Author(s)

Haixia Lv

Corresponding Author

Haixia Lv

Abstract

Campus loan refers to the installment payment or loan obtained by college students through the online loan platform, which solves the problem of shortage of funds for some students. However, due to the non-standard development of "campus loan" and its increasingly hidden nature, malig-nant events still occur frequently, which has a very bad impact on the school and society. From two aspects of soft and hard information on college students' campus loans to investigate analysis, firstly the questionnaire survey was distributed to 2735 college students randomly selected in Shan Dong provincial universities as soft information, and analyze the campus loan according to the results of the study some relevant situations of existence and development of colleges and universities in ShanDong province. Then from the point of view of hard information, this article takes the UCI default of credit card client data sets, using histogram based decision tree algorithm LGBM(Light Gradient Boosting Machine) algorithm of decision tree, training and making prediction of data set. Finally, through the analysis of the essence of campus loan from two aspects of hard and soft in-formation, and from the perspective of the synergistic effect of government, university, enterprise and family, puts forward countermeasures to prevent the risk of campus loan.

Keywords

Campus Loan, Soft and Hard Information, Lgbm, Binary Classification, Analysis and Prediction