AI-Based Handwritten Answer Evaluation System

DOI

https://www.doi.org/10.59256/ijsreat.20260605024

Published In

Volume 6, Issue 5 (2026)

Published Online

September-October 2026

Pages

210-213

Article Type

Original Article

Check for updates

The AI-Based Handwritten Answer Evaluation System is a web-based application designed to assist teachers in evaluating handwritten examination answers. The system provides a platform for teachers to create classes, prepare tests, add questions, and define reference answers. Students can upload handwritten answer sheets in image or PDF format for automated processing. An AI-enabled Optical Character Recognition (OCR) service is used to extract handwritten text from the uploaded documents. The extracted responses are organized question-wise and compared with teacher-provided reference answers using semantic similarity, keyword overlap, and fuzzy text matching techniques. Based on the implemented scoring logic, the system generates preliminary question-wise marks and calculates the overall score. Python and Flask support the web application, while MySQL and SQLAlchemy manage assessment records and evaluation results. Teachers can review the extracted responses and suggested marks, make corrections, and finalize the results. The system aims to support assessment workflows by reducing repetitive evaluation tasks and organizing student performance records. However, the reliability of the generated marks depends on handwriting recognition quality, reference-answer accuracy, and the scoring logic. Teacher supervision remains necessary before finalizing assessment results. Further testing using diverse handwriting samples and teacher-verified scores is required to evaluate the system's accuracy and reliability.