Exploring Implicit Sentiment Evoked by Fine-grained News Events

Cynthia Van Hee, Orphee De Clercq, Veronique Hoste


Abstract
We investigate the feasibility of defining sentiment evoked by fine-grained news events. Our research question is based on the premise that methods for detecting implicit sentiment in news can be a key driver of content diversity, which is one way to mitigate the detrimental effects of filter bubbles that recommenders based on collaborative filtering may produce. Our experiments are based on 1,735 news articles from major Flemish newspapers that were manually annotated, with high agreement, for implicit sentiment. While lexical resources prove insufficient for sentiment analysis in this data genre, our results demonstrate that machine learning models based on SVM and BERT are able to automatically infer the implicit sentiment evoked by news events.
Anthology ID:
2021.wassa-1.15
Volume:
Proceedings of the Eleventh Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis
Month:
April
Year:
2021
Address:
Online
Venues:
EACL | WASSA
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
138–148
Language:
URL:
https://aclanthology.org/2021.wassa-1.15
DOI:
Bibkey:
Cite (ACL):
Cynthia Van Hee, Orphee De Clercq, and Veronique Hoste. 2021. Exploring Implicit Sentiment Evoked by Fine-grained News Events. In Proceedings of the Eleventh Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis, pages 138–148, Online. Association for Computational Linguistics.
Cite (Informal):
Exploring Implicit Sentiment Evoked by Fine-grained News Events (Van Hee et al., WASSA 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.wassa-1.15.pdf