Code-mixing unveiled: Enhancing the hate speech detection in Arabic dialect tweets using machine learning models (PLOS ONE)

Arabic’s numerous dialects and linguistic subtleties necessitate special attention when it comes to detecting hate speech. A further level of complexity is introduced by the common practice known as “code-mixing,” when users seamlessly combine many languages. In order to close gap, the study investigates the detection capabilities of machine learning models using variation characteristics for hate speech, particularly with regard to code-mixing in Arabic tweets. The approach utilized consists of gathering data, pre-processing it, extracting features, building classification models, and assessing the finished models in order to meet the goals. The results of the investigation showed that the TF-IDF feature achieved the best accuracy of 98.21% when used with the SGD model.

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0305657