Version 1
: Received: 20 February 2024 / Approved: 21 February 2024 / Online: 22 February 2024 (01:22:23 CET)
How to cite:
Cholakov, G.; Stoyanova-Doycheva, A. Extending Fraud Detection in Students Exams Using AI. Preprints2024, 2024021204. https://doi.org/10.20944/preprints202402.1204.v1
Cholakov, G.; Stoyanova-Doycheva, A. Extending Fraud Detection in Students Exams Using AI. Preprints 2024, 2024021204. https://doi.org/10.20944/preprints202402.1204.v1
Cholakov, G.; Stoyanova-Doycheva, A. Extending Fraud Detection in Students Exams Using AI. Preprints2024, 2024021204. https://doi.org/10.20944/preprints202402.1204.v1
APA Style
Cholakov, G., & Stoyanova-Doycheva, A. (2024). Extending Fraud Detection in Students Exams Using AI. Preprints. https://doi.org/10.20944/preprints202402.1204.v1
Chicago/Turabian Style
Cholakov, G. and Asya Stoyanova-Doycheva. 2024 "Extending Fraud Detection in Students Exams Using AI" Preprints. https://doi.org/10.20944/preprints202402.1204.v1
Abstract
The Distributed eLearning Center (DeLC) is a portal, providing extensive support in the day-to-day work when it comes to e-learning content – it helps students and teachers organize their learning materials, fill the gaps in knowledge (for students) and educational approaches (for teachers), organize and conduct exams, and overall, with providing proactive and personalized e-learning environment. The scope of DeLC as a project involves many extensions, covering the aspects of learning, teaching, exams, and collecting statistical information. Such extension is an agent-oriented environment, which enriches the functionalities with intelligent components, that are reactive and proactive, referred to as agents or assistants. This paper is focused on presenting the latest step in the evolution of FraudDetector software agent, which started with base functionality for fraud detection, and now it aims at usage of AI to accomplish its tasks, taking advantage not only from its knowledgebase, but from a much larger one, used by ChatGPT – through integration with it, which is the main contribution of this research, although the real results from production environment are still pending. In this process, agent’s architecture should stay open for collaboration with another external AI provider if necessary, trying to decouple the components, responsible for integration. As of now, the experiments show that involving ChatGPT in FraudDetector’s functionality enriches it and the agent’s precision could be improved this way.
Computer Science and Mathematics, Information Systems
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
