Learning Matching Models with Weak Supervision for Response Selection in Retrieval-based Chatbots

Yu Wu, Wei Wu, Zhoujun Li, Ming Zhou


Abstract
We propose a method that can leverage unlabeled data to learn a matching model for response selection in retrieval-based chatbots. The method employs a sequence-to-sequence architecture (Seq2Seq) model as a weak annotator to judge the matching degree of unlabeled pairs, and then performs learning with both the weak signals and the unlabeled data. Experimental results on two public data sets indicate that matching models get significant improvements when they are learned with the proposed method.
Anthology ID:
P18-2067
Volume:
Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
Month:
July
Year:
2018
Address:
Melbourne, Australia
Editors:
Iryna Gurevych, Yusuke Miyao
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
420–425
Language:
URL:
https://aclanthology.org/P18-2067
DOI:
10.18653/v1/P18-2067
Bibkey:
Cite (ACL):
Yu Wu, Wei Wu, Zhoujun Li, and Ming Zhou. 2018. Learning Matching Models with Weak Supervision for Response Selection in Retrieval-based Chatbots. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), pages 420–425, Melbourne, Australia. Association for Computational Linguistics.
Cite (Informal):
Learning Matching Models with Weak Supervision for Response Selection in Retrieval-based Chatbots (Wu et al., ACL 2018)
Copy Citation:
PDF:
https://aclanthology.org/P18-2067.pdf
Presentation:
 P18-2067.Presentation.pdf
Video:
 https://aclanthology.org/P18-2067.mp4
Data
DoubanDouban Conversation Corpus