Identifying Victim Blaming Language in Discussions about Sexual Assaults on Twitter

Increasing instances of sexual assault have presented an opportunity for these heinous crimes to be discussed on social platforms. Oftentimes, victims are slut shamed and held culpable for the assault by the community which further discourages such personal disclosures and assault reporting. Victim Blaming has multiple psychological effects on the victim and further discourages formal reporting of such crimes. Therefore, it is important to devise computationally relevant methods to identify and prevent victim blaming to protect the victims. Additionally, specific datasets to devise models should also be developed. In our work, we present an exhaustive statistical analysis of victim blaming and gender attributes along with a single step transfer learning based classification method to identify victim blaming language on Twitter. Finally, we compare the performance of the proposed model against various deep learning and machine learning models on a manually annotated domain-specific dataset.