Document type
Bachelor thesisPublication date
Publication license
Please use this identifier to cite or link to this item: https://hdl.handle.net/2445/193858
Detecció automàtica de la ironia en espanyol
Journal Title
Authors
Director/Tutor
Journal ISSN
Volume Title
Related resource
Abstract
[cat] La detecció automàtica del llenguatge figurat està generant cada vegada més interès en l’àmbit de la lingüística computacional. Aquest estudi, emmarcat dins els àmbits del Processament del Llenguatge Natural (PLN) i l’Aprenentatge Automàtic, documenta el procés de construcció d’un classificador de la ironia en textos en espanyol a partir de models com un Support Vector Machine (SVM) i BERT multilingüe. Els millors resultats obtinguts pertanyen al SVM, amb un valor F1 de 0.85 punts. Pel que fa al marc teòric, es parteix de les definicions d’ironia i sarcasme que ofereixen autores com Reyes (1994) i Ruiz Gurillo (2012) i de les consideracions de Kerbrat-Orecchioni (1981) i Reus Boyd-Swan (2009), que suggereixen la presència de diverses marques lingüístiques en el to irònic.
[eng] The automatic detection of figurative language is increasingly generating more interest in the field of computational linguistics. This study, framed within the fields of Natural Language Processing and Machine Learning, documents the building process of an automatic irony classifier in Spanish texts based on models such as a Support Vector Machine (SVM) and multilingual BERT. The best results are obtained with the SVM, with an F1 value of 0.85 points. Regarding the theoretical framework, we consider the definitions of irony and sarcasm offered by authors such as Reyes (1994) and Ruiz Gurillo (2012) and the ideas of Kerbrat-Orecchioni (1981) and Reus Boyd-Swan (2009), which suggest the presence of several linguistic marks in the ironic tone.
[eng] The automatic detection of figurative language is increasingly generating more interest in the field of computational linguistics. This study, framed within the fields of Natural Language Processing and Machine Learning, documents the building process of an automatic irony classifier in Spanish texts based on models such as a Support Vector Machine (SVM) and multilingual BERT. The best results are obtained with the SVM, with an F1 value of 0.85 points. Regarding the theoretical framework, we consider the definitions of irony and sarcasm offered by authors such as Reyes (1994) and Ruiz Gurillo (2012) and the ideas of Kerbrat-Orecchioni (1981) and Reus Boyd-Swan (2009), which suggest the presence of several linguistic marks in the ironic tone.
Description
Treballs Finals de Grau de Lingüística. Facultat de Filologia. Universitat de Barcelona, Curs: 2021-2022, Tutor: Mariona Taulé
Citation
Collections
Citation
RAIMUNDO SCHULZ, Emma. Detecció automàtica de la ironia en espanyol. [consulted: 17 of June of 2026]. Available at: https://hdl.handle.net/2445/193858