International Journal of Innovative Research in Computer and Communication Engineering

ISSN Approved Journal | Impact factor: 8.771 | ESTD: 2013 | Follows UGC CARE Journal Norms and Guidelines

| Monthly, Peer-Reviewed, Refereed, Scholarly, Multidisciplinary and Open Access Journal | High Impact Factor 8.771 (Calculated by Google Scholar and Semantic Scholar | AI-Powered Research Tool | Indexing in all Major Database & Metadata, Citation Generator | Digital Object Identifier (DOI) |


TITLE AI-Based Handwritten Answer Sheet Evaluation Using OCR and NLP Semantic Analysis
ABSTRACT Checking answer sheets that are written by hand is a slow and exhausting process, and because of evaluator fatigue and the need to make subjective judgments, it usually leads to inconsistent results—this is indeed a serious issue for organizations which have to evaluate a large number of students. This paper presents an AI-based system which automates the grading of descriptive answers by combining Optical Character Recognition (OCR) with a semantic analysis carried out using Natural Language Processing (NLP). The handwritten responses are first processed by a multimodal large language model, then improved using traditional NLP methods, and then compared to the correct answers using Sentence-BERT (SBERT) embeddings and cosine similarity. The similarity scores obtained in this way are converted into grades according to a grading rubric, and a full and detailed report is produced to ensure complete transparency. The system is based on the most recent advances in OCR, transformer-based similarity models, layout analysis, and LLM-assisted grading, and is designed to speed up the assessment process without missing what the student actually understood. The paper also examines open questions such as differences in handwriting, diagrams, mathematical notation, and validation against human graders, and identifies areas that need further research.
AUTHOR PROF. BASAVARAJA PATIL G V, CHAITRA B S, CHANDANA S, CHANDRALEKHA M S, HARSHITA G K Assistant Professor, Dept. of CSE, Jain Institute of Technology, Davangere, Karnataka, India UG Students, Dept. of CSE, Jain Institute of Technology, Davangere, Karnataka, India
VOLUME 189
DOI DOI: 10.15680/IJIRCCE.2026.1410051
PDF pdf/51_AI-Based Handwritten Answer Sheet Evaluation Using OCR and NLP Semantic Analysis.pdf
KEYWORDS
References [1] A. Kulkarni, R. Patil, S. Endait, G. Kale, and R. Ghatage, "Automated Answer and Diagram Scoring in the STEM Domain: A Literature Review," in Proc. 2024 5th Int. Conf. Emerging Technology (INCET), Karnataka, India, May 2024, doi: 10.1109/INCET61516.2024.10593060.
[2] B. Aarathisree, S. Sarkar, B. D. Dammala, M. Panday, and N. Sharma, "Revolutionizing the Future of Automated Subjective Answer Sheet Evaluation System with Machine Learning and LLMs," in Proc. 2024 IEEE Pune Section Int. Conf. (PuneCon), Pune, India, Dec. 2024, doi: 10.1109/PuneCon63413.2024.10895748.
[3] Palak, Mohit, N. Sharma, Bhushan, and S. Jain, "Python-based OMR Sheet Recognition and Evaluation," in Proc. Int. Conf. Modern Sustainable Systems (CMSS), 2025, doi: 10.1109/CMSS66566.2025.11182432.
[4] N. A. F. Fansab, S. A. I. Alfarozi, and I. Ardiyanto, "Improvement of Small Object Detection of Handwritten Answer Sheet," in Proc. 2025 Int. Conf. Advancement in Data Science, E-learning and Information System (ICADEIS), 2025, doi: 10.1109/ICADEIS65852.2025.10933354.
[5] S. Sarkar, R. Jayadevan, N. Sharma, and M. Pandey, "Enhancing Handwritten Answer Sheet Evaluation Through Large Language Models and Multi-Agent Reasoning," in Proc. 2025 IEEE Pune Section Int. Conf. (PuneCon), Pune, India, Dec. 2025, doi: 10.1109/PUNECON67554.2025.11379082.


Copyright © IJIRCCE 2020.All right reserved