Adaptive Latent Entity Expansion for Document Retrieval

Mackie, I., Chatterjee, S., MacAvaney, S. and Dalton, J. (2024) Adaptive Latent Entity Expansion for Document Retrieval. Knowledge-Enhanced Information Retrieval (KEIR@ECIR 2024), Glasgow, UK, 28 March 2024.

Abstract

Despite considerable progress in neural relevance ranking techniques, search engines still struggle to process complex queries effectively — both in terms of precision and recall. Sparse and dense Pseudo- Relevance Feedback (PRF) approaches have the potential to overcome limitations in recall, but are only effective with high precision in the top ranks. In this work, we tackle the problem of search over complex queries using three complementary techniques. First, we demonstrate that applying a strong neural re-ranker before sparse or dense PRF can improve the retrieval effectiveness by 5–8%. Second, we propose an enhanced expansion model, Latent Entity Expansion (LEE), which applies fine-grained word and entity-based relevance modelling incorporating localized features. Specifically, we find that by including both words and entities for expansion achieve a further 2–8% improvement in NDCG. Our analysis also demonstrates that LEE is largely robust to its parameters across datasets and performs well on entity-centric queries. And third, we include an “adaptive” component in the retrieval process, which iteratively refines the re-ranking pool during scoring using the expansion model and avoids re-ranking additional documents. We find that this combination of techniques achieves the best NDCG, MAP and R@1k results on the TREC Robust 2004 and CODEC document datasets.

• Item Type: Conference or Workshop Item
• Status: Published
• Refereed: Yes
• Glasgow Author(s) Enlighten ID: MacAvaney, Dr Sean and Mackie, Iain and Dalton, Dr Jeff and Chatterjee, Dr Shubham
• Authors: Mackie, I., Chatterjee, S., MacAvaney, S., and Dalton, J.
• College/School: College of Science and Engineering; College of Science and Engineering > School of Computing Science

Related URLs

• Organisation