College Academic Achievement Early Warning Prediction Based on Decision Tree Model

With the transformation of higher education from elitism to popularity, the problem of academically at-risk students has aroused widespread concern in universities and society. Many colleges and universities at home and abroad have implemented the academic achievement early warning system for academically at-risk students, that is, to intervene in the crisis of learning problems and academic difficulties that have occurred or are about to occur during the school period, and take targeted preventive and remedial measures to help solve the academic difficulties. In this paper, based on the education big data, we model the academic achievement early warning prediction as multi-class classification problem and try to give academic achievement warning to students as early as possible. The data pre-precessing and feature engineer is introduced, the decision tree model is adopted to do the academic achievement warning prediction. The prediction model provides decision-making assistance to educational and teaching managers, and provides early warning to students for giving timely protection and intervention to warn them.