Student Feedback Analysis Using Machine Learning and Natural Language Processing: A Real-Time Web-Based System
Anupam Baidya, Subhrangsu Chandra, Ansuman Mahanty, Sukarna Dey Mondal, Sanjay Sengupta, Pabitra Kumar Dey · 2026
Student feedback is a critical component for enhancing teaching quality, refining curriculum design, and supporting institutional improvement in modern education. However, manually analyzing large volumes of feedback is time-consuming and often influenced by subjective interpretation. The objective of this research is to design and implement an automated sentiment analysis system that classifies student feedback as positive, negative, or neutral, thereby enabling educational institutions to make timely, data-driven decisions. The proposed solution is a web-based application that integrates NLP and ML techniques. HTML, CSS and JavaScript is used to design the front-end and Python-Flask is used as a back-end. The collected feedback was processed with the help of tokenisation, stemming, stop-word removal, and TF-IDF-based feature extraction techniques. Two classifiers, Support Vector Machine (SVM) and Multinomial Naive Bayes, were trained and evaluated on a labelled dataset. Performance metrics such as F1-score, recall, accuracy, and precision were used for evaluation. The results indicate that SVM outperformed the other models with an accuracy of 85%, establishing the system as a practical tool for real-time sentiment monitoring in the classroom.