Abstract

With the aim of precisely organizing huge volume of documents into a small amount of considerable clusters, Document clustering has become one among the leading methods by reason of the swift growth of text documents. Conversely, even now there are more than a few complications for document clustering, for instance, accuracy, high dimensionality, scalability, and significant cluster labels, overlapping clusters, and extorting semantics from texts. In this method, semantic associations among phrases are examined and then for characterizing semantic association, lexical chai is utilized. Main phrases are consequently taken out and after that a semantic link graph is constructed on the lexical chai. This research work provides the identification as well as summarization components of the news summarization (NS) system. With the intention of testing this system, Web news pages from the 163 website (www.163.com) with nucleus hints (that are the subject keywords represented by the news authors) are chosen. Experimentation outcomes prove that this technique could perfectly identify Web news pages with a rate of healthier than 96 percent. It is as well proved that the keyword-extraction technique significantly performs better than the techniques rely on term frequency as well as lexical chai. In addition, so as to assess the performance, Experimentations carried out on News datasets. Outcomes confirmed that this technique entirely outdoes the influential news document clustering techniques with improved accuracy. As a result, this approach not only provides more general and meaningful labels for documents, however also efficiently produces overlapping news story clusters.

Paper Details