PAPER EVALUATION USING ARTIFICIAL INTELIGENCE
DOI:
https://doi.org/10.62643/Keywords:
Artificial Intelligence, OCR, Image Processing, KNN Algorithm, Automated Evaluation, MCQ Evaluation, Subjective Answer Analysis, Machine LearningAbstract
The evaluation of student answer sheets is a time-consuming and labor-intensive process, especially in large-scale examinations. This project presents an automated system for evaluating both multiple-choice questions (MCQ) and subjective answer sheets using Artificial Intelligence (AI), Optical Character Recognition (OCR), and image processing techniques. The system allows users to upload scanned images of answer sheets, which are then processed to extract textual and visual information. For MCQ evaluation, image processing techniques are applied to detect marked answers, which are then compared with the correct answers to calculate the score. Correct responses are highlighted using visual indicators, improving result transparency. For subjective answer evaluation, the system utilizes OCR to convert handwritten or printed text from scanned images into machinereadable format. The extracted answers are then compared with the predefined correct answers using a machine learning algorithm, specifically the K-Nearest Neighbors (KNN) algorithm, to determine similarity and assign scores. This approach enables automated and consistent grading, reducing human bias and errors. The system is implemented as a webbased application with modules for user registration, login, MCQ evaluation, and subjective answer evaluation. Experimental results demonstrate that the system can achieve high accuracy in evaluating both objective and subjective answers, with reliable scoring outputs. However, performance may vary depending on image quality and handwriting clarity. Overall, the project highlights the potential of integrating AI, OCR, and image processing to automate academic evaluation processes, making them faster, more efficient, and scalable
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.













