NaijaSenti: A Nigerian Twitter Sentiment Corpus for Multilingual Sentiment Analysis

TL;DR

This paper introduces NaijaSenti, a large-scale, human-annotated Twitter sentiment dataset for four Nigerian languages, enabling research in low-resource multilingual sentiment analysis with promising model performance insights.

Contribution

It provides the first extensive sentiment dataset for Nigerian languages, along with collection, processing methods, and evaluation of models tailored for low-resource multilingual sentiment analysis.

Findings

Language-specific models perform best.

Fine-tuning improves sentiment classification accuracy.

Datasets and models are publicly released for research use.

Abstract

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 Yor\`ub\'a ) consisting of around 30,000 annotated tweets per language (and 14,000 for Nigerian-Pidgin), 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 rangeof pre-trained models and transfer strategies on the dataset. We find that language-specific models and language-adaptivefine-tuning generally perform best. We release the datasets, trained models, sentiment lexicons, and code to incentivizeresearch on sentiment analysis in under-represented languages.