Detection and Analysis of Offensive Online Content in Hausa Language

TL;DR

This work tackles the problem of offensive online content in Hausa, a low-resource language, by conducting two studies—a user study and the creation of the first Hausa offensive-content dataset (HOC)—and by developing a Hausa-specific detection system that surpasses multilingual baselines with recall over $0.70$. It demonstrates that baseline translation approaches struggle due to language nuances and idiomatic usage, underscoring the need for culturally aware NLP in low-resource settings. The paper contributes the HOC dataset, a tailored detection model, and insights from user studies on cyberbullying, revealing that religious and political discourse are particularly prone to offensive content. The research advances safer online environments for Hausa speakers and highlights the importance of diverse, context-aware resources when extending NLP to low-resource languages.

Abstract

Hausa, a major Chadic language spoken by over 100 million people mostly in West Africa is considered a low-resource language from a computational linguistic perspective. This classification indicates a scarcity of linguistic resources and tools necessary for handling various natural language processing (NLP) tasks, including the detection of offensive content. To address this gap, we conducted two set of studies (1) a user study (n=101) to explore cyberbullying in Hausa and (2) an empirical study that led to the creation of the first dataset of offensive terms in the Hausa language. We developed detection systems trained on this dataset and compared their performance against relevant multilingual models, including Google Translate. Our detection system successfully identified over 70% of offensive, whereas baseline models frequently mistranslated such terms. We attribute this discrepancy to the nuanced nature of the Hausa language and the reliance of baseline models on direct or literal translation due to limited data to build purposive detection systems. These findings highlight the importance of incorporating cultural context and linguistic nuances when developing NLP models for low-resource languages such as Hausa. A post hoc analysis further revealed that offensive language is particularly prevalent in discussions related to religion and politics. To foster a safer online environment, we recommend involving diverse stakeholders with expertise in local contexts and demographics. Their insights will be crucial in developing more accurate detection systems and targeted moderation strategies that align with cultural sensitivities.

Figures (4)

• Figure 1: Overview of our approach, which involves (1) conducting user studies and (2) collecting and curating datasets to build detection systems for offensive and threatening content.
• Figure 2: Statistics about language, sentiment and the major themes associated with the HOC dataset.
• Figure 3: Accuracy and F-score performance of the trained models on the task of predicting offensive content (using the HOC dataset).
• Figure 4: A summary of the areas to focus towards improving the detection system. Due to limited resources, the areas highlighted need further studies to build more effective and comprehensive detection systems for offensive content in the Hausa language.