SCOUT: Surveillance and Cyber harassment Observation of Unseen Threats
Sumitra Biswal · 2024
Cyberbullying and cyber harassment pose significant challenges to mental health, with traditional detection methods often falling short, particularly in identifying subtle and passive-aggressive behaviors in textual communication. This research addresses these gaps by developing a Generative Artificial Intelligence (AI)-driven solution that enhances detection capabilities and captures the nuanced nature of cyber harassment across various platforms. Utilizing explainable AI techniques, this approach not only improves detection accuracy but also educates users on behavioral implications, supporting victims and legal authorities. The solution integrates with existing anti-harassment policies in academic institutions, workplaces, and social media environments to provide proactive and tailored interventions. To counter the potential misuse by abusers refining their harassment methods, the solution incorporates adaptive learning algorithms that continuously improve based on new data and detected evasion tactics. Regular audits and updates ensure the system evolves with emerging threats and remains resilient against misuse. This comprehensive approach aims to create safer online and offline spaces by addressing both unresolved and emerging issues in cyberbullying detection.