Administration of Examination Script Evaluation System with Pre-Seeded Answers using AI | IJCSE Volume 9 – Issue 6 | IJCSE-V9I6P5

Grading exams in colleges needs to be quick, precise, and fair – but doing it by hand tends to drag, varies from grader to grader, and suffers from tiredness or personal bias. To fix these issues, this study introduces an AI-driven tool that automatically checks handwritten answers by combining Optical Character Recognition (OCR) with smart algorithms. First, paper scripts get turned into digital format; next, OCR pulls out the written text, which the system then assesses by measuring how closely it matches reference answers – using both meaning-based comparisons and keyword overlap. Instead of just spotting identical phrases, it grasps ideas and context, leading to smarter, steadier scoring. For added clarity and justice, lecturers step in afterward to examine and approve AI- assigned grades before they’re locked in. The system’s rollout sped up assessments, made outcomes more consistent, yet boosted dependability, cutting down teachers’ hands-on tasks by a large margin. Overall, this method using AI offers a straightforward, flexible way to grade exams by itself – giving results that feel fair and solid across any school.

This research shows how an AI-powered exam grading tool could make marking student papers faster and easier in schools. It uses OCR tech to read handwritten answers, while artificial intelligence checks each response for meaning and depth. By combining these two methods, the system tackles common issues in hand-marking – like how long it takes, tired markers, uneven scores, and personal bias. Automating grading cuts down busywork for teachers, so they can spend time boosting learning and connecting with students instead of getting stuck in routine paperwork. Thanks to AI-driven meaning checks, student responses aren’t just scored for keyword matches – ideas and context matter just as much. Because of this, results become more reliable and balanced, especially on open-ended questions where different people might otherwise grade differently. A Human Verification step adds extra strength by keeping things fair and clear, even when automations involved. Mixing smart tech with real human judgment helps teachers and learners feel more confident, while also holding the system responsible for its choices. Thanks to this setup, decisions stay transparent and can be tweaked when needed – ideal for use across big schools or universities. Down the road, this setup might get even better with a few tweaks. Coming updates may add OCR that handles various languages, smarter models learning from teacher input to boost precision gradually – also linking up with school databases or platforms like LMS so grading flows without hiccups. In short, this AI-based grading system offers a flexible, clear, plus smart fix for old-school test scoring problems. Not just faster or more efficient, it also boosts fairness, consistency, along with fresh approaches in how students are assessed – pushing education further into the digital age.

[1]A. Jain and R. Singh, “AI-driven tools for automatic grading of written responses,” International Journal of Computer Applications, vol. 173, no. 3, pp. 24–29, 2020. [2]S. Arora and R. Chauhan, “NLP and ML methods used to check answers automatically,” Journal of Educational Technology, vol. 10, no. 2, pp. 45–56, 2020. [3]S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, 4th ed. Pearson, 2021. [4]I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, 2016. [5]D. Jurafsky and J. H. Martin, Speech and Language Processing, Pearson, 2023. [6]IGI Global, “Use of AI tools for grading and feedback in higher education,” IGI Global Publications, 2023. [7]M. A. Rahaman and H. Mahmud, “Automated grading of handwritten answers through deep learning methods,” Transactions on Engineering and Computing Sciences, vol. 10, no. 4, 2022. [8]S. P. J. Sijimol and S. M. Varghese, “Handwritten Short Answer Evaluation System (HSAES),” International Journal of Scientific Research and Science & Technology (IJSRST), Dept. of CSE, Mar Athanasius College of Engineering, Kerala, 2023. [9]P. M. S. Prerana, S. M. Chavan, R. Bathula, S. Saikumar, and G. Dayalan, “Eval – Automatic grading of answer sheets through deep learning and natural language processing,” International Journal of Intelligent Systems and Applications in Engineering, PES University, Bengaluru, India, n.d. [10]G. P. Tamminedi, S. A. Maganti, and T. Chandra, “Automation of answer script evaluation using AI,” International Journal of Innovative Science and Research Technology, vol. 9, no. 10, Oct. 2024. [11]S. K. Kumari, S. R. B., S. Subedar, S. S. Gupta, and C. N. Sharif, “AI-Driven OCR-Based Script Grading,” International Journal for Research in Applied Science & Engineering Technology (IJRASET), vol. 13, May 2025. [12]H. T. Nguyen, C. T. Nguyen, H. Oka, T. Ishioka, and M. Nakagawa, “Handwriting recognition and automated grading for written responses in Japanese exams,” arXiv preprint, 2022. [13]J. Cvengros and G. Kortemeyer, “Assisting the grading of a handwritten general chemistry exam with artificial intelligence,” arXiv preprint, Sept. 2025. [14]R. Thakur, S. Kaushik, G. Chopra, and H. Rohilla, “TrueGradeAI: A retrieval-augmented system for unbiased digital grading,” arXiv preprint, Sept. 2025. [15]L. I. Sakri, D. N. Kori, S. S. Joshi, A. S. Nayak, and S. Marigoudar, “AI-enabled transformation of online learning through personalization,” Journal of Engineering Education Transformations, vol. 39, no. 1, pp. 251–231, 2025.

Call for Paper (Hub) Fast Paper Publication IJCSE Indexing & Databases Submit Manuscript Call for Paper: Sep–Oct 2025 Call for Paper: Publish Research