Using Word Embeddings for Automatic Query Expansion

TL;DR

"The paper investigates using word2vec word embeddings for automatic query expansion in ad hoc retrieval. It proposes multiple kNN-based QE strategies and an extended query term set that leverage embedding space and compositionality, then evaluates against a strong baseline and RM3 across standard datasets. The results show that while embedding-based QE improves over raw-query retrieval, it generally underperforms RM3, and composition helps; the discussion highlights the complementary strengths of co-occurrence statistics and embedding similarity. The work suggests combining these approaches and exploring localized training to enhance practical effectiveness."

Abstract

In this paper a framework for Automatic Query Expansion (AQE) is proposed using distributed neural language model word2vec. Using semantic and contextual relation in a distributed and unsupervised framework, word2vec learns a low dimensional embedding for each vocabulary entry. Using such a framework, we devise a query expansion technique, where related terms to a query are obtained by K-nearest neighbor approach. We explore the performance of the AQE methods, with and without feedback query expansion, and a variant of simple K-nearest neighbor in the proposed framework. Experiments on standard TREC ad-hoc data (Disk 4, 5 with query sets 301-450, 601-700) and web data (WT10G data with query set 451-550) shows significant improvement over standard term-overlapping based retrieval methods. However the proposed method fails to achieve comparable performance with statistical co-occurrence based feedback method such as RM3. We have also found that the word2vec based query expansion methods perform similarly with and without any feedback information.

Figures (1)

• Figure 1: Difference in AP for individual queries.