Application of Artificial Intelligence in Online Fraud Detection: Research on Intelligent Protection Systems based on College Students' Cybersecurity Education

Abstract

The increasing reliance on digital transactions has exposed college students to heightened fraud risks, necessitating advanced cybersecurity solutions. This meta-analysis examines the effectiveness of Artificial Intelligence (AI)-driven fraud detection systems and their role in protecting college students from online fraud. Findings reveal that deep learning models achieve up to 96.8% accuracy, significantly outperforming traditional rule-based fraud detection methods. AI-driven fraud detection also reduces false positives, enhances response times, and increases user engagement with security alerts. However, regression analysis indicates a strong inverse correlation (r = -0.91) between cybersecurity awareness and fraud incidents, highlighting the critical need for cybersecurity education to complement AI fraud prevention strategies. The study emphasizes the importance of integrating AI-based security frameworks with cybersecurity training programs to enhance digital safety for students. Future efforts should focus on refining AI models and increasing student awareness to mitigate fraud risks effectively.