Social media bullying detection using machine learning on Bangla text

with the popularity of Unicode system and growing use of Internet, the use of Bangla over social media is increasing. However, very few works have been done on Bangla text for social media activity monitoring due to a lack of a large number of annotated corpora, named dictionaries and morphological analyzer, which demands in-depth analysis on Bangladesh's perspective. Moreover, solving the issue by applying available techniques is very content specific, which means that false detection can occur if contents changed from formal English to verbal abuse or sarcasm. Also, performance may vary due to linguistic differences between English and non-English contents and the socio-emotional behaviour of the study population. To combat such issues, this paper proposes the use of machine learning algorithms and the inclusion of user information for cyber bullying detection on Bangla text. For this purpose, a set of Bangla text has been collected from available social media platforms and labelled as either bullied or not bullied for training different machine learning based classification models. Cross-validation results of the models indicate that a support vector machine based algorithm achieves superior performance on Bangla text with a detection accuracy of 97%. Besides, the impact of user specific information such as location, age and gender can further improve the classification accuracy of Bangla cyber bullying detection system.