GraphDialog: Integrating Graph Knowledge into End-to-End Task-Oriented Dialogue Systems

Shiquan Yang, Rui Zhang, Sarah Erfani


Abstract
End-to-end task-oriented dialogue systems aim to generate system responses directly from plain text inputs. There are two challenges for such systems: one is how to effectively incorporate external knowledge bases (KBs) into the learning framework; the other is how to accurately capture the semantics of dialogue history. In this paper, we address these two challenges by exploiting the graph structural information in the knowledge base and in the dependency parsing tree of the dialogue. To effectively leverage the structural information in dialogue history, we propose a new recurrent cell architecture which allows representation learning on graphs. To exploit the relations between entities in KBs, the model combines multi-hop reasoning ability based on the graph structure. Experimental results show that the proposed model achieves consistent improvement over state-of-the-art models on two different task-oriented dialogue datasets.
Anthology ID:
2020.emnlp-main.147
Volume:
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
Month:
November
Year:
2020
Address:
Online
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1878–1888
Language:
URL:
https://aclanthology.org/2020.emnlp-main.147
DOI:
10.18653/v1/2020.emnlp-main.147
Bibkey:
Cite (ACL):
Shiquan Yang, Rui Zhang, and Sarah Erfani. 2020. GraphDialog: Integrating Graph Knowledge into End-to-End Task-Oriented Dialogue Systems. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1878–1888, Online. Association for Computational Linguistics.
Cite (Informal):
GraphDialog: Integrating Graph Knowledge into End-to-End Task-Oriented Dialogue Systems (Yang et al., EMNLP 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.emnlp-main.147.pdf
Video:
 https://slideslive.com/38938852
Code
 shiquanyang/GraphDialog