Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation
BackgroundAutomatically extracting relations between chemicals and diseases plays an important role in biomedical text mining. Chemical-disease relation (CDR) extraction aims at extracting complex semantic relationships between entities in documents, which contain intrasentence and intersentence rel...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
JMIR Publications
2020-07-01
|
| Series: | JMIR Medical Informatics |
| Online Access: | https://medinform.jmir.org/2020/7/e17638 |
| _version_ | 1818900348670050304 |
|---|---|
| author | Wang, Jian Chen, Xiaoyu Zhang, Yu Zhang, Yijia Wen, Jiabin Lin, Hongfei Yang, Zhihao Wang, Xin |
| author_facet | Wang, Jian Chen, Xiaoyu Zhang, Yu Zhang, Yijia Wen, Jiabin Lin, Hongfei Yang, Zhihao Wang, Xin |
| author_sort | Wang, Jian |
| collection | DOAJ |
| description | BackgroundAutomatically extracting relations between chemicals and diseases plays an important role in biomedical text mining. Chemical-disease relation (CDR) extraction aims at extracting complex semantic relationships between entities in documents, which contain intrasentence and intersentence relations. Most previous methods did not consider dependency syntactic information across the sentences, which are very valuable for the relations extraction task, in particular, for extracting the intersentence relations accurately.
ObjectiveIn this paper, we propose a novel end-to-end neural network based on the graph convolutional network (GCN) and multihead attention, which makes use of the dependency syntactic information across the sentences to improve CDR extraction task.
MethodsTo improve the performance of intersentence relation extraction, we constructed a document-level dependency graph to capture the dependency syntactic information across sentences. GCN is applied to capture the feature representation of the document-level dependency graph. The multihead attention mechanism is employed to learn the relatively important context features from different semantic subspaces. To enhance the input representation, the deep context representation is used in our model instead of traditional word embedding.
ResultsWe evaluate our method on CDR corpus. The experimental results show that our method achieves an F-measure of 63.5%, which is superior to other state-of-the-art methods. In the intrasentence level, our method achieves a precision, recall, and F-measure of 59.1%, 81.5%, and 68.5%, respectively. In the intersentence level, our method achieves a precision, recall, and F-measure of 47.8%, 52.2%, and 49.9%, respectively.
ConclusionsThe GCN model can effectively exploit the across sentence dependency information to improve the performance of intersentence CDR extraction. Both the deep context representation and multihead attention are helpful in the CDR extraction task. |
| first_indexed | 2024-12-19T20:02:26Z |
| format | Article |
| id | doaj.art-5f1826674e994a7ca3434d9a87c9db83 |
| institution | Directory Open Access Journal |
| issn | 2291-9694 |
| language | English |
| last_indexed | 2024-12-19T20:02:26Z |
| publishDate | 2020-07-01 |
| publisher | JMIR Publications |
| record_format | Article |
| series | JMIR Medical Informatics |
| spelling | doaj.art-5f1826674e994a7ca3434d9a87c9db832022-12-21T20:07:37ZengJMIR PublicationsJMIR Medical Informatics2291-96942020-07-0187e1763810.2196/17638Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and ValidationWang, JianChen, XiaoyuZhang, YuZhang, YijiaWen, JiabinLin, HongfeiYang, ZhihaoWang, XinBackgroundAutomatically extracting relations between chemicals and diseases plays an important role in biomedical text mining. Chemical-disease relation (CDR) extraction aims at extracting complex semantic relationships between entities in documents, which contain intrasentence and intersentence relations. Most previous methods did not consider dependency syntactic information across the sentences, which are very valuable for the relations extraction task, in particular, for extracting the intersentence relations accurately. ObjectiveIn this paper, we propose a novel end-to-end neural network based on the graph convolutional network (GCN) and multihead attention, which makes use of the dependency syntactic information across the sentences to improve CDR extraction task. MethodsTo improve the performance of intersentence relation extraction, we constructed a document-level dependency graph to capture the dependency syntactic information across sentences. GCN is applied to capture the feature representation of the document-level dependency graph. The multihead attention mechanism is employed to learn the relatively important context features from different semantic subspaces. To enhance the input representation, the deep context representation is used in our model instead of traditional word embedding. ResultsWe evaluate our method on CDR corpus. The experimental results show that our method achieves an F-measure of 63.5%, which is superior to other state-of-the-art methods. In the intrasentence level, our method achieves a precision, recall, and F-measure of 59.1%, 81.5%, and 68.5%, respectively. In the intersentence level, our method achieves a precision, recall, and F-measure of 47.8%, 52.2%, and 49.9%, respectively. ConclusionsThe GCN model can effectively exploit the across sentence dependency information to improve the performance of intersentence CDR extraction. Both the deep context representation and multihead attention are helpful in the CDR extraction task.https://medinform.jmir.org/2020/7/e17638 |
| spellingShingle | Wang, Jian Chen, Xiaoyu Zhang, Yu Zhang, Yijia Wen, Jiabin Lin, Hongfei Yang, Zhihao Wang, Xin Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation JMIR Medical Informatics |
| title | Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation |
| title_full | Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation |
| title_fullStr | Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation |
| title_full_unstemmed | Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation |
| title_short | Document-Level Biomedical Relation Extraction Using Graph Convolutional Network and Multihead Attention: Algorithm Development and Validation |
| title_sort | document level biomedical relation extraction using graph convolutional network and multihead attention algorithm development and validation |
| url | https://medinform.jmir.org/2020/7/e17638 |
| work_keys_str_mv | AT wangjian documentlevelbiomedicalrelationextractionusinggraphconvolutionalnetworkandmultiheadattentionalgorithmdevelopmentandvalidation AT chenxiaoyu documentlevelbiomedicalrelationextractionusinggraphconvolutionalnetworkandmultiheadattentionalgorithmdevelopmentandvalidation AT zhangyu documentlevelbiomedicalrelationextractionusinggraphconvolutionalnetworkandmultiheadattentionalgorithmdevelopmentandvalidation AT zhangyijia documentlevelbiomedicalrelationextractionusinggraphconvolutionalnetworkandmultiheadattentionalgorithmdevelopmentandvalidation AT wenjiabin documentlevelbiomedicalrelationextractionusinggraphconvolutionalnetworkandmultiheadattentionalgorithmdevelopmentandvalidation AT linhongfei documentlevelbiomedicalrelationextractionusinggraphconvolutionalnetworkandmultiheadattentionalgorithmdevelopmentandvalidation AT yangzhihao documentlevelbiomedicalrelationextractionusinggraphconvolutionalnetworkandmultiheadattentionalgorithmdevelopmentandvalidation AT wangxin documentlevelbiomedicalrelationextractionusinggraphconvolutionalnetworkandmultiheadattentionalgorithmdevelopmentandvalidation |