A probabilistic framework for integrating sentence-level semantics via BERT into pseudo-relevance feedback. Issue 1 (January 2022)

• Record Type: Journal Article
• Authors: Pan, Min; Wang, Junmei; Huang, Jimmy X.; Huang, Angela J.; Chen, Qi; Chen, Jinguang
• Is Part Of: Information processing & management. Volume 59:Issue 1(2022)
• Publication Date: 2022-01
• Subjects: Latent semantic information -- Information retrieval -- Query expansion -- Text similarity -- Pseudo-relevance feedback; Information storage and retrieval systems -- Periodicals; Information science -- Periodicals; Systèmes d'information -- Périodiques; Sciences de l'information -- Périodiques; Information science; Information storage and retrieval systems; Periodicals; 658.4038
• DOI: 10.1016/j.ipm.2021.102734
• Languages: English
• ISSNs: 0306-4573

Abstract

Highlights: We propose a probabilistic framework integrating sentence-level semantic into PRF for expanding terms which are more semantically consistent with the query. Three enhanced models are generated by applying the probability framework to Rocchio-based PRF models. Experimental results highlight the proposed framework is robust and effective. Abstract: Existing pseudo-relevance feedback (PRF) methods often divide an original query into individual terms for processing and select expansion terms based on the term frequency, proximity, position, etc. This process may lose some contextual semantic information from the original query. In this work, based on the classic Rocchio model, we propose a probabilistic framework that incorporates sentence-level semantics via Bidirectional Encoder Representations from Transformers (BERT) into PRF. First, we obtain the importance of terms at the term level. Then, we use BERT to interactively encode the query and sentences in the feedback document to acquire the semantic similarity score of a sentence and the query. Next, the semantic scores of different sentences are summed as the term score at the sentence level. Finally, we balance the term-level and sentence-level weights by adjusting factors and combine the terms with the top-k scores to form a new query for the next-round processing. We apply this method to three Rocchio-based models (Rocchio, PRoc2, and KRoc). A series of experiments are conducted based on six official TREC data sets. Various evaluation indicators suggest that the improved models achieve a significant improvement over the corresponding baseline models. Our proposed models provide a promising avenue for incorporating sentence-level semantics into PRF, which is feasible and robust. Through comparison and analysis of a case study, expansion terms obtained from the proposed models are shown to be more semantically consistent with the query.