Extending Fraud Detection in Students Exams Using AI
Extending Fraud Detection in Students Exams Using AI

Summary/Abstract: The objective of the research is to enhance the functionality of the FraudDetector software agent within the Distributed eLearning Center (DeLC), a platform providing extensive support for e-learning activities. DeLC assists students and teachers in organizing learning materials, addressing knowledge gaps, conducting exams, and fostering personalized elearning environments. The project scope encompasses various extensions, including an agent-oriented environment that enriches functionalities with reactive and proactive intelligent components, referred to as agents or assistants. This paper focuses on the latest evolution of the FraudDetector software agent, transitioning from its base functionality for fraud detection to leveraging artificial intelligence (AI) capabilities. The goal is to integrate AI, specifically the knowledgebase provided by ChatGPT, to enhance FraudDetector's effectiveness. This integration is the primary contribution of the research, aimed at improving fraud detection precision. Experimentation reveals promising results, suggesting that involving ChatGPT enriches FraudDetector's functionality and enhances the agent's precision. Moving forward, the agent's architecture should remain open for collaboration with external AI providers, with efforts to decouple components responsible for integration. The real-world implementation of these findings is pending, warranting further validation through production environment testing.

• Details
• Contents