ALT at SemEval-2020 Task 12: Arabic and English Offensive Language Identification in Social Media

Autor:	Hamdy Mubarak, Ahmed Abdelali, Sabit Hassan, Younes Samih
Rok vydání:	2020
Předmět:	Language identification Arabic business.industry Computer science Offensive computer.software_genre language.human_language SemEval Task (project management) Focus (linguistics) language Social media Artificial intelligence business computer Natural language processing
Zdroj:	SemEval@COLING Scopus-Elsevier
DOI:	10.18653/v1/2020.semeval-1.249
Popis:	This paper describes the systems submitted by the Arabic Language Technology group (ALT) at SemEval-2020 Task 12: Multilingual Offensive Language Identification in Social Media. We focus on sub-task A (Offensive Language Identification) for two languages: Arabic and English. Our efforts for both languages achieved more than 90% macro-averaged F1-score on the official test set. For Arabic, the best results were obtained by a system combination of Support Vector Machine, Deep Neural Network, and fine-tuned Bidirectional Encoder Representations from Transformers (BERT). For English, the best results were obtained by fine-tuning BERT.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_dedup___::0dd6a632f2edfa69055b263f71ef40d5 https://doi.org/10.18653/v1/2020.semeval-1.249 Zobrazit plný text záznamu