期刊名称：ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
印刷版ISSN：2194-9042
电子版ISSN：2194-9050
出版年度：2002
卷号：XXXIV Part 4
出版社：Copernicus Publications
摘要：Student populations tend to be located in particular spatial areas and within specific demographic settings. Spatial and non- spatial data such as the university's admission records, census data, transport network data, and university campus location data can be used to describe, explain, and predict university student admission patterns, thus provide information to assist the university with improving its strategy in course marketing. Standard database and statistical methods do not work well with interrelated spatial data. Our study attempts to use Geographic Information System (GIS), spatial statistics, and spatial data mining techniques to explore the associations between the students of a specified course and their demographic characteristics (such as accessibility and proximity to university campus, ethnic background, and socio-economic status). A method of integrating experts' knowledge for mining multi-level association rules iteratively between spatial and non- spatial data is proposed to identify the pattern and predict the spatial trend of student admission, and hence, the potential market areas of students. Historical data are used to evaluate the results and validate the method. This paper also discusses the limitations of the adopted approach and the directions for future study
关键词：Geographic Information System (GIS); university admission; spatial data mining; spatial statistics; potential ; market areas
