Query expansion using Haar wavelet transform

Abstract

Novice users are unable to express their information needs properly, due to this it is difficult to retrieve all the desired relevant documents from the test collection. The problem of word mismatch is fundamental to information retrieval. Query expansion is a technique in which additional terms are added to retrieve relevant documents. In this article, we have expanded the query using pseudo-relevance feedback and Haar wavelet transform. The performance of the proposed technique is evaluated on FIRE 2011 ad hoc English test Collection and Robust dataset. The mean average precision of the proposed model on the FIRE dataset and Robust Track dataset is 0.3334 and 0.2724, respectively.