Generalized Sequential Pattern Mining of Undergraduate Courses
Daniel Leeds, Cody Chen, Yijun Zhao, Fiza Metla, James D. Guest, Gary M. Weiss · Zenodo (CERN European Organization for Nuclear Research) · 2022
University students have a great deal of freedom in deciding the order in which to take their courses. In this paper we apply the Apriori-based Generalized Sequential Pattern (GSP) algorithm to undergraduate course data from a large university in order to identify frequent course sequences. Course sequencing results are primarily generated at the department level, with a special focus on Computer Science courses. This paper also introduces the course sequence flow diagram, which compactly represents a large amount of course sequencing information in an intuitive visual form. Our results and associated flow diagrams can help to answer a variety of important questions, such as: what course sequences are most common, how are courses between different departments ordered, and when are courses taken in an order that may contradict the ad-vice given by academic advisors? In this paper we show that this form of descriptive data mining can identify standard core curriculum and pre-health sequences of study, as well as computer science courses that are either artificially pushed to the end of a student's program of study or taken earlier than would be recommended.