Learning deep relevance couplings for ad-hoc document retrieval

作者：

Highlights：

• A transformation layer is proposed to capture the transformation between terms.

• A dependency layer is proposed to capture the term dependency.

• The proposed model captures the diverse semantic matching between documents.

• Integrating explicit and implicit matching is explored to improve retrieval.

• Performance of the proposed model is evaluated on several TREC datasets.

摘要

•A transformation layer is proposed to capture the transformation between terms.•A dependency layer is proposed to capture the term dependency.•The proposed model captures the diverse semantic matching between documents.•Integrating explicit and implicit matching is explored to improve retrieval.•Performance of the proposed model is evaluated on several TREC datasets.

论文关键词：Document retrieval,Deep relevance matching,Neural network

论文评审过程：Received 9 November 2019, Revised 1 April 2021, Accepted 1 June 2021, Available online 9 June 2021, Version of Record 15 June 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.115335