Text Evaluation Using AI

Evaluation of descriptive answers is often time-consuming and subjective when performed manually, leading to inconsistencies in grading.To address this challenge, the proposed Text Evaluation Using AI system automates the assessment process in a fair and human-like manner.The system utilizes Natural Language Processing (NLP) to analyze semantic similarity between teacher reference answers and student responses based on meaning rather than exact keyword matching.A fuzzy inference system is integrated to manage uncertainty and partial correctness in descriptive answers, while key concept detection and answer length analysis are used to evaluate the completeness and depth of responses.Teacher and student answers are processed in DOCX format on a paragraph-wise basis, and final marks and grades are generated through a user-friendly interface.Experimental results demonstrate improved consistency, reduced evaluation time, and minimized subjectivity, indicating that the proposed system is an effective solution for automated academic assessment with strong potential for use in educational institutions and online evaluation platforms.