Multi‐head attention graph convolutional network model: End‐to‐end entity and relation joint extraction based on multi‐head attention graph convolutional network
Abstract At present, the entity and relation joint extraction task has attracted more and more scholars' attention in the field of natural language processing (NLP). However, most of their methods rely on NLP tools to construct dependency trees to obtain sentence structure information. The adja...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Wiley
2023-06-01
|
Series: | CAAI Transactions on Intelligence Technology |
Subjects: | |
Online Access: | https://doi.org/10.1049/cit2.12086 |
_version_ | 1797641494288400384 |
---|---|
author | Zhihua Tao Chunping Ouyang Yongbin Liu Tonglee Chung Yixin Cao |
author_facet | Zhihua Tao Chunping Ouyang Yongbin Liu Tonglee Chung Yixin Cao |
author_sort | Zhihua Tao |
collection | DOAJ |
description | Abstract At present, the entity and relation joint extraction task has attracted more and more scholars' attention in the field of natural language processing (NLP). However, most of their methods rely on NLP tools to construct dependency trees to obtain sentence structure information. The adjacency matrix constructed by the dependency tree can convey syntactic information. Dependency trees obtained through NLP tools are too dependent on the tools and may not be very accurate in contextual semantic description. At the same time, a large amount of irrelevant information will cause redundancy. This paper presents a novel end‐to‐end entity and relation joint extraction based on the multi‐head attention graph convolutional network model (MAGCN), which does not rely on external tools. MAGCN generates an adjacency matrix through a multi‐head attention mechanism to form an attention graph convolutional network model, uses head selection to identify multiple relations, and effectively improve the prediction result of overlapping relations. The authors extensively experiment and prove the method's effectiveness on three public datasets: NYT, WebNLG, and CoNLL04. The results show that the authors’ method outperforms the state‐of‐the‐art research results for the task of entities and relation extraction. |
first_indexed | 2024-03-11T13:46:28Z |
format | Article |
id | doaj.art-851828b9737e4413ac7327bf5272f820 |
institution | Directory Open Access Journal |
issn | 2468-2322 |
language | English |
last_indexed | 2024-03-11T13:46:28Z |
publishDate | 2023-06-01 |
publisher | Wiley |
record_format | Article |
series | CAAI Transactions on Intelligence Technology |
spelling | doaj.art-851828b9737e4413ac7327bf5272f8202023-11-02T10:21:18ZengWileyCAAI Transactions on Intelligence Technology2468-23222023-06-018246847710.1049/cit2.12086Multi‐head attention graph convolutional network model: End‐to‐end entity and relation joint extraction based on multi‐head attention graph convolutional networkZhihua Tao0Chunping Ouyang1Yongbin Liu2Tonglee Chung3Yixin Cao4School of Computer Science University of South China Hengyang ChinaSchool of Computer Science University of South China Hengyang ChinaSchool of Computer Science University of South China Hengyang ChinaDepartment of Computer Science and technology Tsinghua University Beijing ChinaSchool of Computer and Information Systems Singapore Management University Singapore SingaporeAbstract At present, the entity and relation joint extraction task has attracted more and more scholars' attention in the field of natural language processing (NLP). However, most of their methods rely on NLP tools to construct dependency trees to obtain sentence structure information. The adjacency matrix constructed by the dependency tree can convey syntactic information. Dependency trees obtained through NLP tools are too dependent on the tools and may not be very accurate in contextual semantic description. At the same time, a large amount of irrelevant information will cause redundancy. This paper presents a novel end‐to‐end entity and relation joint extraction based on the multi‐head attention graph convolutional network model (MAGCN), which does not rely on external tools. MAGCN generates an adjacency matrix through a multi‐head attention mechanism to form an attention graph convolutional network model, uses head selection to identify multiple relations, and effectively improve the prediction result of overlapping relations. The authors extensively experiment and prove the method's effectiveness on three public datasets: NYT, WebNLG, and CoNLL04. The results show that the authors’ method outperforms the state‐of‐the‐art research results for the task of entities and relation extraction.https://doi.org/10.1049/cit2.12086information retrievalnatural language processing |
spellingShingle | Zhihua Tao Chunping Ouyang Yongbin Liu Tonglee Chung Yixin Cao Multi‐head attention graph convolutional network model: End‐to‐end entity and relation joint extraction based on multi‐head attention graph convolutional network CAAI Transactions on Intelligence Technology information retrieval natural language processing |
title | Multi‐head attention graph convolutional network model: End‐to‐end entity and relation joint extraction based on multi‐head attention graph convolutional network |
title_full | Multi‐head attention graph convolutional network model: End‐to‐end entity and relation joint extraction based on multi‐head attention graph convolutional network |
title_fullStr | Multi‐head attention graph convolutional network model: End‐to‐end entity and relation joint extraction based on multi‐head attention graph convolutional network |
title_full_unstemmed | Multi‐head attention graph convolutional network model: End‐to‐end entity and relation joint extraction based on multi‐head attention graph convolutional network |
title_short | Multi‐head attention graph convolutional network model: End‐to‐end entity and relation joint extraction based on multi‐head attention graph convolutional network |
title_sort | multi head attention graph convolutional network model end to end entity and relation joint extraction based on multi head attention graph convolutional network |
topic | information retrieval natural language processing |
url | https://doi.org/10.1049/cit2.12086 |
work_keys_str_mv | AT zhihuatao multiheadattentiongraphconvolutionalnetworkmodelendtoendentityandrelationjointextractionbasedonmultiheadattentiongraphconvolutionalnetwork AT chunpingouyang multiheadattentiongraphconvolutionalnetworkmodelendtoendentityandrelationjointextractionbasedonmultiheadattentiongraphconvolutionalnetwork AT yongbinliu multiheadattentiongraphconvolutionalnetworkmodelendtoendentityandrelationjointextractionbasedonmultiheadattentiongraphconvolutionalnetwork AT tongleechung multiheadattentiongraphconvolutionalnetworkmodelendtoendentityandrelationjointextractionbasedonmultiheadattentiongraphconvolutionalnetwork AT yixincao multiheadattentiongraphconvolutionalnetworkmodelendtoendentityandrelationjointextractionbasedonmultiheadattentiongraphconvolutionalnetwork |