Document-level medical relation extraction via edge-oriented graph neural network based on document structure and external knowledge

BMC Medical Informatics and Decision Making

Xiaolong Wang¹  Ying Xiong¹  Tao Li¹  Qingcai Chen²  Buzhou Tang² 

[1] Harbin Institute of Technology, Shenzhen, China;Harbin Institute of Technology, Shenzhen, China;Peng Cheng Laboratory, Shenzhen, China;

关键词: Medical relation extraction;    Graph neural network;    Document structure;    External knowledge;   

DOI  :  10.1186/s12911-021-01733-1

来源: Springer

【 摘 要 】

ObjectiveRelation extraction (RE) is a fundamental task of natural language processing, which always draws plenty of attention from researchers, especially RE at the document-level. We aim to explore an effective novel method for document-level medical relation extraction.MethodsWe propose a novel edge-oriented graph neural network based on document structure and external knowledge for document-level medical RE, called SKEoG. This network has the ability to take full advantage of document structure and external knowledge.ResultsWe evaluate SKEoG on two public datasets, that is, Chemical-Disease Relation (CDR) dataset and Chemical Reactions dataset (CHR) dataset, by comparing it with other state-of-the-art methods. SKEoG achieves the highest F1-score of 70.7 on the CDR dataset and F1-score of 91.4 on the CHR dataset.ConclusionThe proposed SKEoG method achieves new state-of-the-art performance. Both document structure and external knowledge can bring performance improvement in the EoG framework. Selecting proper methods for knowledge node representation is also very important.

【 授权许可】

CC BY