Spatial data mining and university courses marketing
Hong Tang, Simon P. McDonald · Charles Sturt University Research Output (CRO) · 2001
Movement of university admission is not random. Student admission data can be used to define the likely source of students and movement of the source. Thus, it can help the university improve its courses marketing strategy. Standard database and statistical methods do not work well with interrelated spatial data. The ongoing research presented in this paper attempts to use Geographic Information System (GIS), spatial statistical techniques, and spatial data mining to explore the relationships between the source of students (such as area), its spatial components (such as connectivity and distance), and the attribute data (such as family income and education backgrounds). Multiple level spatial classifications and association rules are adopted to identify the patterns and to predict the trend of incoming student source. The preliminary results of this research confirm that sources of students tend to be located in the particular spatial areas and with certain geographical/non-geographical settings. This paper also discusses the limitations of the adopted approach and the directions for future research.