Document-level Relation Extraction via Separate Relation Representation and Logical Reasoning
Open Access
- 21 August 2023
- journal article
- research article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Information Systems
- Vol. 42 (1) , 1-24
Abstract
Document-level relation extraction (RE) extends the identification of entity/mentions’ relation from the single sentence to the long document. It is more realistic and poses new challenges to relation representation and reasoning skills. In this article, we propose a novel model, SRLR , using S eparate Relation R epresentation and L ogical R easoning considering the indirect relation representation and complex reasoning of evidence sentence problems. Specifically, we first expand the judgment of relational facts from the entity-level to the mention-level, highlighting fine-grained information to capture the relation representation for the entity pair. Second, we propose a logical reasoning module to identify evidence sentences and conduct relational reasoning. Extensive experiments on two publicly available benchmark datasets demonstrate the effectiveness of our proposed SRLR as compared to 19 baseline models. Further ablation study also verifies the effects of the key components.Keywords
Funding Information
- National Natural Science Foundation of China (U21B2009)
- Singapore Ministry of Education (MOE) Academic Research Fund (AcRF) Tier 1 grant
- Beijing Institute of Technology Southeast Academy of Information Technology
This publication has 44 references indexed in Scilit:
- ENCORE2017
- Position-aware Attention and Supervised Data Improve Slot Filling2017
- End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures2016
- Neural Machine Translation of Rare Words with Subword Units2016
- Joint Extraction of Events and Entities within a Document Context2016
- Combining relations for information extraction from free textACM Transactions on Information Systems, 2010
- Distant supervision for relation extraction without labeled data2009
- Web-scale information extraction in knowitall2004
- Solving the multiple instance problem with axis-parallel rectanglesArtificial Intelligence, 1997
- Multimedia document presentation, information extraction, and document formation in MINOS: a model and a systemACM Transactions on Information Systems, 1986