In recent years, social media have become a common plattforms for criminals to stalk, intimidate, manipulate and abuse vulnerable citizens, such as women and youth. A recent survey of students in grades 6 to 9 found that the rates of electronic bullying for girls were between 16% and 19%, whereas the rates for boys were between 11% and 19%. 33.47% of sexually abused girls reported experiencing cyberbullying compared to 17.75% of nonsexually abused girls. There are also high rates of other criminal activities observed in social medias, such as scams, fraud and intellectual property crimes.
This PhD project aims to devise novel methodologies for construct an event driven knowledge base to understand and detect criminal activities (eg - sexual grooming, or misrepresentation) in order to provide adequate advice to users. As knowledge regarding criminal activities predominantly exists in digital documents and multimodal social media environments, the key technical challenges herein are to automatically extract events and mine knowledge from existing unstructured/structured data, and exploit the knowledge via neuro-symbolic reasoning for crime prevention (eg -sexual assaults), especially when there is no large scale training data available.
