Dual Attention Guided Graph Convolutional Networks for Relation Extraction

LI Zhi-xin; SUN Ya-ru; TANG Su-qin; ZHANG Can-long; MA Hui-fang

doi:10.12263/DZXB.20191105

您当前的位置：

首页 >

文章列表页 >

Dual Attention Guided Graph Convolutional Networks for Relation Extraction

更新时间：2025-12-08

- Dual Attention Guided Graph Convolutional Networks for Relation Extraction
- Acta Electronica Sinica Vol. 49, Issue 2, Pages: 315-323(2021)
- 作者机构：
  
  1. 广西师范大学广西多源信息挖掘与安全重点实验室,广西,桂林,541004
  2. 西北师范大学计算机科学与工程学院,甘肃,兰州,730070
  3. 广西师范大学广西多源信息挖掘与安全重点实验室,广西,桂林,541004
  4. 西北师范大学计算机科学与工程学院,甘肃,兰州,730070
- 作者简介：
- 基金信息：
- DOI：10.12263/DZXB.20191105
  CLC： TP391
- Published Online：25 February 2021，
  
  Published：2021
- 稿件说明：
移动端阅览
LI Zhi-xin, SUN Ya-ru, TANG Su-qin, et al. Dual Attention Guided Graph Convolutional Networks for Relation Extraction[J]. Acta Electronica Sinica, 2021, 49(2): 315-323.
DOI：

LI Zhi-xin, SUN Ya-ru, TANG Su-qin, et al. Dual Attention Guided Graph Convolutional Networks for Relation Extraction[J]. Acta Electronica Sinica, 2021, 49(2): 315-323. DOI： 10.12263/DZXB.20191105.

摘要

为了更好地学习节点依赖并利用结构信息，本文提出一种以完全依赖树作为直接输入的新方法，利用图卷积网络并结合两个并行的注意力模块，自主学习如何有选择地关注对关系抽取任务有用的信息.该方法将样本表示成图上的各节点，一个模块用于计算节点特征位置之间的影响，使特征向量可以包含更广范围的语义信息，另一个用于计算节点依赖的关系特征，以增强节点间的全局依赖.两个模块并行相互提升，可以得到完整的特征表示.在TACRED和SemEval数据集上的实验结果表明，该方法能够更有效地获取对关系抽取任务有益的信息，在各评价指标上取得了更好的性能.

Abstract

To better learn node dependence and make use of structural information

this paper proposes a new method that takes the tree of complete dependence as the direct input. The method uses the graph convolutional network and combines two parallel attention modules to learn how to select the useful information. The method represents the samples as nodes on the graph. One module is used to compute the influence between positions of node features

which allows the feature vector to contain a wider range of semantic information. The other one is used to compute the relational features of node dependence

so as to enhance the global dependence between nodes. The two modules promote each other in parallel to obtain complete feature representation. The experimental results on the TACRED and SemEval datasets show that the method can obtain more useful information for relation extraction

thus achieve better performances on various evaluation metrics.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A Rumor Detection Approach Based on Multi-Feature Propagation Tree

Neighborhood and Hypergraph Collaboration for Session-Based Recommendation

Object Detection Based on EIMYOLO for High-Resolution Remote Sensing Images

Single-Image High Dynamic Range Reconstruction Based on Multi-Attention and Perceptual Weighted Learning

Related Author

ZHANG Xin-xin

PAN Shan-liang

MAO Qin-jiao

ZHANG Xin-xin

PAN Shan-liang

Mao Qin-jiao

CHEN Rong-yuan

WEN Jie-bin

Related Institution

College of Information Science and Engineering,Ningbo University

College of Cyberspace Security, Ningbo University of Technology

College of Information Science and Engineering，Ningbo University

College of Cyberspace Security，Ningbo University of Technology

College of Frontier Intersection, Hunan University of Technology and Business

⁰