Modeling and Prediction of Graduate Students Enrollment Scale

Study the prediction problem of graduate student enrollment scales.Graduate student enrollment scale problem is affected by policy,political,economic,and social demand.These influencing factors can not be quantified and the relationships between them are complex.It is difficult for traditonal prediction models to predict the future numbers of students and the accuracy of prediction is quite low.In order to improve the prediction accuracy of graduate students enrollment scale,this paper proposed a prediction model based on support vector machine.According to the historical data of graduate student enrollment scale,influence factors were scattered and difficult to be certainly described.The graduate students enrollment scales of 1981～2004 in Heilongjiang Province were analyzed and the support vector machine prediction model was established.The enrollment scales from 2000 to 2004 were predicted with the model,and the prediction precision is quite high.The simulation results show that support vector machine provides a new research method for graduate students enrollment scale.