Ranks Aggregation and Semantic Genetic Approach based Hybrid Model for Query Expansion

DOI

10.2991/ijcis.2017.10.1.4How to use a DOI?

Keywords

Query expansion terms selection; information retrieval; semantic filtering; rank combination; score combination; genetic algorithm

Abstract

Effective query expansion terms selection methods are really very important for improving the accuracy and efficiency of Pseudo-Relevance Feedback (PRF) based automatic query expansion techniques in information retrieval system. These methods remove irrelevant and redundant terms from the top retrieved feedback documents with respect to a user query. Individual terms selection methods have been widely investigated for improving its performance. However, it is always a challenging task to find an individual expansion terms selection method that would outperform other individual methods in most cases. In this paper, first we explore the possibility of improving the overall performance using individual terms selection methods. Second, we propose a model for combining multiple expansion terms selection methods by using a variety of ranks combining approaches. Third, semantic filtering used to filter out semantically irrelevant terms obtained after combining multiple terms selection methods. Fourth, the Genetic Algorithm used to make an optimal combination of query terms and candidate expansion terms obtained by applying ranks combination and semantic filtering approach. Our experimental results demonstrated that our proposed approaches achieved a significant improvement over each individual terms selection methods and related state-of-the-arts approaches.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article under the CC BY-NC license (http://creativecommons.org/licences/by-nc/4.0/).

Download article (PDF)
View full text (HTML)

TY  - JOUR
AU  - Jagendra Singh
PY  - 2017
DA  - 2017/01/01
TI  - Ranks Aggregation and Semantic Genetic Approach based Hybrid Model for Query Expansion
JO  - International Journal of Computational Intelligence Systems
SP  - 34
EP  - 55
VL  - 10
IS  - 1
SN  - 1875-6883
UR  - https://doi.org/10.2991/ijcis.2017.10.1.4
DO  - 10.2991/ijcis.2017.10.1.4
ID  - Singh2017
ER  -