Cyberbullying detection using probabilistic socio-textual information fusion

摘要: Cyberbullying is an important socio-technical challenge in Online Social Networks (OSN). With the growth trends of heterogeneous data OSN, better network characterization, and textual feature sophistication, recent efforts have realized value looking at modes information including features, social image-based features for cyberbullying detection. These approaches, however, still use these either individually or combine them ‘naively’ without considering different confidence levels associated with each interdependencies between features. We propose a novel probabilistic fusion framework that utilizes score uses those to build predictors cyberbullying. The performance proposed approach was compared literature which used similar dataset resulted significant improvements terms