Abstract
Scheduling courses ("timetabling") at a University is a persistent challenge. Allocating course-sections to prescribed "time slots" for courses requires advanced quantitative techniques, such as goal programming, and collecting a large amount of multi-criteria data at least six to eight months in advance of a semester. This study takes an alternate approach. It demonstrates the feasibility of applying the principles of data mining. Specifically it uses association rules to evaluate a non-standard ("aberrant") timetabling pilot study undertaken in one College at a University. The results indicate that 1), inductive methods are indeed applicable, 2), both summary and detailed results can be understood by key decision-makers, and 3), straightforward, repeatable SQL queries can be used as the chief analytical technique on a recurring basis. In addition, this study was one of the first empirical studies to provide an accurate measure of the discernable, but negligible, scheduling exclusionary effects that may impact course availability and diversity negatively.
DOI
10.17705/1CAIS.01623
Recommended Citation
Smith, W. (2005). Applying Data Mining to Scheduling Courses at a University. Communications of the Association for Information Systems, 16, pp-pp. https://doi.org/10.17705/1CAIS.01623
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
