Leveraging Edge AI for Real-Time Detection and Prevention of Online Harassment and Cyberbullying
Kathiravan Pannerselvam, R. Saranya, Shanmugavadivu Pichai · Auerbach Publications eBooks · 2025
In the era of rapid technological advancement, online communication has brought opportunities and challenges to the forefront. One prominent challenge is online harassment and cyberbullying, particularly affecting women and vulnerable communities. This chapter delves into the transformative potential of edge AI in addressing this demanding concern by leveraging real-time detection and prevention mechanisms. The primary focus is to create safer online environments, with a specific emphasis on the well-being and security of women. This chapter explores the foundations of edge AI architecture, which comprises three essential layers: device, edge, and cloud layers. These layers collaboratively process and analyze data swiftly, facilitating the identification of harmful online interactions. The pivotal role of natural language processing (NLP) in enhancing the ability to comprehend and analyze online conversations, further strengthening the cyberbullying detection methods. The chapter emphasizes the ethical considerations in deploying edge AI, emphasizing the importance of data privacy and adherence to regulatory frameworks. By integrating edge AI with anomaly detection systems, it reinforce the efforts in preventing online harassment. This integration enables the swift identification of unusual patterns and deviations, facilitating immediate intervention while ensuring user data privacy through local processing. This chapter contributes to the ongoing discourse on responsible digital engagement and the imperative of safeguarding users’ online experiences. Our comprehensive approach prioritizes safety, security, and compliance with privacy regulations, ultimately offering a promising path toward creating a safer, more inclusive, and secure online ecosystem.