An Encoder-decoder Architecture with Graph Convolutional Networks for Abstractive Summarization

Qiao Yuan, Pin Ni, Junru Liu, Xiangzhi Tong, Hanzhe Lu, Gangmin Li, Steven Guan

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

2 Citations (Scopus)

Abstract

We propose a single-document abstractive summarization system that integrates token relation into a traditional RNN-based encoder-decoder architecture. We employ pointer-wise mutual information to represent the token relation and adopt Graph Convolutional Networks (GCN) to extract token representation from the relation graph. In our experiment on Gigaword, we consider importing two kinds of structural information: token (node) representation from the relation graph. Also, we implement two kinds of GCNs, a spectral-based one and a spatial-based one, to extract structural information. The result shows that the spatial based GCN-enhanced model with node representation outperforms the classical RNN-based encoder-decoder model.

Original languageEnglish
Title of host publication2021 IEEE 4th International Conference on Big Data and Artificial Intelligence, BDAI 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages91-97
Number of pages7
ISBN (Electronic)9781665412704
DOIs
Publication statusPublished - 2 Jul 2021
Event2021 IEEE 4th International Conference on Big Data and Artificial Intelligence, BDAI 2021 - Qingdao, China
Duration: 2 Jul 20214 Jul 2021

Publication series

Name2021 IEEE 4th International Conference on Big Data and Artificial Intelligence, BDAI 2021

Conference

Conference2021 IEEE 4th International Conference on Big Data and Artificial Intelligence, BDAI 2021
Country/TerritoryChina
CityQingdao
Period2/07/214/07/21

Keywords

  • GCN
  • Seq2Seq
  • natural language processing
  • text summarization

Cite this