Document Expansion by Query Prediction

Citations Over Time

Abstract

One technique to improve the retrieval effectiveness of a search engine is to expand documents with terms that are related or representative of the documents' content.From the perspective of a question answering system, this might comprise questions the document can potentially answer. Following this observation, we propose a simple method that predicts which queries will be issued for a given document and then expands it with those predictions with a vanilla sequence-to-sequence model, trained using datasets consisting of pairs of query and relevant documents. By combining our method with a highly-effective re-ranking component, we achieve the state of the art in two retrieval tasks. In a latency-critical regime, retrieval results alone (without re-ranking) approach the effectiveness of more computationally expensive neural re-rankers but are much faster.

Related Papers

• → Document Retrieval in the Context of Question Answering(2003)41 cited
• → Human question answering performance using an interactive document retrieval system(2012)6 cited
• The Ranking Algorithms of Search Engine(2010)
• → Bilingual Search Engine Augmented with Query Expansion(2007)1 cited
• → Bilingual Search Engine Augmented with Query Expansion(2007)