A Novel Term Selection based Automatic Query Expansion Approach using PRF and Semantic Filtering
Yogesh Gupta1, Ashish Saini2
1Yogesh Gupta, Department of Computer Science and Engineering, Manipal University Jaipur (Rajasthan), India.
2Ashish Saini, Department of Electrical Engineering, Dayalbagh Educational Institute, Agra (U.P), India.
Manuscript received on 28 March 2019 | Revised Manuscript received on 07 April 2019 | Manuscript Published on 11 April 2019 | PP: 130-137 | Volume-8 Issue-4C, April 2019 | Retrieval Number: D24330484C19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Query expansion is the technique to make user’s query precise by providing more relevant terms and term selection is one of the methods of it. This method removes irrelevant and redundant terms from the top ranked documents and enhances the efficiency of Information Retrieval System. There are several term selection methods and each method has its own strength and weakness. This paper introduces an approach which utilizes the strengths of each term selection method and overcomes the weaknesses. The proposed approach is based on pseudo relevance feedback and fuzzy logic. A novel semantic filter is also developed in this work to avoid query drifting problem. Three benchmark datasets CACM, CISI and TREC-3 are used to perform all the experiments and the results are compared with recent state of art in terms of MAP, precision-recall and F-measure. The results demonstrate the superiority of proposed approach over other compared approaches.
Keywords: Automatic Query Expansion, Fuzzy Logic, Pseudo Relevance Feedback, Semantic Filter, Term Selection Method.
Scope of the Article: Semantic Web