Exploring Boundaries and Intensities in Offensive and Hate Speech: Unveiling the Complex Spectrum of Social Media Discourse
This work addresses the need for more nuanced detection of hate and offensive speech in Amharic social media, which is incremental as it builds on existing methods with new data and tasks.
The study tackled the problem of oversimplifying hate and offensive speech detection by creating an Amharic benchmark dataset with 8,258 tweets annotated for category, target, and intensity tasks, revealing that most tweets are less offensive and hateful, and showing that the Afro-XLMR-large model achieved F1-scores up to 75.30% and a correlation coefficient of 80.22%.
The prevalence of digital media and evolving sociopolitical dynamics have significantly amplified the dissemination of hateful content. Existing studies mainly focus on classifying texts into binary categories, often overlooking the continuous spectrum of offensiveness and hatefulness inherent in the text. In this research, we present an extensive benchmark dataset for Amharic, comprising 8,258 tweets annotated for three distinct tasks: category classification, identification of hate targets, and rating offensiveness and hatefulness intensities. Our study highlights that a considerable majority of tweets belong to the less offensive and less hate intensity levels, underscoring the need for early interventions by stakeholders. The prevalence of ethnic and political hatred targets, with significant overlaps in our dataset, emphasizes the complex relationships within Ethiopia's sociopolitical landscape. We build classification and regression models and investigate the efficacy of models in handling these tasks. Our results reveal that hate and offensive speech can not be addressed by a simplistic binary classification, instead manifesting as variables across a continuous range of values. The Afro-XLMR-large model exhibits the best performances achieving F1-scores of 75.30%, 70.59%, and 29.42% for the category, target, and regression tasks, respectively. The 80.22% correlation coefficient of the Afro-XLMR-large model indicates strong alignments.