AfriHate: A Multilingual Collection of Hate Speech and Abusive Language Datasets for African Languages
Shamsuddeen Hassan Muhammad, Idris Abdulmumin, Abinew Ali Ayele, David Ifeoluwa Adelani, Ibrahim Said Ahmad, Saminu Mohammad Aliyu, Nelson Odhiambo Onyango, Lilian D. A. Wanzare, Samuel Rutunda, Lukman Jibril Aliyu, Esubalew Alemneh, Oumaima Hourrane, Hagos Tesfahun Gebremichael, Elyas Abdi Ismail, Meriem Beloucif, Ebrahim Chekol Jibril, Andiswa Bukula, Rooweither Mabuya, Salomey Osei, Abigail Oppong, Tadesse Destaw Belay, Tadesse Kebede Guge
2025-01-15
Summary
This paper talks about AfriHate, a new collection of datasets that helps identify hate speech and abusive language in 15 different African languages. It's like creating a dictionary of mean words and phrases, but for multiple African languages and cultures.
What's the problem?
In many parts of Africa and other developing regions, there's a big issue with online hate speech and bullying. Current methods to spot and stop this are not working well because they don't understand the local languages and cultures. They either miss a lot of hate speech or end up censoring things that aren't actually harmful. Also, these systems often focus too much on what famous people say and miss widespread hate campaigns against minority groups.
What's the solution?
The researchers created AfriHate, which is like a huge collection of examples of hate speech and mean language in 15 African languages. They got people who speak these languages and understand the local culture to look at each example and label it as hate speech, abusive, or okay. This helps teach computers to recognize hate speech in these languages more accurately. The researchers also shared all their data and results online so other people can use it and improve it.
Why it matters?
This matters because it can help make the internet safer and fairer for people who speak African languages. Better hate speech detection can protect minorities from online bullying and harassment. It can also prevent unnecessary censorship of harmless speech. By including local languages and cultures, it helps ensure that online spaces are welcoming for everyone, not just English speakers. This research could lead to better social media moderation in African countries, helping millions of people have a better online experience.
Abstract
Hate speech and abusive language are global phenomena that need socio-cultural background knowledge to be understood, identified, and moderated. However, in many regions of the Global South, there have been several documented occurrences of (1) absence of moderation and (2) censorship due to the reliance on keyword spotting out of context. Further, high-profile individuals have frequently been at the center of the moderation process, while large and targeted hate speech campaigns against minorities have been overlooked. These limitations are mainly due to the lack of high-quality data in the local languages and the failure to include local communities in the collection, annotation, and moderation processes. To address this issue, we present AfriHate: a multilingual collection of hate speech and abusive language datasets in 15 African languages. Each instance in AfriHate is annotated by native speakers familiar with the local culture. We report the challenges related to the construction of the datasets and present various classification baseline results with and without using LLMs. The datasets, individual annotations, and hate speech and offensive language lexicons are available on https://github.com/AfriHate/AfriHate