摘要
近年来,互联网金融在国内发展迅速,各种类型的小额网贷平台更是层出不穷,但是随着平台跑路,欺诈等各种问题的涌现,暴露出对小额网贷平台监管的不足。本文以C5.0决策树算法为核心建立起风险监控模型,首先利用大数据技术抓取平台数据、公开信息数据和征信机构数据,然后对数据分析,筛选关键指标体系,建立了一种风控模型,最后评估和验证模型及算法的有效性,并通过对模型的参数进行调整,提出以保证整体错误率的前提下,尽可能的降低α错误率的评价标准。通过实验发现,该风控模型对问题平台有着很好的预测能力。
Recently,Internet Finance has become more and more popular. However,its booming brings a series of risk control problems. Based on the C5.0 Decision Tree algorithm,a model to monitor the P2P platforms is proposed in this paper. It collected and analyzed the data including the platform data,the public enterprise data and the credit data so on. After that we studied a model including the key evaluation factors. And by adjusting and testing the parameters of the model,we made full use of C5.0 algorithm to predict and evaluate the P2P enterprise risks. Finally,we verified and proved that this model had a good ability of evaluating and predicting the platforms risks.
出处
《中国管理科学》
CSSCI
北大核心
2016年第S1期345-352,共8页
Chinese Journal of Management Science
基金
中央财经大学科研课题(020676116004)
