Intelligent sense-enabled lexical search on text documents

Abstract

This paper proposes a new methodology for intelligent sense-enabled lexical search on text documents. The proposed methodology extracts words from an input text document which are semantically related to a particular sense of the query word. The entire methodology is divided in to two tasks namely, Word Sense disambiguation (WSD) of each word in the input text followed by semantic search i.e, extracting those words that are semantically related to a particular sense of the query word. The significance of the proposed methodology is that, to the best of our knowledge this is the first work that supports sense-enabled lexical search in a text document simultaneously considering the problems with polysemous words. Extraction of semantically related words to a given query word has role in many applications such as document indexing, vocabulary learning for humans, machine translation, etc. Experimental results show that the proposed system surpasses the existing system in terms of precision and computational time. This improved precision and execution time enhances the end user’s experience quality in using the system.