Copyright © 2026 Authors retain the copyright of this article. This article 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.

@article{209841,
        author = {Alok Dubey and Asst . prof Nikesh Aote and Ayush Verma and Chetna Khapre and Durga Kashyap and Om Thombre},
        title = {LearnInsight: Class-Level Learning Assessment and Gap Detection System},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {13},
        number = {5},
        pages = {3598-3605},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=209841},
        abstract = {Marks and grades show how a student performed but not which concepts were misunderstood, and in large classes teachers cannot manually inspect every response to find shared weaknesses. This paper presents LearnInsight, a class-level learning gap detection system that analyses assessment responses to find weak concepts and shared misconceptions. Multiple-choice answers are checked directly, while free-text and OCR-read handwritten answers are graded by a large language model (Claude API) that names the misconception behind a wrong answer. A transparent rule engine then computes concept mastery per student and per class, flags a concept as a class-wide gap when mastery is below 50%, and flags a shared misconception when 40% or more of the class choose the same wrong answer. Flagged gaps are matched with videos, notes, tutorials and practice sets and shown on a teacher dashboard. A prototype on a sample Grade 8 class test is described, together with the planned evaluation protocol. The system is intended to support, not replace, teacher judgement.},
        keywords = {Learning Analytics; Learning Gap Detection; Misconceptions; Large Language Models; Educational Data Mining; Explainable Analytics; Teacher Dashboard.},
        month = {October},
        }