Inter-Personal Relation Extraction Model Based on Bidirectional GRU and Attention Mechanism

Yuming Li, Pin Ni, Gangmin Li, Xutao Wang, Zhenjin Dai

Research output: Chapter in Book or Report/Conference proceedingConference Proceedingpeer-review

1 Citation (Scopus)

Abstract

Inter-Personal Relationship Extraction is an important part of knowledge extraction and is also the fundamental work of constructing the knowledge graph of people's relationships. Compared with the traditional pattern recognition methods, the deep learning methods are more prominent in the relation extraction (RE) tasks. At present, the research of Chinese relation extraction technology is mainly based on the method of kernel function and Distant Supervision. In this paper, we propose a Chinese relation extraction model based on Bidirectional GRU network and Attention mechanism. Combining with the structural characteristics of the Chinese language, the input vector is input in the form of word vectors. Aiming at the problem of context memory, a Bidirectional GRU neural network is used to fuse the input vectors. The feature information of the word level is extracted from a sentence, and the sentence feature is extracted through the Attention mechanism of the word level. To verify the feasibility of this method, we use the distant supervision method to extract data from websites and compare it with existing relationship extraction methods. The experimental results show that Bi-directional GRU with Attention mechanism model can make full use of all the feature information of sentences, and the accuracy of Bi-directional GRU model is significantly higher than that of other neural network models without Attention mechanism.

Original languageEnglish
Title of host publication2019 IEEE 5th International Conference on Computer and Communications, ICCC 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1867-1871
Number of pages5
ISBN (Electronic)9781728147437
DOIs
Publication statusPublished - Dec 2019
Event5th IEEE International Conference on Computer and Communications, ICCC 2019 - Chengdu, China
Duration: 6 Dec 20199 Dec 2019

Publication series

Name2019 IEEE 5th International Conference on Computer and Communications, ICCC 2019

Conference

Conference5th IEEE International Conference on Computer and Communications, ICCC 2019
Country/TerritoryChina
CityChengdu
Period6/12/199/12/19

Keywords

  • Attention
  • Bi-GRU
  • Relation Extraction

Fingerprint

Dive into the research topics of 'Inter-Personal Relation Extraction Model Based on Bidirectional GRU and Attention Mechanism'. Together they form a unique fingerprint.

Cite this