NLP-CIC at HASOC 2020: Multilingual offensive language detection using all-in-one model

Segun Taofeek Aroyehun, Alexander Gelbukh

Producción científica: Contribución a una revistaArtículo de la conferenciarevisión exhaustiva

3 Citas (Scopus)

Resumen

We describe our deep learning model submitted to the HASOC 2020 shared task on detection of offensive language in social media in three Indo-European languages: English, German, and Hindi. We fine-tune a pre-trained multilingual encoder on the combination of data provided for the competition. Our submission received a competitive macro- average F1 score of 0.4980 on the English Subtask A as well as comparatively strong performance on the German data.

Idioma originalInglés
Páginas (desde-hasta)331-335
Número de páginas5
PublicaciónCEUR Workshop Proceedings
Volumen2826
EstadoPublicada - 2020
EventoWorking Notes of FIRE - 12th Forum for Information Retrieval Evaluation, FIRE-WN 2020 - Hyderabad, India
Duración: 16 dic. 202020 dic. 2020

Huella

Profundice en los temas de investigación de 'NLP-CIC at HASOC 2020: Multilingual offensive language detection using all-in-one model'. En conjunto forman una huella única.

Citar esto