Application of Data Mining in Predicting College Graduates Employment

With the popularization and enrollment expansion of higher education in China, the employment of college students has been the focus of public attentions. In this study, taking the graduates of Grade 2016 in Guilin University of Technology as example, data mining techniques are applied in predicting employment by using five influencing factors. Based on the CART algorithm, the Gini indexes were calculated and decision tree was constructed. Furthermore, the random forest algorithm is used to improve accuracy rate of employment prediction. After data collection, cleaning and conversion, 496 employment records were obtained, 70% of them were taken as the training samples. The constructed model was tested by the remaining samples and the accuracy reaches 81%. Finally, the features academic achievement and graduation qualification are identified as the significant factors of students' employability. The combination model of decision tree and random forest provides a new method for employment forecasting, which is feasible and adaptable to employment guidance of schools.