Term Selection Methods for Query Expansion in Pseudo Relevance Feedback

Query expansion term selection methods are very essential to increase the precision and the efficiency of Pseudo-Relevance Feedback (PRF) based automatic query expansion for information retrieval system by excluding distinct and redundant terms from the top retrieved feedback documents collection with respect to the user query. Specific query e n term methods selection have been broadly explored for improving its performance. Still, it is always a challenging task to find a specific query expansion term selection method which would outperform other individual query expansion terms selection methods in maximum cases. In this paper, we explore the individual query expansion term selection methods and compare the performance of the individual term selection methods for query expansion. Our experimental results show that the information delivered by each method is of contrasting nature and, and finally concluding the term selection method for query expansion.