基于缴费行为轨迹追踪的渠道风险监控模型
|
摘要
电费收取过程中的第三方缴费渠道隐藏着巨大的风险监控漏洞,对电网企业构成严重的资金安全隐患。在已明确风险类型、成因机理和表现形式的基础之上,从追踪客户的缴费行为轨迹出发,提出了一个针对缴费渠道的风险监控模型。先分析客户缴费行为记录并追踪其缴费行为轨迹,使用决策树算法与正则表达式相结合的方法判断出客户潜在的风险类型;然后确定各风险类型的评估指标,建立递阶的评估指标体系;最后使用加权最小二乘法创建风险评估模型,计算渠道各风险类型的风险指数,进而得到渠道的各风险排名。该模型将定性识别与定量评估分离,可以有效地监管渠道风险且获得了可信度较高的评估结果。
It is found that there are still huge hidden risks in the third-party payment channels during the process of electricity charge collection, which poses serious capital security risks to power grid enterprises. Based on risk types, cause mechanisms and expression forms, a risk monitoring model of payment channels is proposed from the perspective on tracking the trajectory of customer's payment. The process is firstly to analyze the customer's payment records and track the trajectory of their payment behaviors, by combining the decision tree algorithm with regular expressions to judge potential risk types in the customer, then to confirm evaluation indicators and establish a hierarchical evaluation index system, and finally to establish a risk assessment model with the weighted least square method(WLS) to obtain the risk ranking of the channels from the risk index of each risk type. The model separates qualitative cognition from quantitative assessment, and can effectively monitor the channel's risk in the process of electricity charge collection and obtain higher reliability risk assessment results.
It is found that there are still huge hidden risks in the third-party payment channels during the process of electricity charge collection, which poses serious capital security risks to power grid enterprises. Based on risk types, cause mechanisms and expression forms, a risk monitoring model of payment channels is proposed from the perspective on tracking the trajectory of customer's payment. The process is firstly to analyze the customer's payment records and track the trajectory of their payment behaviors, by combining the decision tree algorithm with regular expressions to judge potential risk types in the customer, then to confirm evaluation indicators and establish a hierarchical evaluation index system, and finally to establish a risk assessment model with the weighted least square method(WLS) to obtain the risk ranking of the channels from the risk index of each risk type. The model separates qualitative cognition from quantitative assessment, and can effectively monitor the channel's risk in the process of electricity charge collection and obtain higher reliability risk assessment results.
引文
[1] 陈媛. 电网企业电费管理研究[D].北京:华北电力大学(北京),2011.
[2] 张燕. 电力客户信用风险评估管理系统的设计与实现[D].保定:华北电力大学,2013.
[3] NIU D, LIU Y, JI L, et al. The study of electric power customer risk evaluation based on entropy method and MADA[C]// 2011 IEEE International Conference on Computer Science and Automation Engineering. Shanghai: CSAE, 2011, 3: 401-404.
[4] 曹清山,郑梦莲,丁一,等. 新电改背景下基于多属性决策的电力客户评估和选择研究[J]. 电网技术,2018,42(1):117-125.
[5] 张吉军.模糊层次分析法(FAHP)[J].模糊系统与数学, 2000,14(2):80-88.
[6] 陈婷婷,姜健,张丽华. 基于Fuzzy AHP的电力客户信用等级评价[J].科技和产业,2013,13(5):107-109.
[7] 刘凌燕,谢振平,陈洵,等. 基于异源交叉回归分析的电力客户缴费风险建模[J].江苏科技大学学报(自然科学版), 2017,31(2):200-206.
[8] 薛振宇,胡航海,宋毅,等. 基于大数据分析的县公司综合评价策略[J].电力自动化设备,2017,37(9):199-204.
[9] PENG X G, LAI J W, CHEN Y. Application of clustering analysis in typical power consumption profile analysis[J]. Power System Protection and Control, 2014, 42(19): 68-73.
[10] 王炼,孙和平,谢振平,等.一种基于缴费积极性的电力客户信用评估模型[J].计算机工程与应用,2016,52(22):253-259.
[11] ORESKI S, ORESKI G. Genetic algorithm-based heuristic for feature selection in credit risk assessment[J]. Expert Systems with Applications, 2014, 41(4): 2052-2064.
[12] YU L, YAO X, WANG S, et al. Credit risk evaluation using a weighted least squares SVM classifier with design of experiment for parameter selection[J]. Expert Systems with Applications, 2011, 38(12):15392-15399.
[13] 徐树方,高立,张平文.数值线性代数(第二版)[M].北京:北京大学出版社,2013:76-84.
[14] 王少鹏,闻英友,赵宏,等.一种滑动窗口下数据流Disjoint查询的增量处理算法[J].计算机学报,2017,40(10):2381-2403.
[2] 张燕. 电力客户信用风险评估管理系统的设计与实现[D].保定:华北电力大学,2013.
[3] NIU D, LIU Y, JI L, et al. The study of electric power customer risk evaluation based on entropy method and MADA[C]// 2011 IEEE International Conference on Computer Science and Automation Engineering. Shanghai: CSAE, 2011, 3: 401-404.
[4] 曹清山,郑梦莲,丁一,等. 新电改背景下基于多属性决策的电力客户评估和选择研究[J]. 电网技术,2018,42(1):117-125.
[5] 张吉军.模糊层次分析法(FAHP)[J].模糊系统与数学, 2000,14(2):80-88.
[6] 陈婷婷,姜健,张丽华. 基于Fuzzy AHP的电力客户信用等级评价[J].科技和产业,2013,13(5):107-109.
[7] 刘凌燕,谢振平,陈洵,等. 基于异源交叉回归分析的电力客户缴费风险建模[J].江苏科技大学学报(自然科学版), 2017,31(2):200-206.
[8] 薛振宇,胡航海,宋毅,等. 基于大数据分析的县公司综合评价策略[J].电力自动化设备,2017,37(9):199-204.
[9] PENG X G, LAI J W, CHEN Y. Application of clustering analysis in typical power consumption profile analysis[J]. Power System Protection and Control, 2014, 42(19): 68-73.
[10] 王炼,孙和平,谢振平,等.一种基于缴费积极性的电力客户信用评估模型[J].计算机工程与应用,2016,52(22):253-259.
[11] ORESKI S, ORESKI G. Genetic algorithm-based heuristic for feature selection in credit risk assessment[J]. Expert Systems with Applications, 2014, 41(4): 2052-2064.
[12] YU L, YAO X, WANG S, et al. Credit risk evaluation using a weighted least squares SVM classifier with design of experiment for parameter selection[J]. Expert Systems with Applications, 2011, 38(12):15392-15399.
[13] 徐树方,高立,张平文.数值线性代数(第二版)[M].北京:北京大学出版社,2013:76-84.
[14] 王少鹏,闻英友,赵宏,等.一种滑动窗口下数据流Disjoint查询的增量处理算法[J].计算机学报,2017,40(10):2381-2403.