Abstract
Structured Sentiment Analysis is the task of extracting sentiment tuples in a graph structure commonly from review texts. We adapt the Aspect-Based Sentiment Analysis pointer network BARTABSA to model this tuple extraction as a sequence prediction task and extend their output grammar to account for the increased complexity of Structured Sentiment Analysis. To predict structured sentiment tuples in languages other than English we swap BART for a multilingual mT5 and introduce a novel Output Length Regularization to mitigate overfitting to common target sequence lengths, thereby improving the performance of the model by up to 70%. We evaluate our approach on seven datasets in five languages including a zero shot crosslingual setting.- Anthology ID:
- 2022.semeval-1.183
- Volume:
- Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022)
- Month:
- July
- Year:
- 2022
- Address:
- Seattle, United States
- Venue:
- SemEval
- SIGs:
- SIGLEX | SIGSEM
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1313–1323
- Language:
- URL:
- https://aclanthology.org/2022.semeval-1.183
- DOI:
- 10.18653/v1/2022.semeval-1.183
- Cite (ACL):
- Jan Pfister, Sebastian Wankerl, and Andreas Hotho. 2022. SenPoi at SemEval-2022 Task 10: Point me to your Opinion, SenPoi. In Proceedings of the 16th International Workshop on Semantic Evaluation (SemEval-2022), pages 1313–1323, Seattle, United States. Association for Computational Linguistics.
- Cite (Informal):
- SenPoi at SemEval-2022 Task 10: Point me to your Opinion, SenPoi (Pfister et al., SemEval 2022)
- PDF:
- https://preview.aclanthology.org/auto-file-uploads/2022.semeval-1.183.pdf
- Data
- MPQA Opinion Corpus