Enhancing Student Feedback Analysis: NLP-Based Machine Learning Classification of Satisfaction Levels at Universitas Terbuka Surakarta

Dwianika, Agustine
Paramitasari, Ratih

JURISMA: Jurnal Riset Bisnis & Manajemen
Vol. 15 No. 2: Oktober 2025

This research aims to classify the level of student satisfaction at Universitas Terbuka (UT) Surakarta using Natural Language Processing (NLP) techniques and machine learning algorithms. The study utilizes textual responses from student satisfaction surveys and processes them through a supervised classification approach. By applying methods such as TF-IDF for feature extraction and classification algorithms like Naïve Bayes, Support Vector Machine (SVM), and Random Forest, the research seeks to identify which algorithm best categorizes sentiment into satisfaction levels. Results indicate that the SVM model outperforms other algorithms in accuracy, precision, and F1-score. This approach demonstrates the practical application of NLP in higher education quality assurance and offers valuable insights for policymakers. Keywords: Machine Learning; Natural Language Processing; Student Satisfaction; Classification; Sentiment Analysis