". A cluster is therefore a collection of objects which are coherent internally, but clearly dissimilar to the objects belonging to other clusters. Document clustering is used in many fields such as data mining and information retrieval. Thus, the main goals of this paper are to identify the comparison of the performance of criterion function in the context of partition clustering approach, k means, and agglomerative hierarchical approach. By comparing all this we establish right clustering algorithm to produce qualitative clustering of real world document. And also modify existing algorithm to establish right algorithm which we try to make more efficient than existing algorithms which we are study in this paper.