DefinePK

Abstract

Owing to the fastest data growth, this era can be claimed as the era of zettabytes. An effective mechanism is the need of this hour, to manage all the available data efficiently. Clustering is a technique to group relevant documents together. This work takes the semantics into account and clusters the document with the hybrid of bisecting k-means and UPGMA algorithm. The semantic analysis is made possible by the inclusion of Wordnet, which is a lexical database. The outcome of this algorithm is more accurate, as the clusters are meaningful. The performance of the proposed algorithm is evaluated with respect to precision, recall, F-measure, accuracy and misclassification rate. The experimental results of the proposed work are satisfactory.