AI-DRIVEN AUTOMATED ACADEMIC EVOLUTION USING NLP

Abstract

The rapid growth of digital education platforms has created a strong need for automated systems that can efficiently evaluate student responses while maintaining accuracy and fairness. The proposed Automated Academic Assessment Using Natural Language Processing and Artificial Intelligence system is designed to automatically evaluate descriptive answers submitted by students through a web-based platform. The system integrates Natural Language Processing (NLP) techniques and similarity analysis to compare student responses with predefined model answers stored in a database.The architecture of the system is implemented using the Flask web framework, where different modules handle authentication, student interaction, administrative management, and automated evaluation. Students can register, log in, and submit answers to questions through the web interface. Administrators can add and manage questions and reference answers through the admin panel. All submitted answers are stored in a structured database for further analysis.During the evaluation phase, the system preprocesses both the student answers and the reference answers using NLP techniques such as text cleaning, tokenization, and normalization. A similarity computation module then compares the processed texts to determine the semantic similarity between the student's response and the expected answer. Based on the calculated similarity score, the system automatically assigns marks and generates feedback for the student using predefined grading and feedback functions.The proposed system significantly reduces the manual effort required for grading descriptive answers and ensures faster and more consistent evaluation. By integrating AI-driven text analysis and automated grading mechanisms, the system enhances the scalability and efficiency of academic assessments in modern educational environments.