An Efficient Multi-Dimensional Level based Semantic Relational Depthness Clustering for Enhancing Web Document Clustering

Abstract

The problem of web document clustering has been discussed in different situation and there are number of solution has been recommended for the problem of web document clustering. Each method has used different measures to identify the similarity of documents that suffers with the problem of false indexing and produces poor clustering. The existing methods results in more time complexity and poor accuracy in clustering to solve the problems identified in web document clustering. In this paper, an multi-dimensional level based semantic relational depthness clustering is proposed. This proposed method identifies the Scopus terms that has more meaning by removing the stop words and by performing stemming operation. This method uses semantic ontology for each domain of category being considered and the word netsynset to collect the semantic domain ontology. The multi-dimensional level based semantic relational depthness measure is computed for each term present in the term set with the retrieved Scopus terms. The category of the document is identified by using the multi-attribute relational depthness measure. The proposed method improves the performance of web document clustering and reduces the false indexing ratio and time complexity of clustering.