HERDPhobia: A Dataset for Hate Speech against Fulani in Nigeria

TL;DR

This paper introduces HERDPhobia, a multilingual dataset of hate speech against Fulani herders in Nigeria, and benchmarks models for hate speech detection, achieving high accuracy with the XML-T model.

Contribution

It provides the first annotated hate speech dataset on Fulani herders in Nigeria across three languages and evaluates pre-trained models for classification.

Findings

XML-T model achieves 99.83% weighted F1 score.

The dataset covers English, Nigerian-Pidgin, and Hausa.

Benchmark results establish a baseline for future research.

Abstract

Social media platforms allow users to freely share their opinions about issues or anything they feel like. However, they also make it easier to spread hate and abusive content. The Fulani ethnic group has been the victim of this unfortunate phenomenon. This paper introduces the HERDPhobia - the first annotated hate speech dataset on Fulani herders in Nigeria - in three languages: English, Nigerian-Pidgin, and Hausa. We present a benchmark experiment using pre-trained languages models to classify the tweets as either hateful or non-hateful. Our experiment shows that the XML-T model provides better performance with 99.83% weighted F1. We released the dataset at https://github.com/hausanlp/HERDPhobia for further research.