Cyberbullying Victimization among Adolescents: Machine Learning Analysis of Survey Data

Cyberbullying has emerged as a pervasive issue affecting children and adolescents globally. This research article aims to empirically explore the prevalence of cyberbullying among adolescents in the U.S., elucidating its various forms and manifestations, and adolescents’ perception of cyberbullying victimization. It also examines the adolescents’ views of the cyberbullying risks and the potential protective factors. This study is based on quantitative data collected through survey interviews with 380 adolescent children aged 12-17 years living in the Hampton Roads area in Virginia, U.S. Three machine learning models, a Principal Component Analysis (PCA) model, a decision-tree model, and a K-Nearest Neighbors (KNN) model, are processed to examine the patterns in their cyberbullying experiences. The deciding factors that impact adolescent children’s perception of cyberbullying victimization are also analyzed. This study contributes to shedding light on understanding cyberbullying experiences among adolescents by highlighting the adolescents’ perception of cyberbullying victimization and what they think would be effective measures to help avoid cyberbullying.