A Pseudo-relevance feedback framework combining relevance matching and semantic matching for information retrieval

作者：

Highlights：

• Relevance matching plays a more important role than semantic matching in information retrieval.

• The proposed framework, which combines relevance matching and semantic matching, is more effective than using either relevance matching or semantic matching.

• Five enhanced models are generated by merging the framework with probability-based PRF models and language-model-based PRF models.

• Our PRF framework combines relevance matching and semantic matching to improve the quality of the feedback documents.

摘要

•Relevance matching plays a more important role than semantic matching in information retrieval.•The proposed framework, which combines relevance matching and semantic matching, is more effective than using either relevance matching or semantic matching.•Five enhanced models are generated by merging the framework with probability-based PRF models and language-model-based PRF models.•Our PRF framework combines relevance matching and semantic matching to improve the quality of the feedback documents.

论文关键词：Information retrieval,Pseudo-relevance feedback,Text similarity,Semantic matching

论文评审过程：Received 23 January 2020, Revised 14 June 2020, Accepted 14 June 2020, Available online 27 June 2020, Version of Record 27 June 2020.

论文官网地址：https://doi.org/10.1016/j.ipm.2020.102342