Cyberbullying severity detection: A machine learning approach.

With widespread usage of online social networks and its popularity, social networking platforms have given us incalculable opportunities than ever before, and its benefits are undeniable. Despite benefits, people may be humiliated, insulted, bullied, and harassed by anonymous users, strangers, or peers. In this study, we have proposed a cyberbullying detection framework to generate features from Twitter content by leveraging a pointwise mutual information technique. Based on these features, we developed a supervised machine learning solution for cyberbullying detection and multi-class categorization of its severity in Twitter. In the study we applied Embedding, Sentiment, and Lexicon features along with PMI-semantic orientation. Extracted features were applied with Naïve Bayes, KNN, Decision Tree, Random Forest, and Support Vector Machine algorithms. Results from experiments with our proposed framework in a multi-class setting are promising both with respect to Kappa, classifier accuracy and f-measure metrics, as well as in a binary setting. These results indicate that our proposed framework provides a feasible solution to detect cyberbullying behavior and its severity in online social networks. Finally, we compared the results of proposed and baseline features with other machine learning algorithms. Findings of the comparison indicate the significance of the proposed features in cyberbullying detection.

Association of Cyberbullying Involvement With Subsequent Substance Use Among Adolescents.

PURPOSE: Adolescent involvement in cyberbullying is common and involves several roles (witness, perpetrator, or victim). Whether different cyberbullying roles are differentially associated with substance use is unknown. The present study examined the associations of adolescent cyberbullying involvement with use and polyuse of various substances. METHODS: A longitudinal cohort of students in Los Angeles, California (N = 2,768) completed surveys at baseline (10th grade, 2014, mean age = 15.5 years) and 12-month follow-up (11th grade, 2015). Five mutually exclusive cyberbullying roles were identified at baseline-no involvement; witness only; witness and victim; witness and perpetrator; and witness, victim, and perpetrator. Past 6-month use of nine substances and poly-use of multiple substances were assessed at baseline and follow-up. RESULTS: Most students (52.2%) were involved in >1 cyberbullying roles. Relative to no involvement, all cyberbullying roles, including witnessing only, were associated with increased odds of using most substances and polysubstance use at follow-up, after adjusting for sociodemographics and baseline substance use (odds ratios: 1.44 [95% confidence interval: 1.18-1.76] to 5.24 [2.73-10.05]). Relative to the witness-only role, students involved in all three roles were at greater odds of using several substances at follow-up (odds ratios: 1.47 [95% confidence interval: 1.05-2.05] to 2.96 [1.60-5.50]). CONCLUSIONS: Cyberbullying involvement, even witnessing, may be associated with future substance use in adolescence. All cyberbullying roles warrant consideration in understanding and preventing youth substance use.