基于数据挖掘的个人征信系统异常查询实时监测模型及其应用

doi:10.11959/j.issn.2096-0271.2016044

Abstract

Abstract:

The data selected contained 900 million query records in the latest 36 months from the personal credit reference system database.The model was subdivided according to different volatility characteristics of each user’s query behavior,and four types of real-time anomaly inquiry monitoring models were discussed and developed.Results indicate that the anomaly inquiry monitoring model is feasible to apply on predicting anomaly query behaviors and showed positive effects.The successful application and constant perfection of the model would definitely exert deterrent effect on illegal query behaviors,force commercial banks to strengthen internal management,protect individual’s private information and right,and promote the healthy development of the credit reference market.

Key words: data mining, personal credit reference system, anomaly inquiry, illegal inquiry, real-time monitoring

Qian YAO, Huamei XIE, Zhigang JING, Qingqing HU, Enzhe SI. Real-time data-mining-based anomaly inquiry monitoring model of personal credit reference system and it's application[J]. Big Data Research, 2016, 2(4): 83-92.

休眠天数T	0	2	84	369	973
用户数占比	0	25%	50%	65%	100%

	9月份状态	当前活跃	休眠1个月	休眠2个月	休眠3个月	休眠4个月	休眠5个月	休眠6个月	休眠7个月	休眠8个月	休眠9个月	休眠10个月	休眠11个月	休眠12个月
	9月份用户数/万	6.7	0.4	0.3	0.4	0.8	0.4	0.2	0.1	0.1	0.2	0.1	0.1	6.1
	9月份占比	42%	2%	2%	3%	5%	3%	1%	1%	1%	1%	1%	0	38%
10月份状态	再次活跃占比	93%	42%	18%	5%	2%	3%	4%	4%	3%	1%	3%	2%	4%
	休眠1个月	7%
	休眠2个月		58%
	休眠3个月			82%
	休眠4个月				95%
	休眠5个月					98%
	休眠6个月						98%
	休眠7个月							96%
	休眠8个月								96%
	休眠9个月									97%
	休眠10个月										99%
	休眠11个月											97%
	休眠12个月												98%
	休眠＞12个月													96%

时间	当前活跃	休眠1个月	休眠2个月	休眠3个月	休眠4个月	休眠5个月	休眠6个月	休眠7个月	休眠8个月	休眠9个月	休眠10个月	休眠11个月	休眠12个月
2015年9月	93%	42%	18%	5%	2%	3%	4%	4%	3%	1%	3%	2%	4%
2015年8月	94%	38%	9%	3%	4%	6%	6%	5%	2%	2%	3%	3%	5%
2015年7月	93%	22%	6%	5%	7%	5%	5%	3%	4%	5%	3%	2%	5%
2015年6月	91%	17%	9%	10%	8%	9%	3%	5%	4%	3%	4%	0	6%
2015年5月	85%	20%	17%	10%	9%	4%	5%	6%	5%	4%	0	1%	6%
2015年4月	91%	38%	22%	20%	8%	9%	10%	5%	4%	0	1%	1%	8%
2015年3月	94%	41%	30%	12%	13%	12%	9%	6%	0%	1%	1%	2%	8%
2015年2月	96%	65%	24%	22%	18%	14%	8%	0%	2%	2%	2%	2%	9%
2015年1月	89%	25%	18%	12%	9%	5%	0%	1%	2%	2%	1%	1%	5%
均值	92%	34%	17%	11%	9%	7%	6%	4%	3%	2%	2%	1%	6%

序号	类型	特征
1	活跃无断点	观察期内均有查询记录
2	活跃有断点	观察期内存在无查询记录的月份，但连续不超过3个月
3	新开用户	只有最近6个月的查询记录
4	次新用户	有超过6个月但小于两年的查询记录
5	当前休眠用户	观察期的最近3个月连续无查询记录
6	历史休眠用户	其他

序号	类型	补充缺失值
1	活跃无断点	无需处理
2	活跃有断点	用相邻月份的均值填充
3	新开用户	待查询记录满6个月后再开始监测
4	次新用户	截取最新6个月的数据建模
5	当前休眠用户	直接设定阈值为200次，此类用户再次启用即需关注
6	历史休眠用户	用24个月非0查询量的均值填充

Real-time data-mining-based anomaly inquiry monitoring model of personal credit reference system and it's application

RichHTML

PDF下载

Knowledge

Cited

Abstract

Cite this article

share this article

Figures/Tables 10

References 4

Related Articles 13

Metrics

Recommended 0

类别	波动特征	算法
平稳型（全国性商业银行）	10.11959/j.issn.2096-0271.2016044.F004	简单平均法备注：查询量锐减系当月开通接口查询
增长型（中国人民银行）	10.11959/j.issn.2096-0271.2016044.F005	月度差分加权移动平均算法
小幅跳跃型（城市商业银行）	10.11959/j.issn.2096-0271.2016044.F006	加权移动平均
周期跳跃型（财务公司）	10.11959/j.issn.2096-0271.2016044.F007	年度差分算法

	最大相对误差	次大相对误差	最小相对误差	平均精度	泰勒不等系数
平稳型	23.71%	23.47%	0.65%	84.45%	0.085
增长型	45.52%	25.49%	25.02%	68.21%	0.213
小幅跳跃型	66.20%	34.63%	2.96%	73.86%	0.128
周期跳跃型	31.30%	28.08%	0.57%	85.09%	0.139

[1]	HAN J , KAMBER M . Data mining concepts and techniques[M]. Translated by FAN M,MENG X F. Beijing: China Machine PressPress, 2012.
[2]	PANG-NING T , STEINBACH M , KUMAR V . Introduction to data mining[M]. Translated by FAN M,FAN H J. Beijing: Posts ＆ Telecom PressPress, 2011.
[3]	中华人民共和国国务院. Introduction to data mining[M]. 北京: 中国法制出版社, 2013.中华人民共和国国务院.. The State Council of the People’s Republic of China[M]. Beijing: China Legal Publishing HousePress, 2013.
[4]	汪路 . 论征信的本质及其主要特征[J]. 西部金融, 2010(6): 60-62. WANG L . On the essence and main features of credit reference[J]. West China Finance, 2010(6): 60-62.

[1]	Deren LI, Guo ZHANG, Yonghua JIANG, Xin SHEN, Weiling LIU. Opportunities and challenges of geo-spatial information science from the perspective of big data [J]. Big Data Research, 2022, 8(2): 3-14.
[2]	Shuai REN, Dandan CHEN, Genshen CHU, He BAI, Huizhao LI, Yuanjie HE, Changjun HU. Discovery of irradiation mechanism based on big data of material simulation [J]. Big Data Research, 2021, 7(6): 3-18.
[3]	An WANG, Shuai REN, Xue MIAO, Lingyu DONG, Ying ZHU, Dandan CHEN, Changjun HU. Big data of numerical nuclear reactor and its application [J]. Big Data Research, 2021, 7(5): 40-56.
[4]	Liangchen XU, Chonghui GUO. Analysis and utilization framework of electronic medical records under the background of smart hospital construction [J]. Big Data Research, 2021, 7(4): 141-156.
[5]	Xi CHEN, Yinhai WANG, Zhuang DAI, Xiaolei MA. Research on demand identification for customized bus based on multi-source mobility data [J]. Big Data Research, 2020, 6(6): 105-118.
[6]	Yuhang YAO, Junjie OU, Yang LI, Luoyi FU, Xinbing WANG, Guihai CHEN. Turing index:cross-domain and cross-generation metric of unraveling scholars’ impact in academic big data [J]. Big Data Research, 2019, 5(5): 25-37.
[7]	Pingwen ZHANG, Weinan E, Xiaoru YUAN, Yiming FU. Big data analysis and application technology innovation platform [J]. Big Data Research, 2018, 4(4): 86-93.
[8]	Yong SHI, Fan MENG. Credit scoring:basic theory and applications [J]. Big Data Research, 2017, 3(1): 19-26.
[9]	Jiye LIANG, Yuhua QIAN, Deyu Li, Qinghua HU. Research development on granular computing theory and method for big data [J]. Big Data Research, 2016, 2(4): 13-23.
[10]	Binxing FANG, Yan JIA, Aiping LI, Rong JIANG. Privacy preservation in big data:a survey [J]. Big Data Research, 2016, 2(1): 1-18.
[11]	Libin YANG, Hailin LI, Feibo ZHANG. Research on development of management information system in big data environment [J]. Big Data Research, 2016, 2(1): 86-98.
[12]	. Data Mining in the Era of Big Data:From the Application Perspective [J]. Big Data Research, 2015, 1(4): 57-80.
[13]	Yun Xiong, Yangyong Zhu. Abnormal Group Mining:Framework and Applications [J]. Big Data Research, 2015, 1(2): 66-77.