Recommending the right academic programs: An interest mining approach using BERTopic
Alessandro Hill, Kalen Goo, Puneet Agarwal
TL;DR
The paper tackles the problem of helping students choose university programs by introducing TopProRec, a BERTopic-based recommender that mines interest topics from course descriptions and backtracks on a knowledge map to rank programs. It leverages an embedding-based topic modeling pipeline (BERTopic) to generate $h$ interest topics with up to $\\gamma$ keywords, presenting them as word clouds for user feedback and using a backtracking algorithm to compute $PIS$, $R-PIS$, and $SCORE$ to output a top $\\tau$ list of programs. The authors validate the approach in a case study with 84 programs, 4,251 courses, and 19,283 keywords, reporting high program coverage ($\\rho$ around 0.98) and personalization ($SCORE$ and $0.77$), along with strong user acceptance (>$98\%$ alignment and ~$94\%$ willingness to reuse). Qualitative feedback highlights serendipity and trust, while the study also discusses explainability, controllability, fairness, robustness, privacy, and potential extensions to multi-university settings and course-level recommendations, underscoring practical impact for student support services.
Abstract
Prospective students face the challenging task of selecting a university program that will shape their academic and professional careers. For decision-makers and support services, it is often time-consuming and extremely difficult to match personal interests with suitable programs due to the vast and complex catalogue information available. This paper presents the first information system that provides students with efficient recommendations based on both program content and personal preferences. BERTopic, a powerful topic modeling algorithm, is used that leverages text embedding techniques to generate topic representations. It enables us to mine interest topics from all course descriptions, representing the full body of knowledge taught at the institution. Underpinned by the student's individual choice of topics, a shortlist of the most relevant programs is computed through statistical backtracking in the knowledge map, a novel characterization of the program-course relationship. This approach can be applied to a wide range of educational settings, including professional and vocational training. A case study at a post-secondary school with 80 programs and over 5,000 courses shows that the system provides immediate and effective decision support. The presented interest topics are meaningful, leading to positive effects such as serendipity, personalization, and fairness, as revealed by a qualitative study involving 65 students. Over 98% of users indicated that the recommendations aligned with their interests, and about 94% stated they would use the tool in the future. Quantitative analysis shows the system can be configured to ensure fairness, achieving 98% program coverage while maintaining a personalization score of 0.77. These findings suggest that this real-time, user-centered, data-driven system could improve the program selection process.