基于CPSO-XGboost的个人信用评估
|
摘要
在大数据时代的背景下,个人信用数据指标日益繁杂。为降低个人信用数据冗余性,使用基于随机森林与梯度提升决策树组合的特征选取方法;提出混沌粒子群算法优化XGboost信用评估模型参数,提高个人信用评估准确性。实例分析结果表明,CPSO-XGboost相比XGboost、Logistic和SVM在个人信用评估中具有更高的稳定性和准确性。
In the context of the era of big data,since the personal credit data indicators are increasingly complex,actions need to be taken to improve the performance of the personal credit assessment.Feature selection methods based on combination of random forests and gradient boosting decision tree were used to reduce the redundancy of personal credit data.Chaos particle swarm optimization algorithm was proposed to optimize the XGboost credit assessment model parameters,which improved the accuracy of personal credit assessment.Example analysis shows that,compared with XGboost,Logistic,and SVM,CPSO-XGboost has higher stability and accuracy in personal credit assessment.
In the context of the era of big data,since the personal credit data indicators are increasingly complex,actions need to be taken to improve the performance of the personal credit assessment.Feature selection methods based on combination of random forests and gradient boosting decision tree were used to reduce the redundancy of personal credit data.Chaos particle swarm optimization algorithm was proposed to optimize the XGboost credit assessment model parameters,which improved the accuracy of personal credit assessment.Example analysis shows that,compared with XGboost,Logistic,and SVM,CPSO-XGboost has higher stability and accuracy in personal credit assessment.
引文
[1]SHI Xiaokang,MA Xuejun.The index selection method of personal credit rating model[J].Statistics and Decision,2014(23):41-43(in Chinese).[史小康,马学俊.个人信用评级模型的指标选择方法[J].统计与决策,2014(23):41-43.]
[2]Hoque N,Bhattacharyya DK,Kalita JK.MIFS-ND:A mutual information-based feature selection method[J].Expert Systems with Applications,2014,41(14):6371-6385.
[3]Li J,Cheng K,Wang S,et al.Feature selection:A data perspective[J].ACM Computing Surveys(CSUR),2017,50(6):1-45.
[4]Fernandes GB,Artes R.Spatial dependence in credit risk and its improvement in credit scoring[J].European Journal of Operational Research,2016,249(2):517-524.
[5]Blanco A,Pino-Mejías R,Lara J,et al.Credit scoring models for the microfinance industry using neural networks:Evidence from Peru[J].Expert Systems with Applications,2013,40(1):356-364.
[6]Harris T.Credit scoring using the clustered support vector machine[J].Expert Systems with Applications,2015,42(2):741-750.
[7]Zhu Y,Xie C,Wang GJ,et al.Comparison of individual,ensemble and integrated ensemble machine learning methods to predict China’s SME credit risk in supply chain finance[J].Neural Computing&Applications,2017,28(1):41-50.
[8]Deng H,Runger G.Gene selection with guided regularized random forest[J].Pattern Recognition,2013,46(12):3483-3489.
[9]Si S,Zhang H,Keerthi SS,et al.Gradient boosted decision trees for high dimensional sparse output[C]//International Conference on Machine Learning,2017:3182-3190.
[10]ZHANG Renpeng,YANG Jinxiao,PAN Jiahua,et al.Three-dimensional route planning of UAV based on improved particle swarm algorithm[J].Computer Simulation,2014,31(3):65-69(in Chinese).[张仁鹏,杨金孝,潘佳华,等.基于改进粒子群算法的无人机三维航迹规划[J].计算机仿真,2014,31(3):65-69.]
[11]YANG Xiyun,GUAN Wenyuan,LIU Yuqi,et al.Prediction intervals forecasts of wind power based on PSO-KELM[J].Proceedings of the CSEE,2015,35(S1):146-153(in Chinese).[杨锡运,关文渊,刘玉奇,等.基于粒子群优化的核极限学习机模型的风电功率区间预测方法[J].中国电机工程学报,2015,35(S1):146-153.]
[12]Li H,Zhu H,Ma D.Demographic information inference through meta-data analysis of Wi-Fi traffic[J].IEEE Transactions on Mobile Computing,2017,17(5):1033-1047.
[13]Chen T,Guestrin C.Xgboost:A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,2016:785-794.
[14]Xia Y,Liu C,Li YY,et al.A boosted decision tree approach using Bayesian hyper-parameter optimization for credit scoring[J].Expert Systems with Applications,2017,78(78):225-241.
[15]I-Cheng Yeh.Default of credit card clients data set[EB/OL].[2016-01-26].http://archive.ics.uci.edu/ml/datasets/default+of+credit+card+clients.
[2]Hoque N,Bhattacharyya DK,Kalita JK.MIFS-ND:A mutual information-based feature selection method[J].Expert Systems with Applications,2014,41(14):6371-6385.
[3]Li J,Cheng K,Wang S,et al.Feature selection:A data perspective[J].ACM Computing Surveys(CSUR),2017,50(6):1-45.
[4]Fernandes GB,Artes R.Spatial dependence in credit risk and its improvement in credit scoring[J].European Journal of Operational Research,2016,249(2):517-524.
[5]Blanco A,Pino-Mejías R,Lara J,et al.Credit scoring models for the microfinance industry using neural networks:Evidence from Peru[J].Expert Systems with Applications,2013,40(1):356-364.
[6]Harris T.Credit scoring using the clustered support vector machine[J].Expert Systems with Applications,2015,42(2):741-750.
[7]Zhu Y,Xie C,Wang GJ,et al.Comparison of individual,ensemble and integrated ensemble machine learning methods to predict China’s SME credit risk in supply chain finance[J].Neural Computing&Applications,2017,28(1):41-50.
[8]Deng H,Runger G.Gene selection with guided regularized random forest[J].Pattern Recognition,2013,46(12):3483-3489.
[9]Si S,Zhang H,Keerthi SS,et al.Gradient boosted decision trees for high dimensional sparse output[C]//International Conference on Machine Learning,2017:3182-3190.
[10]ZHANG Renpeng,YANG Jinxiao,PAN Jiahua,et al.Three-dimensional route planning of UAV based on improved particle swarm algorithm[J].Computer Simulation,2014,31(3):65-69(in Chinese).[张仁鹏,杨金孝,潘佳华,等.基于改进粒子群算法的无人机三维航迹规划[J].计算机仿真,2014,31(3):65-69.]
[11]YANG Xiyun,GUAN Wenyuan,LIU Yuqi,et al.Prediction intervals forecasts of wind power based on PSO-KELM[J].Proceedings of the CSEE,2015,35(S1):146-153(in Chinese).[杨锡运,关文渊,刘玉奇,等.基于粒子群优化的核极限学习机模型的风电功率区间预测方法[J].中国电机工程学报,2015,35(S1):146-153.]
[12]Li H,Zhu H,Ma D.Demographic information inference through meta-data analysis of Wi-Fi traffic[J].IEEE Transactions on Mobile Computing,2017,17(5):1033-1047.
[13]Chen T,Guestrin C.Xgboost:A scalable tree boosting system[C]//Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,2016:785-794.
[14]Xia Y,Liu C,Li YY,et al.A boosted decision tree approach using Bayesian hyper-parameter optimization for credit scoring[J].Expert Systems with Applications,2017,78(78):225-241.
[15]I-Cheng Yeh.Default of credit card clients data set[EB/OL].[2016-01-26].http://archive.ics.uci.edu/ml/datasets/default+of+credit+card+clients.