Abstract. This paper analyzed the quality of teaching using the Student Evaluation of Teaching (SET). The Valence Aware Dictionary and Sentiment Reasoner (VADER) was utilized to assess textual comments, providing a comprehensive view of teaching effectiveness beyond numerical ratings. The objectives were to identify faculty strengths and areas for improvement based on student feedback, analyze sentiment toward teaching methods, and determine the optimal number of clusters within the dataset. The analysis included 28,222 student comments from three semesters, preprocessed through tokenization, stopword removal, part-of-speech tagging, and lemmatization. A word cloud visualized common terms, while K-means clustering and the Elbow method identified five as the optimal number of clusters. Results indicate that most comments are positive, emphasizing effective teaching methods' role in creating a positive educational experience. The findings suggest integrating machine learning with VADER and expanding the dataset for broader insights. Institutions should develop effective teaching strategies, prioritizing regular feedback collection and analysis.

Keywords: Teaching effectiveness; Student Evaluation of Teaching (SET); Sentiment analysis; Feedback collection.