Font Size: a A A

Multimodal Labeling and Characterization of Social Network Data for Detection and Prediction of Cyberbullying

Posted on:2017-09-22Degree:Ph.DType:Thesis
University:University of Colorado at BoulderCandidate:Hosseinmardi, HFull Text:PDF
GTID:2448390005976200Subject:Computer Science
Abstract/Summary:
This dissertation provides insight into the problem cyberbullying in social networks by investigating profanity usage, ground truth labeling of cyberbullying, and characterization of relationships between cyberbullying and a variety of factors, including profane word usage, social graph features, temporal commenting behavior, linguistic content, and multimedia modality. This thesis investigates how labeled cyberbullying compares to labeled cyberaggression, the utility of labeling images involved in cyberaggression and cyberbullying incidents, the propagation of cyberbullying in social networks, and detecting/predicting cyberbullying in social networks.
Keywords/Search Tags:Cyberbullying, Social, Labeling
Related items