A Sequence-to-Structure Approach to Document-level Targeted Sentiment Analysis

Nan Song, Hongjie Cai, Rui Xia, Jianfei Yu, Zhen Wu, Xinyu Dai


Abstract
Most previous studies on aspect-based sentiment analysis (ABSA) were carried out at the sentence level, while the research of document-level ABSA has not received enough attention. In this work, we focus on the document-level targeted sentiment analysis task, which aims to extract the opinion targets consisting of multi-level entities from a review document and predict their sentiments. We propose a Sequence-to-Structure (Seq2Struct) approach to address the task, which is able to explicitly model the hierarchical structure among multiple opinion targets in a document, and capture the long-distance dependencies among affiliated entities across sentences. In addition to the existing Seq2Seq approach, we further construct four strong baselines with different pretrained models. Experimental results on six domains show that our Seq2Struct approach outperforms all the baselines significantly. Aside from the performance advantage in outputting the multi-level target-sentiment pairs, our approach has another significant advantage - it can explicitly display the hierarchical structure of the opinion targets within a document. Our source code is publicly released at https://github.com/NUSTM/Doc-TSA-Seq2Struct.
Anthology ID:
2023.findings-emnlp.515
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2023
Month:
December
Year:
2023
Address:
Singapore
Editors:
Houda Bouamor, Juan Pino, Kalika Bali
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
7687–7698
Language:
URL:
https://aclanthology.org/2023.findings-emnlp.515
DOI:
10.18653/v1/2023.findings-emnlp.515
Bibkey:
Cite (ACL):
Nan Song, Hongjie Cai, Rui Xia, Jianfei Yu, Zhen Wu, and Xinyu Dai. 2023. A Sequence-to-Structure Approach to Document-level Targeted Sentiment Analysis. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 7687–7698, Singapore. Association for Computational Linguistics.
Cite (Informal):
A Sequence-to-Structure Approach to Document-level Targeted Sentiment Analysis (Song et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-emnlp.515.pdf