Abstract
Clickbait spoiling is a task of generating or retrieving a fairly short text with a purpose to satisfy curiosity of a content consumer without their addressing to the document linked to a clickbait post or headline. In this paper we introduce an ensemble approach to clickbait spoiling task at SemEval-2023. The tasks consists of spoiler classification and retrieval on Webis-Clickbait-22 dataset. We show that such an ensemble solution is quite successful at classification, whereas it might perform poorly at retrieval with no additional features. In conclusion we outline our thoughts on possible directions to improving the approach and shape a set of suggestions to the said features.- Anthology ID:
- 2023.semeval-1.289
- Volume:
- Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023)
- Month:
- July
- Year:
- 2023
- Address:
- Toronto, Canada
- Editors:
- Atul Kr. Ojha, A. Seza Doğruöz, Giovanni Da San Martino, Harish Tayyar Madabushi, Ritesh Kumar, Elisa Sartori
- Venue:
- SemEval
- SIG:
- SIGLEX
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 2100–2106
- Language:
- URL:
- https://aclanthology.org/2023.semeval-1.289
- DOI:
- 10.18653/v1/2023.semeval-1.289
- Cite (ACL):
- Maksim Shmalts. 2023. John Boy Walton at SemEval-2023 Task 5: An Ensemble Approach to Spoiler Classification and Retrieval for Clickbait Spoiling. In Proceedings of the 17th International Workshop on Semantic Evaluation (SemEval-2023), pages 2100–2106, Toronto, Canada. Association for Computational Linguistics.
- Cite (Informal):
- John Boy Walton at SemEval-2023 Task 5: An Ensemble Approach to Spoiler Classification and Retrieval for Clickbait Spoiling (Shmalts, SemEval 2023)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2023.semeval-1.289.pdf