基于TF-IDF算法的P2P贷款违约预测模型

doi:10.11772/j.issn.1001-9081.2018030673

计算机应用 ›› 2018, Vol. 38 ›› Issue (10): 3042-3047.DOI: 10.11772/j.issn.1001-9081.2018030673

• 应用前沿、交叉与综合 • 上一篇下一篇

基于TF-IDF算法的P2P贷款违约预测模型

章宁¹, 陈钦^1,2

1. 中央财经大学信息学院, 北京 100081;
2. 国银金融租赁股份有限公司信息化管理部, 广东深圳 518038

收稿日期:2018-04-02 修回日期:2018-05-15 出版日期:2018-10-10 发布日期:2018-10-13
通讯作者: 陈钦
作者简介:章宁(1975-),女,江西临川人,教授,博士,主要研究方向:互联网金融、个人信息保护、服务外包;陈钦(1977-),男,江西南昌人,高级工程师,博士研究生,主要研究方向:金融科技、智能投资、大数据分析、信息检索。
基金资助:
国家重点研发计划项目（2017YFB1400701）。

P2P loan default prediction model based on TF-IDF algorithm

ZHANG Ning¹, CHEN Qin^1,2

1. School of Information, Central University of Finance and Economics, Beijing 100081, China;
2. IT Department, China Development Bank Financial Leasing Company Limited, Shenzhen Guangdong 518038, China

Received:2018-04-02 Revised:2018-05-15 Online:2018-10-10 Published:2018-10-13
Supported by:
This work is partially supported by the National Key Research and Development Program of China (2017YFB1400701).

摘要/Abstract

摘要： 针对目前P2P贷款违约预测模型受限于借贷双方信息不对称性，未考虑投资人之间差异性的问题，提出了基于信息检索词频-逆文本频率（TF-IDF）算法的P2P贷款违约预测模型。首先以投资效用理论为基础，利用投资人历史投资收益率、贷款利率出价等信息，建立基于投资人效用的贷款违约预测模型；然后，借鉴信息检索TF-IDF算法，构造投资人逆向投资比例因子，对投资人差异性进行量化度量，优化模型中投资人权重计算因子。实验结果表明，该模型预测准确度与其他模型相比平均提高了6%左右，并在不同的测试数据集上都保持最优。

关键词: 贷款违约预测, 效用理论, 信息检索, 词频-逆文本频率, 个人对个人借贷, 曲线下面积

Abstract: Concerning that current P2P loan default prediction models are limited by the information asymmetry of lenders and borrowers, and do not take differences between loan lenders into account, a P2P loan default prediction model based on Term Frequency-Inverse Document Frequency (TF-IDF) algorithm of information retrieval was proposed. Firstly, based on the investment utility theory, a loan default prediction model was established by using the information such as lender's historical investment profit rate and loan bid interest rate. Secondly, referred to TF-IDF algorithm of information retrieval, loan lender's reverse investment scale factor was constructed to quantify the lender's differences, and the weight factor in the model were optimized. Experimental results show that the prediction effect of this model is better than those of other models on different data sets, its prediction accuracy increases by an average of 6% compared with other models.

Key words: loan default prediction, utility theory, information retrieval, Term Frequency-Inverse Document Frequency (TF-IDF), Peer-to-Peer lending (P2P lending), Area Under Curve (AUC)

中图分类号:

章宁, 陈钦. 基于TF-IDF算法的P2P贷款违约预测模型[J]. 计算机应用, 2018, 38(10): 3042-3047.

ZHANG Ning, CHEN Qin. P2P loan default prediction model based on TF-IDF algorithm[J]. Journal of Computer Applications, 2018, 38(10): 3042-3047.

参考文献

[1] WANG H, CHEN K, ZHU W, et al. A process model on P2P lending[EB/OL].[2018-03-10]. https://link.springer.com/content/pdf/10.1186%2Fs40854-015-0002-9.pdf.
[2] 刘轶, 赵宣, 罗春蓉. P2P网络借贷研究:一个文献综述[J]. 金融理论与实践, 2015(6):106-112. (LIU Y, ZHAO X, LUO C R. Research on P2P lending:a literature review[J]. Financial Theory and Practice, 2015(6):106-112.)
[3] 和讯网.2017年P2P网贷行业年报.[EB/OL].[2018-03-31]. http://p2p.hexun.com/2018-01-01/192131797.html. (Hexun.com. 2017 P2P network loan industry annual report[EB/OL].[2018-03-31]. http://p2p.hexun.com/2018-01-01/192131797.html.)
[4] WANG P, ZHENG H, CHEN D, et al. Exploring the critical factors influencing online lending intentions[EB/OL].[2018-01-10]. https://jfin-swufe.springeropen.com/track/pdf/10.1186/s40854-015-0010-9.
[5] BERGER S C, GLEISNER F. Emergence of financial intermediaries in electronic markets:the case of online P2P lending[J]. Business Research, 2010, 2(1):39-65.
[6] EMEKTER R, TU Y. Evaluating credit risk and loan performance in online Peer-to-Peer (P2P) lending[J]. Applied Economics, 2015, 47(1):54-70.
[7] 谈超, 孙本芝, 王冀宁. P2P网络借贷平台中的逾期行为研究[J]. 财会通讯, 2015(2):49-51. (TAN C, SUN B Z, WANG J N. Research on overdue behavior in P2P lending platform[J]. Communication of Finance and Accounting, 2015(2):49-51.)
[8] 高见. 基于核的P2P信贷风险评估模型研究[D]. 武汉:华中科技大学, 2013:17-21. (GAO J. Research on kernel-based model in P2P lending credit risk assessment[D]. Wuhan:Huazhong University of Science and Technology, 2013:17-21.)
[9] 吴小英. P2P网络借贷中的数据分析与决策模型研究[D]. 厦门:厦门大学, 2013:27. (WU X Y. Research on data analysis and decision model in P2P lending[D]. Xiamen:Xiamen University, 2013:27.)
[10] 邓帆帆, 薛菁, 闫海鑫.商业银行参与P2P网络借贷的路径分析及建议——基于贝叶斯网络投资模型的测算结果[J]. 集美大学学报(哲学与社会科学版), 2015, 18(2):53-58. (DENG F F, XUE J, YAN H X. Analysis and suggestions of commercial banks' participation in P2P lending based on the measurement results of Bayesian network model[J]. Journal of Jimei University (Philosophy and Social Sciences), 2015, 18(2):53-58.)
[11] EVERETT C R. Information asymmetry in relationship versus transactional debt markets:evidence from peer-to-peer lending[D]. West Lafayette:Purdue University, 2011:63-66.
[12] 雒春雨. P2P网络借贷中的投资决策模型研究[D]. 大连:大连理工大学, 2012:18-40. (LUO C Y. Research on investment decision model in P2P lending[D]. Dalian:Dalian University of Technology, 2012:18-40.)
[13] LUO C, XIONG H, ZHOU W, et al. Enhancing investment decisions in P2P lending:an investor composition perspective[C]//Proceedings of the 17th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York:ACM, 2011:292-300.
[14] ZHAO H, WU L, LIU Q, et al. Investment recommendation in P2P lending:a portfolio perspective with risk management[C]//Proceedings of the 2014 IEEE International Conference on Data Mining. Washington, DC:IEEE Computer Society, 2014:1109-1114.
[15] WEI Z, LIN M. Auction vs. posted-price:market mechanism, lender behaviors, and transaction outcomes in online crowd-funding[EB/OL].[2018-02-10]. http://www.law.northwestern.edu/research-faculty/searlecenter/events/internet/documents/Wei_Lin_Auction_vs_Posted-Price.pdf.
[16] WIKIPEDIA. Utility[EB/OL].[2018-03-31]. https://en.wikipedia.org/wiki/Utility.
[17] WIKIPEDIA. Market clearing[EB/OL].[2018-02-25]. https://en.wikipedia.org/wiki/Market_clearing.
[18] ROBERTSON S. Understanding inverse document frequency:on theoretical arguments for IDF[J]. Journal of Documentation, 2004, 60(5):503-520.
[19] 吴军.数学之美[M]. 北京:人民邮电出版社, 2014:105-110. (WU J. The Beauty of Mathematics[M]. Beijing:Posts and TeleCom Press, 2014:105-110.)
[20] PROSPER. Prosper data export specification[EB/OL].[2017-10-16]. https://www.prosper.com/Downloads/Services/Documentation/ProsperDataExport_Specification.pdf.
[21] FAWCETT T. An introduction to ROC analysis[J]. Pattern Recognition Letters, 2006, 27(8):861-874.
[22] WIKIPEDIA. Receiver operating characteristic[EB/OL].[2018-03-30]. https://en.wikipedia.org/wiki/Receiver_operating_characteristic.
[23] MYERSON J, GREEN L, WARUSAWITHARANA M, et al. Area under the curve as a measure of discounting[J]. Journal of the Experimental Analysis of Behavior, 2001, 76(2):235-243.
[24] DIETTERICH T G. An experimental comparison of three methods for constructing ensembles of decision trees:bagging, boosting, and randomization[J]. Machine Learning, 2000, 40(2):139-157.
[25] 王玲娣, 徐华. AdaBoost的多样性分析及改进[J]. 计算机应用, 2018, 38(3):650-654. (WANG L D, XU H. Diversity analysis and improvement of AdaBoost[J]. Journal of Computer Applications, 2018, 38(3):650-654.)

基于TF-IDF算法的P2P贷款违约预测模型

P2P loan default prediction model based on TF-IDF algorithm

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	宁进, 陈雷霆, 罗子娟, 周川, 曾慧茹. 离群点检测算法的评价指标[J]. 计算机应用, 2020, 40(9): 2622-2627.
[2]	万志远, 刘勤明, 叶春明, 刘文溢. 突发事件下的医院应急资源冗余配置优化模型[J]. 计算机应用, 2020, 40(2): 584-588.
[3]	章宁, 陈钦. 基于AUC及Q统计值的集成学习训练方法[J]. 计算机应用, 2019, 39(4): 935-939.
[4]	沈力, 刘洪星, 李勇华. 基于版本控制的中文文档到源代码的自动跟踪方法[J]. 计算机应用, 2018, 38(10): 2996-3001.
[5]	袁大曾, 何明星, 李虓, 曾晟珂. 基于点函数秘密共享的私有信息检索协议[J]. 计算机应用, 2017, 37(2): 494-498.
[6]	李岩, 张博文, 郝红卫. 基于语义向量表示的查询扩展方法[J]. 计算机应用, 2016, 36(9): 2526-2530.
[7]	孙芯宇, 吴江, 蒲强. 基于稳定性语义聚类的相关模型估计[J]. 计算机应用, 2016, 36(5): 1313-1318.
[8]	胡小生钟勇. 一种两层加权融合的排序算法[J]. 计算机应用, 2012, 32(12): 3331-3334.
[9]	鲁强李效恋王智广. 程序算法识别研究综述[J]. 计算机应用, 2012, 32(10): 2863-2868.
[10]	李劲张华吴浩雄向军. 基于特定领域的中文微博热点话题挖掘系统BTopicMiner[J]. 计算机应用, 2012, 32(08): 2346-2349.
[11]	李劲张华吴浩雄向军辜希武. 基于社会标注质量的文本分类模型框架[J]. 计算机应用, 2012, 32(05): 1335-1339.
[12]	曾文华黄桦. 基于网页信息检索的地理信息变化检测方法[J]. 计算机应用, 2010, 30(4): 1132-1134.
[13]	袁柳张龙波. 基于统计主题模型的多粒度Web文档标注[J]. 计算机应用, 2010, 30(12): 3401-3406.
[14]	李清茂杨兴江周相兵马洪江. 基于主题图的本体信息检索模型研究[J]. 计算机应用, 2010, 30(1): 240-242.
[15]	付崇国汤志忠. 基于P2P的个性化Web信息检索[J]. 计算机应用, 2010, 30(1): 114-117.