Volume 11, Number 01, January 2021

A Data-Driven Strategy to Combine Word Embeddings in Information Retrieval

  Authors

Alfredo Silva^{1, 2} and Marcelo Mendoza^{1, 2}, ¹Universidad Técnica Federico Santa María, Chile, ²Millenium Institute of Foundational Research on Data, Chile

  Abstract

Word embeddings are vital descriptors of words in unigram representations of documents for many tasks in natural language processing and information retrieval. The representation of queries has been one of the most critical challenges in this area because it consists of a few terms and has little descriptive capacity. Strategies such as average word embeddings can enrich the queries' descriptive capacity since they favor the identification of related terms from the continuous vector representations that characterize these approaches. We propose a datadriven strategy to combine word embeddings. We use Idf combinations of embeddings to represent queries, showing that these representations outperform the average word embeddings recently proposed in the literature. Experimental results on benchmark data show that our proposal performs well, suggesting that data-driven combinations of word embeddings are a promising line of research in ad-hoc information retrieval.

  Keywords

Word embeddings, information retrieval, query representation.