Unveiling Insights: A Machine Learning Approach to Decipher Student Sentiments in Computer University of Myanmar

This propose system addresses the crucial aspect of student feedback analysis in computer science courses at the Computer University of Myanmar. Acknowledging the significance of genuine feedback for educational improvement, the study integrates machine learning techniques and natural language processing for a comprehensive analysis. Three Google forms, tailored for curriculum, assessment, and learning outcomes, are utilized to collect both quantitative and qualitative data from students. The analysis pipeline involves preprocessing, text normalization, Bag of Words model construction, sentiment labeling, and TF-IDF feature extraction. Classification employs Support Vector Machine (SVM) and Naive Bayes classifiers, with varied strengths demonstrated across questionnaire forms. The results, presented the importance of selecting classifiers based on data context. SVM excels in positive class accuracy, particularly in assessment and learning outcomes, while Naive Bayes performs well in the curriculum. The study concludes by emphasizing the value of a combined SVM and Naive Bayes approach for a robust analysis and offers insights and methodologies for universities to enhance educational programs through meaningful feedback interpretation.