A Study of Applying Data Mining Techniques on the Drop out Prediction Model- Using a Northern University of Taiwan as Example

碩士 === 銘傳大學 === 資訊管理學系碩士在職專班 === 100 === According to the latest statistics from the Ministry of Education, one hundred and ten thousand students dropped out from colleges in 2010. The rate is increasing from 6.5% to 14.1%, about 2 times higher than in 2002. The universities are in face of the pro...

Full description

Bibliographic Details
Main Authors: Cai-Ling Jian, 簡采羚
Other Authors: Yung-Sun Lee
Format: Others
Language:zh-TW
Published: 2012
Online Access:http://ndltd.ncl.edu.tw/handle/96758646560177656037
Description
Summary:碩士 === 銘傳大學 === 資訊管理學系碩士在職專班 === 100 === According to the latest statistics from the Ministry of Education, one hundred and ten thousand students dropped out from colleges in 2010. The rate is increasing from 6.5% to 14.1%, about 2 times higher than in 2002. The universities are in face of the problem that students refuse to come, and refuse to study as they come. The main purpose of this study was to explore the critical factors that affect student attrition from university. Through the related literature review, we inducted conduct accomplishments, academic achievement, attendance rate, entrance test results, and student loan as the independent variables to analysis the characteristics of drop out students in order to identify the students at-risk of dropping out as early as possible and facilitate a counseling intervention procedure to prevent student attrition. The samples of data are students of a Northern University between 2007 and 2010. We utilized decision tree and association rules to conduct data mining with different admission channels. The results showed that the attrition factors are different from admission channels. The common main influence factors are “semester grade” and “conduct accomplishments”. The universities should facilitate a counsel intervention procedure for different admission students to prevent student attrition.