Author(s)

Lei Yang, Liwei Tian, Liang Yu and Yungui Chen

Corresponding Author

Lei Yang

ABSTRACT

The registration rate of new students has always been a concern of all colleges, and it is a difficult problem to accurately predict the number of new students before they are registered. At present, there are no researchers using machine learning method to predict the registration of new students, because the intuitive feeling is that whether students register or not, which is a very subjective thing, affected by many subjective factors. At present, the traditional methods are used to predict the number of new students in Colleges, that is, telephone inquiry and tuition payment status inquiry. According to the historical enrollment data of a university, this research uses the method of random forest to study it. The results show that whether the freshmen register or not can be predicted, and the enrollment data of universities over the years is valuable.

KEYWORDS

Prediction; enrollment; random forest