Query Expansion with Locally-Trained Word Embeddings

TL;DR

This paper investigates the effectiveness of locally-trained word embeddings for query expansion in information retrieval, showing they outperform global embeddings like word2vec and GloVe in retrieval tasks.

Contribution

It introduces the use of corpus and query-specific embeddings for query expansion, demonstrating their superior performance over traditional global embeddings.

Findings

Locally-trained embeddings outperform global embeddings in retrieval tasks.

Query-specific embeddings improve the effectiveness of query expansion.

Global embeddings like word2vec and GloVe underperform in this context.

Abstract

Continuous space word embeddings have received a great deal of attention in the natural language processing and machine learning communities for their ability to model term similarity and other relationships. We study the use of term relatedness in the context of query expansion for ad hoc information retrieval. We demonstrate that word embeddings such as word2vec and GloVe, when trained globally, underperform corpus and query specific embeddings for retrieval tasks. These results suggest that other tasks benefiting from global embeddings may also benefit from local embeddings.