A Study of Applying Data Mining Techniques on the Influence of Entrance Paths and Student's Background on Student's Academic Performance

• Main Authors: Da-Lin Chan, 詹德麟
• Other Authors: Ching-Chih Tseng
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/17410311911394405855

碩士 === 大葉大學 === 事業經營研究所碩士在職專班 === 96 === In the past years, the ways of recruiting new students of university are only based on past experiences, on which the maximal utility cannot be reached. Hence, it is a very important issue for how to recruit the most suitably new students subject to the limited quota of multi-channel entrance program. In recent years, data mining techniques have been applied to many domains and there are a lot of successful applications in various industries. However, the work of application on education industry is rare. For an ever- and stably- increasing quantity and completely holding of student data, it will be useful for developing the strategies of recruiting new students and other relative activities if the data can be effectively analyzed. In order to understand whether the learning per-formances of the enrolled students in Dayeh University are related to their entrance channels, so we can develop the strategies of recruiting new students. Therefore, the objectives of this study are two. The first is, by applying the technique of data mining, we try to explore all potential knowledge such as the possible reference table of per-formances among different channel of entrance students and the relationship of the learning performances of the enrolled students in Dayeh University among different en-trance channels. Based on these analyzed results, are developed. The second is to for-mulate a mathematical programming model to find the best ratios of the different chan-nel of entrance students and the strategies of recruiting new students.