Criteria evaluation and selection in non-native language MBA students admission based on machine learning methods

摘要: Although the research on student selection criteria has been very rich up to now, role of level foreign language played in admission a non-native spoken program is still receiving little attention. This study intends explore issue through three methods: (1) two-sample test hypothesis; (2) multiple linear regression analysis; (3) machine learning algorithms (Ridge regression, SVM, Random forest, GBDT). The case about 549 students enrolled Shanghai International MBA Program China from 2007 2014 was used as empirical samples. Through methods analysis and comparison, it found that Oral English fluency key China. It confirmed criteria, such Rank graduated university, Company Nature, Latest Highest Degree, Math Exam, Sponsor (Tuition provider) Stress management, have good effect predicting final grades when graduation. also shows based algorithm modeling ridge SVM are suitable for decision modeling.