Unsupervised Aspect-Level Sentiment Controllable Style Transfer

Mukuntha Narayanan Sundararaman, Zishan Ahmad, Asif Ekbal, Pushpak Bhattacharyya


Abstract
Unsupervised style transfer in text has previously been explored through the sentiment transfer task. The task entails inverting the overall sentiment polarity in a given input sentence, while preserving its content. From the Aspect-Based Sentiment Analysis (ABSA) task, we know that multiple sentiment polarities can often be present together in a sentence with multiple aspects. In this paper, the task of aspect-level sentiment controllable style transfer is introduced, where each of the aspect-level sentiments can individually be controlled at the output. To achieve this goal, a BERT-based encoder-decoder architecture with saliency weighted polarity injection is proposed, with unsupervised training strategies, such as ABSA masked-language-modelling. Through both automatic and manual evaluation, we show that the system is successful in controlling aspect-level sentiments.
Anthology ID:
2020.aacl-main.33
Volume:
Proceedings of the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing
Month:
December
Year:
2020
Address:
Suzhou, China
Venue:
AACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
303–312
Language:
URL:
https://aclanthology.org/2020.aacl-main.33
DOI:
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2020.aacl-main.33.pdf