Building semantic richness among natural language content

Abstract

In this work we propose Inclusive vector to keep the key words available in natural language database. The inclusive vectors are generated by the process of extraction of words given in the source and the cited items of records published in the ISI Thompson Citation Indexes. The proposed inclusive vector exhibits related words and the degree of their relationships. In this work we present the results of the implications of using vectors on the automatic classification of natural language text. In this system, preprocessed documents, extra words as well as word stems are at first found. We have used an enhanced algorithm to bring further semantic relations between the cited and source items in citation databases.