Selecting Potential Medical Professional Ability Students in Chinese NCEE by Predicting GPA through Data Mining

Abstract

The “Trinity” Comprehensive Evaluation Enrollment (TCEE) of colleges and universities is one enrollment reform of Chinese National College Entrance Examination (NCEE) since 2012. But how to select students with professional qualifications by predicting academic performance after admission is a problem requiring an urgent solution. This paper analyzed the academic achievements and registration datasets of students admitted by Wenzhou Medical University via TCEE and established a classifier by using a decision tree, support vector machine and Bayesian network. A nested integrated studying method was then adopted and accuracy, precision, recall, and specificity indicators were used to evaluate model performance. Model performance of the bagged classifier for nested decision tree was found the most suitable. Findings are conducive to optimizing the selection scheme of the TCEE and improving managerial decision making.