JISARA

Abstract: In recent years, online professor review platforms have become increasingly prevalent in higher education. While previous studies have examined various aspects of these platforms, such as review sentiment and content validity, their potential as a source of information for academic success has been largely unexplored. This paper investigates the use of Large Language Models to analyze anonymous professor reviews and identify common themes related to effective teaching practices, course design, and student engagement. The goal is to provide students with actionable suggestions on how to succeed in specific courses rather than focusing on elements that do not directly impact educational outcomes. Our study analyzed reviews of nearly 40,000 computer science instructors, producing meaningful insights into course experiences. Although we realized our analysis or publicly available professor reviews the proposed methodology can be utilized in the context of official Student Evaluation of Teaching. We discuss how the proposed method can be utilized to process instructors’ reviews, highlight teaching strategies, and elicit actionable information for both students and educators. Also, we describe how the same approach could also be utilized to identify areas for potential improvement.

Download this article: JISARA - V18 N4 Page 20.pdf

Recommended Citation: Caporusso, N., Doan, M., Acharya, B., Pandit, P., Shrestha, S., Khatri, R., Pond, W., Le, N., (2025). From Angry Reviews to Classroom Success: Using LLMs to Synthesize RateMyProfessors.com Data. Journal of Information Systems Applied Research and Analytics 18(4) pp 20-35. https://doi.org/10.62273/JUTH9224