2025
doi: 10.12720/jait.16.10.1430-1441
Enhancing Automated Exam Creation with Retrieval-Augmented Generation for Scalable Educational Assessment
Charaf Hamidi1
,
Mohamed Badiy2 ,
Salma Gaou1 ,
Fatima Amounas2 ,
Mourade Azrour2,* ,
Hicham Tribak3 ,
Abdullah M. Alnajim4 ,
Abdulatif Alabdulatif5
- 1Laboratory of Engineering Sciences, Faculty of Science Agadir, University Ibn Zohr, Agadir, Morocco
- 2MSIA Team, IMIA Laboratory, Faculty of Sciences and Techniques, Moulay Ismail University of Meknes, Errachidia, Morocco
- 3Physics, Energy, and Information Processing, Multidisciplinary Faculty, University Ibn Zohr, Ouarzazate, Morocco
- 4Department of Information Technology, College of Computer, Qassim University, Buraydah, Saudi Arabia
- 5Department of Computer Science, College of Computer, Qassim University, Buraydah, Saudi Arabia
- Manuscript receivedFebruary 24, 2025
- revisedApril 30, 2025
- acceptedMay 20, 2025
- publishedOctober 14, 2025
Abstract
The integration of Artificial Intelligence (AI) into educational assessment has led to significant advancements in automated exam generation. This study proposes a novel Retrieval-Augmented Generation (RAG)-based system designed to automate the creation of exam questions while ensuring contextual accuracy and adaptability to pedagogical needs. By leveraging state-of-the-art Natural Language Processing (NLP) techniques, including LangChain, Facebook AI Similarity Search (FAISS), and Large Language Model Meta AI (LLaMA) 3.2, the system retrieves domain-specific knowledge and generates multiple-choice and open-ended questions dynamically. The framework incorporates an interactive user interface built with Next.js, enabling customizable quiz settings, real-time answer verification, and performance tracking. A structured knowledge base is developed through automated web scraping and PDF content extraction using LLaMA Parse, ensuring the generation of precise and curriculum-aligned questions. The proposed system enhances educational assessment by reducing manual effort, ensuring scalability, and providing immediate feedback to learners. Experimental evaluations demonstrate its effectiveness in producing pedagogically sound Structured Query Language (SQL) assessments in French, highlighting the potential for broader applications in automated exam generation. This research contributes to the growing field of AI-driven educational tools, paving the way for future enhancements in adaptive learning and intelligent assessment methodologies.
Keywords
- retrieval-augmented generation
- natural language processing
- Artificial Intelligence (AI) in education
- automated exam generation
- LangChain
- Facebook AI Similarity Search (FAISS)
- Large Language Model Meta AI (LLaMA) 3
- educational assessment
How to Cite
Charaf Hamidi, Mohamed Badiy, Salma Gaou, Fatima Amounas, Mourade Azrour, Hicham Tribak, Abdullah M. Alnajim, and Abdulatif Alabdulatif, "Enhancing Automated Exam Creation with Retrieval-Augmented Generation for Scalable Educational Assessment," Journal of Advances in Information Technology, vol. 16, no. 10, pp. 1430-1441, 2025. https://doi.org/10.12720/jait.16.10.1430-1441
Copyright & License
Copyright © 2025 by the authors. 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 (CC BY 4.0).
