摘要
为了解决互联网金融行业面临的风险压缩与风险控制问题,本文研究了以互联网金融大数据为基础的风控模型,同时为互联网金融反欺诈问题提供了新型的风控方案。提出的风控模型融合了机器学习技术,主要包括异常用户检测和用户信用评分两部分。前一个部分的输出结果作为后一个部分的输入内容,从而得到最终的信用评分输出结果。
A study was presented in the paper about a risk model based on internet financial big data and a new risk control program against online financial fraudulence to deal with the issues of risk compression and control threatening the internet financial industry. The model,incorporating the technology of machine learning,comprises 2 major parts,i. e. abnormal user detection and user credit rating. The output of the former serves as the input for the later to obtain the final credit rating.
引文
[1]马云.金融行业需要搅局者[J].市场观察,2013(7):11-11.
[2]王达.美国互联网金融的发展及中美互联网金融的比较---基于网络经济学视角的研究与思考[J].国际金融研究,2014,332(12):47-57.
[3]王越,曹长修.KDD方法在金融欺诈检测中的应用研究[J].计算机工程与设计,2002,23(5):43-45.
[4]马杰.大数据征信应用于互联网金融风控研究[D].北京:对外经济贸易大学,2015.
[5]Chen J,Tao Y,Wang H,et al.Big data based fraud risk management at Alibaba[J].Journal of Finance&Data Science,2015,1(1):1-10.
[6]Chandola V,Banerjee A,Kumar V.Anomaly detection:Asurvey[J].Acm Computing Surveys,2009,41(3):1-58.
[7]Anscombe F J,Guttman I.Rejection of Outliers[J].Technometrics,1960,2(2):123-146.
[8]Eskin E.Anomaly Detection over Noisy Data using Learned Probability Distributions[C].International Conference on Machine Learning.2000,255-262.
[9]Desforges M J,Jacob P J,Cooper J E.Applications of probability density estimation to the detection of abnormal conditions in engineering[J].Proceedings of the Institution of Mechanical Engineers-Part C,1998,212(8):687-703.