Detecting Hate Speech on Social Media with Respect to Adolescent Vulnerability

Social media has become one of the biggest plat-forms for children and young adolescents to spend their free time. Due to the rising age gap between people using social media, tweens can be exposed to offensive and provocative posts made by others that are older. Twitter allows children aged 13 and up to create an account on their platform. While this is a common age restriction on most social media platforms, it can be damaging to young adolescents who are at a crucial time developing their morals and beliefs based on what they see online. Identifying hate speech within a timely matter is crucial for censoring hate speech for kids. This can impact the overall environment on social media to be less toxic and uplifting to all users. In this work, we propose to use multiple machine learning techniques: SVM, k-nearest neighbor, Naive Bayes, and soft-voting ensemble classifier.