Joint Extraction of Entities and Relations via Entity and Relation Heterogeneous Graph Attention Networks
Entity and relation extraction (ERE) is a core task in information extraction. This task has always faced the overlap problem. It was found that heterogeneous graph attention networks could enhance semantic analysis and fusion between entities and relations to improve the ERE performance in our prev...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/2/842 |
| _version_ | 1827629086756831232 |
|---|---|
| author | Bo Jiang Jia Cao |
| author_facet | Bo Jiang Jia Cao |
| author_sort | Bo Jiang |
| collection | DOAJ |
| description | Entity and relation extraction (ERE) is a core task in information extraction. This task has always faced the overlap problem. It was found that heterogeneous graph attention networks could enhance semantic analysis and fusion between entities and relations to improve the ERE performance in our previous work. In this paper, an entity and relation heterogeneous graph attention network (ERHGA) is proposed for joint ERE. A heterogeneous graph attention network with a gate mechanism was constructed containing word nodes, subject nodes, and relation nodes to learn and enhance the embedding of parts for relational triple extraction. The ERHGA was evaluated on the public relation extraction dataset named WebNLG. The experimental results demonstrate that the ERHGA, by taking subjects and relations as a priori information, can effectively handle the relational triple extraction problem and outperform all baselines to 93.3%, especially overlapping relational triples. |
| first_indexed | 2024-03-09T13:44:46Z |
| format | Article |
| id | doaj.art-f4daeb8ac0c74748a8b1d34a511597c9 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-09T13:44:46Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-f4daeb8ac0c74748a8b1d34a511597c92023-11-30T21:02:17ZengMDPI AGApplied Sciences2076-34172023-01-0113284210.3390/app13020842Joint Extraction of Entities and Relations via Entity and Relation Heterogeneous Graph Attention NetworksBo Jiang0Jia Cao1School of Information Science and Technology, Beijing Forestry University, Beijing 100083, ChinaSchool of Information Science and Technology, Beijing Forestry University, Beijing 100083, ChinaEntity and relation extraction (ERE) is a core task in information extraction. This task has always faced the overlap problem. It was found that heterogeneous graph attention networks could enhance semantic analysis and fusion between entities and relations to improve the ERE performance in our previous work. In this paper, an entity and relation heterogeneous graph attention network (ERHGA) is proposed for joint ERE. A heterogeneous graph attention network with a gate mechanism was constructed containing word nodes, subject nodes, and relation nodes to learn and enhance the embedding of parts for relational triple extraction. The ERHGA was evaluated on the public relation extraction dataset named WebNLG. The experimental results demonstrate that the ERHGA, by taking subjects and relations as a priori information, can effectively handle the relational triple extraction problem and outperform all baselines to 93.3%, especially overlapping relational triples.https://www.mdpi.com/2076-3417/13/2/842entity and relation extractionoverlapping triplesheterogeneous graph attention networks |
| spellingShingle | Bo Jiang Jia Cao Joint Extraction of Entities and Relations via Entity and Relation Heterogeneous Graph Attention Networks Applied Sciences entity and relation extraction overlapping triples heterogeneous graph attention networks |
| title | Joint Extraction of Entities and Relations via Entity and Relation Heterogeneous Graph Attention Networks |
| title_full | Joint Extraction of Entities and Relations via Entity and Relation Heterogeneous Graph Attention Networks |
| title_fullStr | Joint Extraction of Entities and Relations via Entity and Relation Heterogeneous Graph Attention Networks |
| title_full_unstemmed | Joint Extraction of Entities and Relations via Entity and Relation Heterogeneous Graph Attention Networks |
| title_short | Joint Extraction of Entities and Relations via Entity and Relation Heterogeneous Graph Attention Networks |
| title_sort | joint extraction of entities and relations via entity and relation heterogeneous graph attention networks |
| topic | entity and relation extraction overlapping triples heterogeneous graph attention networks |
| url | https://www.mdpi.com/2076-3417/13/2/842 |
| work_keys_str_mv | AT bojiang jointextractionofentitiesandrelationsviaentityandrelationheterogeneousgraphattentionnetworks AT jiacao jointextractionofentitiesandrelationsviaentityandrelationheterogeneousgraphattentionnetworks |