Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach

Jingjing Xu; Xu Sun; Qi Zeng; Xuancheng Ren; Xiaodong Zhang; Houfeng Wang; Wenjie Li

Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach

Jingjing Xu, Xu Sun, Qi Zeng, Xuancheng Ren, Xiaodong Zhang, Houfeng Wang, Wenjie Li

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

133 Citations (Scopus)

Abstract

The goal of sentiment-to-sentiment “translation” is to change the underlying sentiment of a sentence while keeping its content. The main challenge is the lack of parallel data. To solve this problem, we propose a cycled reinforcement learning method that enables training on unpaired data by collaboration between a neutralization module and an emotionalization module. We evaluate our approach on two review datasets, Yelp and Amazon. Experimental results show that our approach significantly outperforms the state-of-the-art systems. Especially, the proposed method substantially improves the content preservation performance. The BLEU score is improved from 1.64 to 22.46 and from 0.56 to 14.06 on the two datasets, respectively.

Original language	English
Title of host publication	ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)
Publisher	Association for Computational Linguistics (ACL)
Pages	979-988
Number of pages	10
ISBN (Electronic)	9781948087322
Publication status	Published - 1 Jan 2018
Event	56th Annual Meeting of the Association for Computational Linguistics, ACL 2018 - Melbourne, Australia Duration: 15 Jul 2018 → 20 Jul 2018

Publication series

Name	ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)
Volume	1

Conference

Conference	56th Annual Meeting of the Association for Computational Linguistics, ACL 2018
Country/Territory	Australia
City	Melbourne
Period	15/07/18 → 20/07/18

ASJC Scopus subject areas

Software
Computational Theory and Mathematics

More information

http://hdl.handle.net/10397/89106

Cite this

Xu, J., Sun, X., Zeng, Q., Ren, X., Zhang, X., Wang, H., & Li, W. (2018). Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach. In ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers) (pp. 979-988). (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers); Vol. 1). Association for Computational Linguistics (ACL). http://hdl.handle.net/10397/89106

Xu, Jingjing ; Sun, Xu ; Zeng, Qi et al. / Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach. ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers). Association for Computational Linguistics (ACL), 2018. pp. 979-988 (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)).

@inproceedings{536f8c4732534d7a8c729745d4c4cd06,

title = "Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach",

abstract = "The goal of sentiment-to-sentiment “translation” is to change the underlying sentiment of a sentence while keeping its content. The main challenge is the lack of parallel data. To solve this problem, we propose a cycled reinforcement learning method that enables training on unpaired data by collaboration between a neutralization module and an emotionalization module. We evaluate our approach on two review datasets, Yelp and Amazon. Experimental results show that our approach significantly outperforms the state-of-the-art systems. Especially, the proposed method substantially improves the content preservation performance. The BLEU score is improved from 1.64 to 22.46 and from 0.56 to 14.06 on the two datasets, respectively.",

author = "Jingjing Xu and Xu Sun and Qi Zeng and Xuancheng Ren and Xiaodong Zhang and Houfeng Wang and Wenjie Li",

year = "2018",

month = jan,

day = "1",

language = "English",

series = "ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)",

publisher = "Association for Computational Linguistics (ACL)",

pages = "979--988",

booktitle = "ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)",

note = "56th Annual Meeting of the Association for Computational Linguistics, ACL 2018 ; Conference date: 15-07-2018 Through 20-07-2018",

}

Xu, J, Sun, X, Zeng, Q, Ren, X, Zhang, X, Wang, H & Li, W 2018, Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach. in ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers). ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers), vol. 1, Association for Computational Linguistics (ACL), pp. 979-988, 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018, Melbourne, Australia, 15/07/18. <http://hdl.handle.net/10397/89106>

Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach. / Xu, Jingjing; Sun, Xu; Zeng, Qi et al.
ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers). Association for Computational Linguistics (ACL), 2018. p. 979-988 (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers); Vol. 1).

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach

AU - Xu, Jingjing

AU - Sun, Xu

AU - Zeng, Qi

AU - Ren, Xuancheng

AU - Zhang, Xiaodong

AU - Wang, Houfeng

AU - Li, Wenjie

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The goal of sentiment-to-sentiment “translation” is to change the underlying sentiment of a sentence while keeping its content. The main challenge is the lack of parallel data. To solve this problem, we propose a cycled reinforcement learning method that enables training on unpaired data by collaboration between a neutralization module and an emotionalization module. We evaluate our approach on two review datasets, Yelp and Amazon. Experimental results show that our approach significantly outperforms the state-of-the-art systems. Especially, the proposed method substantially improves the content preservation performance. The BLEU score is improved from 1.64 to 22.46 and from 0.56 to 14.06 on the two datasets, respectively.

AB - The goal of sentiment-to-sentiment “translation” is to change the underlying sentiment of a sentence while keeping its content. The main challenge is the lack of parallel data. To solve this problem, we propose a cycled reinforcement learning method that enables training on unpaired data by collaboration between a neutralization module and an emotionalization module. We evaluate our approach on two review datasets, Yelp and Amazon. Experimental results show that our approach significantly outperforms the state-of-the-art systems. Especially, the proposed method substantially improves the content preservation performance. The BLEU score is improved from 1.64 to 22.46 and from 0.56 to 14.06 on the two datasets, respectively.

UR - http://www.scopus.com/inward/record.url?scp=85055705783&partnerID=8YFLogxK

M3 - Conference article published in proceeding or book

AN - SCOPUS:85055705783

T3 - ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)

SP - 979

EP - 988

BT - ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)

PB - Association for Computational Linguistics (ACL)

T2 - 56th Annual Meeting of the Association for Computational Linguistics, ACL 2018

Y2 - 15 July 2018 through 20 July 2018

ER -

Xu J, Sun X, Zeng Q, Ren X, Zhang X, Wang H et al. Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach. In ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers). Association for Computational Linguistics (ACL). 2018. p. 979-988. (ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers)).

Unpaired sentiment-to-sentiment translation: A cycled reinforcement learning approach

Abstract

Publication series

Conference

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this