基于GBDT的个人信用评估方法

被引：73

作者：

王黎 ^{[1
,2
]}

廖闻剑 ^{[1
,2
]}

机构：

[1] 武汉邮电科学研究院

[2] 烽火通信科技股份有限公司南京研发部

来源：

电子设计工程 | 2017年 / 25卷 / 15期

关键词：

信用评估; 分类算法; GBDT;

D O I：

暂无

中图分类号：

F830.5 [信贷]; TP301.6 [算法理论];

学科分类号：

020219 [财政学（含：税收学）]; 080201 [机械制造及其自动化];

摘要：

近年来,个人信用评估问题成为信贷行业的研究热点,针对当前应用于信用评估的分类算法大多存在只对某种类型的信用数据集具有较好的分类效果的问题,提出了基于Gradient Boosted Decision Tree(GBDT)的个人信用评估方法。GBDT天然可处理混合数据类型的数据集,可以发现多种有区分性的特征以及特征组合,不需要做复杂的特征变换,对于特征类型复杂的信用数据集有明显的优势,且其通过其损失函数可以很好地处理异常点。在基于两个UCI公开信用审核数据集上的对比实验表明,GBDT明显优于传统常用的支持向量机(Support Vector Machine,SVM)以及逻辑回归(Logistic Regression,LR)的信用评估效果,具有较好的稳定性和普适性。

引用

页码：68 / 72

页数：5

共 10 条

[1]

Example-dependent cost-sensitive decision trees [J].

Bahnsen, Alejandro Correa ;

Aouada, Djamila ;

Ottersten, Bjoern .

EXPERT SYSTEMS WITH APPLICATIONS, 2015, 42 (19) :6609-6619

[2]

Enhancing accuracy and interpretability of ensemble strategies in credit risk assessment. A correlated-adjusted decision forest proposal [J].

Florez-Lopez, Raquel ;

Manuel Ramon-Jeronimo, Juan .

EXPERT SYSTEMS WITH APPLICATIONS, 2015, 42 (13) :5737-5753

[3]

Credit scoring models for the microfinance industry using neural networks: Evidence from Peru [J].

Blanco, Antonio ;

Pino-Mejias, Rafael ;

Lara, Juan ;

Rayo, Salvador .

EXPERT SYSTEMS WITH APPLICATIONS, 2013, 40 (01) :356-364

[4]

Credit scoring, risk, and consumer lendingscapes in emerging markets [J].

Burton, Dawn .

ENVIRONMENT AND PLANNING A-ECONOMY AND SPACE, 2012, 44 (01) :111-124

[5]

An experimental comparison of classification algorithms for imbalanced credit scoring data sets.[J].Iain Brown;Christophe Mues.Expert Systems With Applications.2011, 3

[6]

An introduction to ROC analysis [J].

Fawcett, Tom .

PATTERN RECOGNITION LETTERS, 2006, 27 (08) :861-874

[7]

Credit scoring with a data mining approach based on support vector machines.[J].Cheng-Lung Huang;Mu-Chen Chen;Chieh-Jen Wang.Expert Systems With Applications.2006, 4

[8]

Credit scoring using the hybrid neural discriminant technique [J].

Lee, TS ;

Chiu, CC ;

Lu, CJ ;

Chen, IF .

EXPERT SYSTEMS WITH APPLICATIONS, 2002, 23 (03) :245-254

[9]

Greedy function approximation: A gradient boosting machine [J].

Friedman, JH .

ANNALS OF STATISTICS, 2001, 29 (05) :1189-1232

[10]

Neural network credit scoring models.[J].David West.Computers and Operations Research.2000, 11

← 1 →