AI-Powered Study Resource and PYQ Analysis System: A Centralized Platform for Sharing Study Material and Prioritizing Revision Topics
Year: 2026
Type: preprint
Abstract: Before an exam, most students end up doing two chores. They hunt for notes, previous year question papers (PYQs) and PDFs scattered over WhatsApp groups, Telegram channels and personal Drive links, and then they read several years of papers by hand to guess which topics deserve their limited revision time. Resource-sharing platforms handle the first chore and stop there. Research on question-paper analysis comes close to the second, but usually aims to generate new papers rather than to guide a student. This paper proposes one platform that does both. Uploaded PYQs are split into individual questions, mapped to syllabus topics, and scored on how often and how recently each topic was examined and, when marks are printed, how much weight it carried. The ranked topics are linked to notes and PDFs already stored on the platform and shown as a revision list on a plain student dashboard. The system only describes past patterns; it does not predict the questions of an upcoming exam. We present the architecture, the scoring method, a plan for evaluating topic mapping, ranking quality, resource matching and usability, and the copyright and data-quality limits of the approach. The work is mainly a design: a notes-summarization module and a basic topic-frequency count have been prototyped, while the full ranking pipeline and its evaluation remain future work.
Institution: MIT Art, Design and Technology University
Language: English
Subfield: Education
Field: Social Sciences
Domain: Social Sciences
SDG: Quality education
Open Access status: green
OpenAlex ID: W7236300105
