NaijaSenti: A Nigerian Twitter Sentiment Corpus for Multilingual Sentiment Analysis

• Tipo: Artigo em Livro de Atas de Conferência Internacional
• Data: 2022

Autores

Muhammad, SH (Autor) Outra; Adelani, DI (Autor) Outra; Ruder, S (Autor) Outra; Ahmad, IS (Autor) Outra; Abdulmumin, I (Autor) Outra; Bello, BS (Autor) Outra; Choudhury, M (Autor) Outra; Emezue, CC (Autor) Outra; Abdullahi, SS (Autor) Outra; Aremu, A (Autor) Outra; Jorge, AM (Autor) FCUP; Pavel Brazdil (Autor) FEP

Ata de Conferência Internacional

Título: Proceedings of the 13th Language Resources and Evaluation Conference (LREC 2022)

Páginas: 590-602

LREC 2022: Language Resources and Evaluation Conference, 13

Marseille, 2022

Indexação

ISI Web of Knowledge - 0 Citações

Outras Informações

ID Authenticus: P-00W-5M1

Abstract (EN): Sentiment analysis is one of the most widely studied applications in NLP, but most work focuses on languages with large amounts of data. We introduce the first large-scale human-annotated Twitter sentiment dataset for the four most widely spoken languages in Nigeria-Hausa, Igbo, Nigerian-Pidgin, and Yoruba-consisting of around 30,000 annotated tweets per language, including a significant fraction of code-mixed tweets. We propose text collection, filtering, processing, and labeling methods that enable us to create datasets for these low-resource languages. We evaluate a range of pre-trained models and transfer strategies on the dataset. We find that language-specific models and language-adaptive fine-tuning generally perform best. We release the datasets, trained models, sentiment lexicons, and code to incentivize research on sentiment analysis in under-represented languages.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 13