AI-Based Framework for Question Paper Preparation

View at publisher Summarise with VIRAT
Authors: N. R. Sathis Kumar, M Vigneshkumar, AK Reddy, G. Karthik kumar Reddy, Bathula Manikanta, Amir Raza
Cited by: 0
References: 6

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Traditionally, faculty convene to create question papers manually or work individually as a part of their teaching and learning cycle. This can take time and does not always address content coverage, difficulty level or topic alignment with the syllabus. This project highlights the initiate and develop Quizzy AI Paper Forge, a system that generates examination questions and answers from the content of study materials in PDF format. The system uses PDF.js to obtain the text, completes some light preprocessing, then relies on combination of AI adherer (Claude, Gemini and Llama) with a fallback rule-based model to sequence the subject content into question formatting. Supabase is used for authenticating, storage and maintaining the question metadata. Each question item is run through a vetting component to check for relevance, and then the difficulty level is assigned, Bloom's taxonomy and topic mapping per question. The selected questions build a composite exam display; customized sourcing semantically consolidated. Each of the twelve course documents (representing approximately 450 questions) was evaluated through relevant subject, adequate grammar, coverage of topics (noting, further work would be required to develop last two points) along with a noticeable decrease in preparation time over manual work.

Fields of Study & Subjects

Artificial Intelligence in Education Educational Technology and Assessment Topic Modeling

Publication details

Journal / source
International Conference on Communication, Computing and Electronics Systems (ICCCES)
Pages
1024–1029
Publication date
21 January 2026
Language
EN