Abstract
Sustainable development requires a significant change in our dietary habits. Argument mining can help achieve this goal by both affecting and helping understand people’s behavior. We design an annotation scheme for argument mining from online discourse around sustainable diets, including novel evidence types specific to this domain. Using Twitter as a source, we crowdsource a dataset of 597 tweets annotated in relation to 5 topics. We benchmark a variety of NLP models on this dataset, demonstrating strong performance in some sub-tasks, while highlighting remaining challenges.- Anthology ID:
- 2022.nlp4pi-1.5
- Volume:
- Proceedings of the Second Workshop on NLP for Positive Impact (NLP4PI)
- Month:
- December
- Year:
- 2022
- Address:
- Abu Dhabi, United Arab Emirates (Hybrid)
- Editors:
- Laura Biester, Dorottya Demszky, Zhijing Jin, Mrinmaya Sachan, Joel Tetreault, Steven Wilson, Lu Xiao, Jieyu Zhao
- Venue:
- NLP4PI
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 40–58
- Language:
- URL:
- https://aclanthology.org/2022.nlp4pi-1.5
- DOI:
- 10.18653/v1/2022.nlp4pi-1.5
- Cite (ACL):
- Marcus Hansen and Daniel Hershcovich. 2022. A Dataset of Sustainable Diet Arguments on Twitter. In Proceedings of the Second Workshop on NLP for Positive Impact (NLP4PI), pages 40–58, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics.
- Cite (Informal):
- A Dataset of Sustainable Diet Arguments on Twitter (Hansen & Hershcovich, NLP4PI 2022)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-4/2022.nlp4pi-1.5.pdf