A spanish dataset for targeted sentiment analysis of political headlines

dc.creatorPérez, Juan Manuel
dc.creatorRecart Zapata, Emilio
dc.creatorAlves Salgueiro, Tomás
dc.creatorFurman, Damián
dc.creatorFernández Larrosa, Pablo Nicolá
dc.date2023-05
dc.date2023-08-23T17:54:10Z
dc.date.accessioned2024-07-24T03:42:21Z
dc.date.available2024-07-24T03:42:21Z
dc.identifierhttp://sedici.unlp.edu.ar/handle/10915/156750
dc.identifier.urihttps://repositorioslatinoamericanos.uchile.cl/handle/2250/9534925
dc.descriptionSubjective texts have been extensively studied due to their potential to influence behaviors. While most research has focused on user-generated texts in social networks, other types of texts, such as news headlines expressing opinions on certain topics, can also influence judgment criteria during political decisions. In this paper, we address the task of Targeted Sentiment Analysis for news headlines related to the 2019 Argentinean Presidential Elections, published by major news outlets. To facilitate research in this area, we present a polarity dataset comprising 1,976 headlines that mention candidates at the target level. Our experiments using state-of-the-art classification algorithms based on pre-trained language models demonstrate the usefulness of target information for this task. We also provide public access to our data and models to foster further research.
dc.descriptionSociedad Argentina de Informática e Investigación Operativa
dc.formatapplication/pdf
dc.format53-66
dc.languageen
dc.rightshttp://creativecommons.org/licenses/by-nc/4.0/
dc.rightsCreative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
dc.subjectCiencias Informáticas
dc.subjectTargeted Sentiment Analysis
dc.subjectlanguage models
dc.subjectpolitical headlines
dc.titleA spanish dataset for targeted sentiment analysis of political headlines
dc.typeArticulo
dc.typeArticulo


Este ítem pertenece a la siguiente institución