Smart Paper Assessment System

The increasing demand for rapid and accurate assessment in large-scale examinations has driven the development of automated grading systems. This paper introduces a Smart Paper Assessment System that leverages Optical Character Recognition (OCR), Natural Language Processing (NLP), and Large Language Models (LLMs) to convert physical exam papers into digital formats, analyze student responses, and autonomously assign scores. The system operates by first applying OCR to digitize handwritten or printed text on exam papers, enabling the identification and organization of student answers. NLP techniques then align these responses with their corresponding questions, while LLMs assess the content by comparing it to a predefined answer key. Using keyword matching, sentence structure, and semantic analysis, the model assigns marks based on the relevance and accuracy of each response. This approach accelerates grading, minimizes human errors and biases, and ensures consistent and fair evaluations. The scalability of the system allows it to manage exams of varying sizes, making it suitable for educational institutions ranging from schools to universities. The proposed Smart Paper Assessment System streamlines the grading process, enhances precision, and provides timely feedback, contributing to a more efficient educational assessment framework. keywords- Optical Character Recognition (OCR), Natural Language Processing (NLP), Large Language Models (LLM), Automated Grading, Paper-Based Assessment, Keyword Matching, Machine Learning in Education, Text Recognition.