Towards an improved of teaching practice using Sentiment Analysis in Student Evaluation

Student Evaluation of Teaching (SET) serves as an ad hoc way of assessing teaching effectiveness within higher education institutions. This paper introduces an approach to analyzing sentiments expressed in SET comments using a Large Language Model (LLM). By employing natural language processing techniques, the polarity conveyed by students upon course completion is extracted and analyzed, aiming to furnish educators and stakeholders with valuable insights into teaching quality and areas for improvement in teaching practice. This study showcases the effectiveness of LLMs in sentiment analysis of comments, underscoring their potential to enhance the evaluation process. The development of a prototype tool, collaborative labeling of end -of -course evaluations, and a comparison with LLM-based labeling are experimentally explored. Subsequently, the implications for educational institutions are discussed, and future research directions in this domain are proposed.