A HIERARCHICAL DIVISIVE CLUSTERING BASED MULTI-VIEW POINT SIMILARITY MEASURE FOR DOCUMENT CLUSTERING
B. Amuthajanaki, K. P. Jayalakshmi · 2013
As we know a cluster is a collection of similar objects situated together and are divergent to other cluster objects. In this manuscript, we establish divisive based Multi-view point clustering that is based on different similarity measures. With multiple viewpoints, more informative measurement of similarity could be accomplished. Two criterion functions for document clustering are proposed based on this new measure they are, inter cluster and intra-cluster relation between objects. The previous clustering process focused on hierarchical clustering of Multi-view point documents, which are not spotlighted on sparse and high dimensional data. The difficulty this manuscript spotlights on is the classical problem of unsupervised clustering of a data-set. Especially, the bisecting divisive clustering approach is here considered. This advance consists in recursively splitting a cluster into two sub-clusters, starting from the main dataset. We evaluated our approach with previous model on a variety of document collections to validate the advantages of our proposed method.