A Relational Adaptive Neural Model for Joint Entity and Relation Extraction
Relation extraction is a popular subtask in natural language processing (NLP). In the task of entity relation joint extraction, overlapping entities and multi-type relation extraction in overlapping triplets remain a challenging problem. The classification of relations by sharing the same probabilit...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2021-03-01
|
| Series: | Frontiers in Neurorobotics |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnbot.2021.635492/full |
| _version_ | 1818910590720016384 |
|---|---|
| author | Guiduo Duan Guiduo Duan Jiayu Miao Jiayu Miao Tianxi Huang Wenlong Luo Wenlong Luo Dekun Hu |
| author_facet | Guiduo Duan Guiduo Duan Jiayu Miao Jiayu Miao Tianxi Huang Wenlong Luo Wenlong Luo Dekun Hu |
| author_sort | Guiduo Duan |
| collection | DOAJ |
| description | Relation extraction is a popular subtask in natural language processing (NLP). In the task of entity relation joint extraction, overlapping entities and multi-type relation extraction in overlapping triplets remain a challenging problem. The classification of relations by sharing the same probability space will ignore the correlation information among multiple relations. A relational-adaptive entity relation joint extraction model based on multi-head self-attention and densely connected graph convolution network (which is called MA-DCGCN) is proposed in the paper. In the model, the multi-head attention mechanism is specifically used to assign weights to multiple relation types among entities so as to ensure that the probability space of multiple relation is not mutually exclusive. This mechanism also predicts the strength of the relationship between various relationship types and entity pairs flexibly. The structure information of deeper level in the text graph is extracted by the densely connected graph convolution network, and the interaction information of entity relation is captured. To demonstrate the superior performance of our model, we conducted a variety of experiments on two widely used public datasets, NYT and WebNLG. Extensive results show that our model achieves state-of-the-art performance. Especially, the detection effect of overlapping triplets is significantly improved compared with the several existing mainstream methods. |
| first_indexed | 2024-12-19T22:45:14Z |
| format | Article |
| id | doaj.art-d61af2b78d854a12b82cc6c5d47ad2cc |
| institution | Directory Open Access Journal |
| issn | 1662-5218 |
| language | English |
| last_indexed | 2024-12-19T22:45:14Z |
| publishDate | 2021-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurorobotics |
| spelling | doaj.art-d61af2b78d854a12b82cc6c5d47ad2cc2022-12-21T20:02:58ZengFrontiers Media S.A.Frontiers in Neurorobotics1662-52182021-03-011510.3389/fnbot.2021.635492635492A Relational Adaptive Neural Model for Joint Entity and Relation ExtractionGuiduo Duan0Guiduo Duan1Jiayu Miao2Jiayu Miao3Tianxi Huang4Wenlong Luo5Wenlong Luo6Dekun Hu7School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaTrusted Cloud Computing and Big Data Key Laboratory of Sichuan Province, Chengdu, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaTrusted Cloud Computing and Big Data Key Laboratory of Sichuan Province, Chengdu, ChinaDepartment of Fundamental Courses, Chengdu Textile College, Chengdu, ChinaSchool of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, ChinaTrusted Cloud Computing and Big Data Key Laboratory of Sichuan Province, Chengdu, ChinaCollege of Computer, Chengdu University, Chengdu, ChinaRelation extraction is a popular subtask in natural language processing (NLP). In the task of entity relation joint extraction, overlapping entities and multi-type relation extraction in overlapping triplets remain a challenging problem. The classification of relations by sharing the same probability space will ignore the correlation information among multiple relations. A relational-adaptive entity relation joint extraction model based on multi-head self-attention and densely connected graph convolution network (which is called MA-DCGCN) is proposed in the paper. In the model, the multi-head attention mechanism is specifically used to assign weights to multiple relation types among entities so as to ensure that the probability space of multiple relation is not mutually exclusive. This mechanism also predicts the strength of the relationship between various relationship types and entity pairs flexibly. The structure information of deeper level in the text graph is extracted by the densely connected graph convolution network, and the interaction information of entity relation is captured. To demonstrate the superior performance of our model, we conducted a variety of experiments on two widely used public datasets, NYT and WebNLG. Extensive results show that our model achieves state-of-the-art performance. Especially, the detection effect of overlapping triplets is significantly improved compared with the several existing mainstream methods.https://www.frontiersin.org/articles/10.3389/fnbot.2021.635492/fullentity relation joint extractionoverlapping triplets detectionDCGCNrelational-adaptive mechanismgraph convolutional networks |
| spellingShingle | Guiduo Duan Guiduo Duan Jiayu Miao Jiayu Miao Tianxi Huang Wenlong Luo Wenlong Luo Dekun Hu A Relational Adaptive Neural Model for Joint Entity and Relation Extraction Frontiers in Neurorobotics entity relation joint extraction overlapping triplets detection DCGCN relational-adaptive mechanism graph convolutional networks |
| title | A Relational Adaptive Neural Model for Joint Entity and Relation Extraction |
| title_full | A Relational Adaptive Neural Model for Joint Entity and Relation Extraction |
| title_fullStr | A Relational Adaptive Neural Model for Joint Entity and Relation Extraction |
| title_full_unstemmed | A Relational Adaptive Neural Model for Joint Entity and Relation Extraction |
| title_short | A Relational Adaptive Neural Model for Joint Entity and Relation Extraction |
| title_sort | relational adaptive neural model for joint entity and relation extraction |
| topic | entity relation joint extraction overlapping triplets detection DCGCN relational-adaptive mechanism graph convolutional networks |
| url | https://www.frontiersin.org/articles/10.3389/fnbot.2021.635492/full |
| work_keys_str_mv | AT guiduoduan arelationaladaptiveneuralmodelforjointentityandrelationextraction AT guiduoduan arelationaladaptiveneuralmodelforjointentityandrelationextraction AT jiayumiao arelationaladaptiveneuralmodelforjointentityandrelationextraction AT jiayumiao arelationaladaptiveneuralmodelforjointentityandrelationextraction AT tianxihuang arelationaladaptiveneuralmodelforjointentityandrelationextraction AT wenlongluo arelationaladaptiveneuralmodelforjointentityandrelationextraction AT wenlongluo arelationaladaptiveneuralmodelforjointentityandrelationextraction AT dekunhu arelationaladaptiveneuralmodelforjointentityandrelationextraction AT guiduoduan relationaladaptiveneuralmodelforjointentityandrelationextraction AT guiduoduan relationaladaptiveneuralmodelforjointentityandrelationextraction AT jiayumiao relationaladaptiveneuralmodelforjointentityandrelationextraction AT jiayumiao relationaladaptiveneuralmodelforjointentityandrelationextraction AT tianxihuang relationaladaptiveneuralmodelforjointentityandrelationextraction AT wenlongluo relationaladaptiveneuralmodelforjointentityandrelationextraction AT wenlongluo relationaladaptiveneuralmodelforjointentityandrelationextraction AT dekunhu relationaladaptiveneuralmodelforjointentityandrelationextraction |