CN113256402A - Risk control rule determination method and device and electronic equipment - Google Patents
Risk control rule determination method and device and electronic equipment Download PDFInfo
- Publication number
- CN113256402A CN113256402A CN202110621370.7A CN202110621370A CN113256402A CN 113256402 A CN113256402 A CN 113256402A CN 202110621370 A CN202110621370 A CN 202110621370A CN 113256402 A CN113256402 A CN 113256402A
- Authority
- CN
- China
- Prior art keywords
- credit business
- control rule
- decision tree
- risk control
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012954 risk control Methods 0.000 title claims abstract description 148
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000003066 decision tree Methods 0.000 claims abstract description 143
- 238000012549 training Methods 0.000 claims abstract description 68
- 230000011218 segmentation Effects 0.000 claims abstract description 63
- 238000012360 testing method Methods 0.000 claims abstract description 47
- 238000012545 processing Methods 0.000 claims abstract description 45
- 238000012795 verification Methods 0.000 claims description 37
- 230000008569 process Effects 0.000 claims description 19
- 238000010200 validation analysis Methods 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 7
- 238000000638 solvent extraction Methods 0.000 claims description 6
- 238000012550 audit Methods 0.000 claims description 5
- 238000012216 screening Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 6
- 230000006870 function Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- General Business, Economics & Management (AREA)
- Finance (AREA)
- Data Mining & Analysis (AREA)
- Accounting & Taxation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The application provides a risk control rule determination method, a risk control rule determination device and electronic equipment, and relates to the technical field of credit wind control. In the application, firstly, a plurality of pieces of credit business sample information are obtained; secondly, carrying out segmentation processing on a plurality of pieces of credit business sample information to form a training sample set and a test sample set; then, performing segmentation processing on credit business sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit business risk control rule; and finally, verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting the target conditions, and judging whether the target credit business is approved or not. Based on the method, the problem of poor risk control effect in the existing credit wind control technology can be solved.
Description
Technical Field
The application relates to the technical field of credit wind control, in particular to a risk control rule determination method, a risk control rule determination device and electronic equipment.
Background
The consumer credit market is showing tremendous potential, and various financial institutions are deploying personal online credit business. In online credit consumption wind control, different wind control rules need to be designed in a customized manner aiming at different passenger groups and different products, so as to ensure that fraudulent passenger groups and passenger groups with low repayment capacity are excluded in a credit-before-credit granting stage. However, through research by the inventor, it is found that in the prior art, because the wind control rule set based on experience has many artificial uncertainty factors, different people often set different thresholds, which may cause inaccuracy of the wind control strategy, such as too high disturbance rate or too low accuracy rate, and thus, there is a problem of poor risk control effect. .
Disclosure of Invention
In view of the above, an object of the present application is to provide a method, an apparatus and an electronic device for determining a risk control rule, so as to solve the problem of poor risk control effect existing in the existing credit wind control technology.
In order to achieve the above purpose, the embodiment of the present application adopts the following technical solutions:
a risk control rule determination method is applied to electronic equipment and comprises the following steps:
obtaining a plurality of pieces of credit business sample information, wherein each piece of credit business sample information is generated based on each history after credit business is processed, each piece of credit business sample information has label information, and the label information is used for representing whether a credit customer is overdue or not;
performing segmentation processing on the multiple pieces of credit business sample information to form a training sample set and a test sample set, wherein the training sample set comprises the multiple pieces of credit business sample information, and the test sample set comprises the multiple pieces of credit business sample information;
performing segmentation processing on the credit business sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit business risk control rule;
and verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting target conditions, wherein the at least one target credit business risk control rule is used for judging whether the overdue probability of the target credit business reaches a preset probability, and when the overdue probability is greater than the preset probability, determining that the target credit business is not approved.
In a preferred selection of the embodiment of the present application, in the method for determining a risk control rule, the step of performing segmentation processing on the plurality of pieces of credit business sample information to form a training sample set and a testing sample set includes:
acquiring predetermined target proportion information;
and performing segmentation processing on the plurality of pieces of credit business sample information based on the target proportion information to form a training sample set comprising a first number of pieces of credit business sample information and a testing sample set comprising a second number of pieces of credit business sample information, wherein the ratio between the first number and the second number is equal to the target proportion information, and the first number is larger than the second number.
In a preferred selection of the embodiment of the present application, in the method for determining a risk control rule, the step of performing segmentation processing on the credit business sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit business risk control rule includes:
obtaining a first predetermined decision tree parameter and a second predetermined decision tree parameter, wherein the first decision tree parameter is used for representing the minimum purity of node division of the decision tree during the segmentation processing, the second decision tree parameter is used for representing the minimum sample number required by the node division of the decision tree during the segmentation processing, the first decision tree parameter is less than 0.001 and greater than 0.01, and the second decision tree parameter is less than 10 and greater than 100;
and segmenting the credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one credit business risk control rule.
In a preferred selection of the embodiment of the present application, in the method for determining a risk control rule, the step of performing segmentation processing on the credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter, and a target decision tree algorithm to obtain at least one credit business risk control rule includes:
performing first segmentation processing on credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one initial credit business risk control rule;
determining whether the interpretability of the at least one initial credit business risk control rule reaches a target degree, and adjusting the first decision tree parameter and the second decision tree parameter to obtain a new first decision tree parameter and a new second decision tree parameter when the interpretability does not reach the target degree;
and performing second segmentation processing on the credit business sample information in the training sample set based on the new first decision tree parameters, the new second decision tree parameters and a target decision tree algorithm to obtain at least one credit business risk control rule.
In a preferred option of an embodiment of the present application, in the above risk control rule determining method, the step of performing verification processing on the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting a target condition includes:
verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain a verification result;
judging whether the verification result meets a target condition;
and if the verification result meets the target condition, determining the at least one credit business risk control rule as a target credit business risk control rule.
In a preferred option of an embodiment of the present application, in the method for determining a risk control rule, the step of performing verification processing on the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting a target condition further includes:
if the verification result does not meet the target condition, screening a plurality of sample features included in at least one piece of credit service sample information in the training sample set to obtain a new training sample set, wherein the number of the credit service sample information included in the new training sample set is the same as the number of the credit service sample information included in the training sample set;
and segmenting the credit business sample information in the new training sample set based on the target decision tree algorithm, and taking the obtained at least one credit business risk control rule as at least one target credit business risk control rule meeting the target condition.
An embodiment of the present application further provides a risk control rule determining apparatus, which is applied to an electronic device, and the risk control rule determining apparatus includes:
the credit business system comprises a sample information acquisition module, a credit business analysis module and a credit business analysis module, wherein the sample information acquisition module is used for acquiring a plurality of pieces of credit business sample information, each piece of credit business sample information is generated based on each history after credit business is processed, each piece of credit business sample information has label information, and the label information is used for representing whether a credit customer is overdue or not;
the sample information segmentation module is used for segmenting the plurality of pieces of credit business sample information to form a training sample set and a test sample set, wherein the training sample set comprises the plurality of pieces of credit business sample information, and the test sample set comprises the plurality of pieces of credit business sample information;
the sample information segmentation module is used for carrying out segmentation processing on the credit business sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit business risk control rule;
and the control rule verification module is used for verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set so as to obtain at least one target credit business risk control rule meeting target conditions, wherein the at least one target credit business risk control rule is used for judging whether the overdue probability of the target credit business reaches a preset probability, and when the overdue probability is greater than the preset probability, the target credit business is determined not to pass the audit.
In a preferred option of the embodiment of the present application, in the risk control rule determining device, the sample information segmentation module includes:
a decision tree parameter obtaining submodule, configured to obtain a first decision tree parameter and a second decision tree parameter that are predetermined, where the first decision tree parameter is used to represent a minimum degree of uncertainty of node partitioning of a decision tree during segmentation processing, the second decision tree parameter is used to represent a minimum number of samples required by node partitioning of the decision tree during segmentation processing, the first decision tree parameter is less than 0.001 and greater than 0.01, and the second decision tree parameter is less than 10 and greater than 100;
and the sample information segmentation submodule is used for carrying out segmentation processing on the credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one credit business risk control rule.
In a preferable selection of the embodiment of the present application, in the risk control rule determining device, the control rule verifying module includes:
the control rule verification submodule is used for verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain a verification result;
the verification result judgment submodule is used for judging whether the verification result meets the target condition;
and the control rule determining submodule is used for determining the at least one credit business risk control rule as a target credit business risk control rule when the verification result meets the target condition.
On the basis, an embodiment of the present application further provides an electronic device, including:
a memory for storing a computer program;
and the processor is connected with the memory and is used for executing the computer program stored in the memory so as to realize the risk control rule determination method.
According to the risk control rule determining method, the risk control rule determining device and the electronic equipment, on one hand, the credit business sample information is segmented through the decision tree, so that the credit business risk control rule can be obtained, and therefore compared with a conventional technical scheme based on artificial experience rule setting, the reliability and accuracy of the credit business risk control rule can be improved. On the other hand, the credit business sample information is divided into the training sample set and the testing sample set, so that the credit business risk control rule obtained based on the training sample set can be verified based on the testing sample set, and the target credit business risk control rule is obtained.
In order to make the aforementioned objects, features and advantages of the present application more comprehensible, preferred embodiments accompanied with figures are described in detail below.
Drawings
Fig. 1 is a block diagram of an electronic device according to an embodiment of the present disclosure.
Fig. 2 is a schematic flow chart of a risk control rule determination method according to an embodiment of the present application.
Fig. 3 is a block diagram illustrating a risk control rule determination apparatus according to an embodiment of the present application.
Icon: 10-an electronic device; 12-a memory; 14-a processor; 100-risk control rule determination means; 110-a sample information acquisition module; 120-a sample information segmentation module; 130-sample information segmentation module; 140-control rule validation module.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all the embodiments. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations.
Thus, the following detailed description of the embodiments of the present application, presented in the accompanying drawings, is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
As shown in fig. 1, an electronic device 10 according to an embodiment of the present disclosure may include a memory 12, a processor 14, and a risk control rule determining apparatus 100.
Wherein the memory 12 and the processor 14 are electrically connected directly or indirectly to realize data transmission or interaction. For example, they may be electrically connected to each other via one or more communication buses or signal lines. The risk control rule determination means 100 comprises at least one software function module which may be stored in the memory 12 in the form of software or firmware. The processor 14 is configured to execute an executable computer program stored in the memory 12, for example, a software functional module and a computer program included in the risk control rule determination apparatus 100, so as to implement the risk control rule determination method provided in the embodiment of the present application.
Alternatively, the Memory 12 may be, but is not limited to, a Random Access Memory (RAM), a Read Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Read-Only Memory (EPROM), an electrically Erasable Read-Only Memory (EEPROM), and the like.
The Processor 14 may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), a System on Chip (SoC), and the like; but may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.
It is understood that the structure shown in fig. 1 is only an illustration, and the electronic device 10 may further include more or less components than those shown in fig. 1, or have a different configuration from that shown in fig. 1, for example, a communication unit for information interaction with other devices (such as a database server) may also be included.
Wherein the electronic device 10 may be a server having data processing capabilities.
With reference to fig. 2, an embodiment of the present application further provides a risk control rule determining method applicable to the electronic device 10. Wherein the method steps defined by the flow related to the risk control rule determination method may be implemented by the electronic device 10.
The specific process shown in FIG. 2 will be described in detail below.
Step S110, obtaining a plurality of pieces of credit business sample information.
In this embodiment, the electronic device 10 may first obtain a plurality of pieces of credit service sample information.
Each piece of credit business sample information is generated based on each history after credit business is processed, and each piece of credit business sample information has label information which is used for representing whether a credit customer is overdue or not. It is understood that tag information corresponding to the occurrence of a timeout (e.g., 30 days or more) may be represented by a first value, such as 1, and tag information corresponding to the non-occurrence of a timeout (e.g., 30 days or less) may be represented by a second value, such as 0. It is understood that the credit business sample information may include basic information of credit customers including age, sex, whether marriage is present, unit of residence, etc., credit information including contents of credit report of bank, and three-party information including multi-throw information such as loans at a plurality of banks or financial institutions, etc.
And step S120, carrying out segmentation processing on the plurality of pieces of credit business sample information to form a training sample set and a testing sample set.
In this embodiment, after obtaining the pieces of credit business sample information based on step S110, the electronic device may perform a segmentation process on the pieces of credit business sample information, so that a corresponding training sample set and a corresponding testing sample set may be formed.
Wherein the training sample set may include a plurality of pieces of credit business sample information, the testing sample set may include a plurality of pieces of credit business sample information, and there is no intersection between the training sample set and the testing sample set.
Step S130, performing segmentation processing on the credit business sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit business risk control rule.
In this embodiment, after obtaining the training sample set based on step S120, the electronic device 10 may perform a segmentation process on the credit business sample information in the training sample set based on a target decision tree algorithm, so that at least one credit business risk control rule may be obtained.
It is understood that the number of the credit business risk control rules may be determined based on the number of nodes of the decision tree after the segmentation process, so that at least one credit business risk control rule may be obtained by integrating the rule corresponding to each node on the decision tree.
Step S140, verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting target conditions.
In this embodiment, after obtaining the test sample set based on step S120 and obtaining the at least one credit business risk control rule based on step S130, the electronic device 10 may perform a verification process on the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting a target condition. The at least one target credit business risk control rule is used for judging whether the overdue probability of the target credit business reaches a preset probability or not, and when the overdue probability is larger than the preset probability, the target credit business is determined not to be approved. The preset probability may be generated based on configuration operations performed by a user, and specific numerical values are not limited.
Based on the method, on one hand, the credit business sample information is segmented through the decision tree, so that the credit business risk control rule can be obtained, and therefore, compared with the conventional technical scheme based on the artificial experience setting rule, the reliability and the accuracy of the credit business risk control rule can be improved. On the other hand, the credit business sample information is divided into the training sample set and the testing sample set, so that the credit business risk control rule obtained based on the training sample set can be verified based on the testing sample set, and the target credit business risk control rule is obtained.
It is understood that, in the above example, when step S110 is executed, the plurality of pieces of credit service sample information may be acquired based on the following steps:
firstly, acquiring a plurality of pieces of original credit business sample information, wherein the original credit business sample information can comprise a plurality of kinds of information; then, the original credit business sample information can be cleaned to obtain credit business sample information, wherein the cleaning can be to screen out information which is not related to credit auditing and the like; finally, tag information, such as 0 or 1, may be added to each piece of credit service sample information.
It is to be understood that, in the above example, when performing step S120, the training sample set and the test sample set may be formed based on the following steps:
firstly, obtaining predetermined target proportion information; secondly, the plurality of pieces of credit business sample information are subjected to segmentation processing based on the target proportion information to form a training sample set comprising a first number of pieces of credit business sample information and a testing sample set comprising a second number of pieces of credit business sample information, wherein the ratio between the first number and the second number is equal to the target proportion information, and the first number is greater than the second number, for example, in an alternative example, the ratio between the first number and the second number can be 7: 3.
It is understood that, in the above example, when step S130 is executed, at least one credit business risk control rule may be obtained based on the following steps:
firstly, obtaining a first predetermined decision Tree parameter And a second predetermined decision Tree parameter, where the first decision Tree parameter is used to represent a minimum impure degree of node partitioning of a decision Tree during a segmentation process (for example, a parameter min _ impurity _ hierarchical in a CART decision Tree, where the CART decision Tree refers to a Classification Regression Tree And a Classification And Regression Tree), the second decision Tree parameter is used to represent a minimum sample number required by node partitioning of the decision Tree during the segmentation process (for example, a parameter min _ samples _ split in the CART decision Tree), the first decision Tree parameter may be greater than 0.001 And less than 0.01, And the second decision Tree parameter may be greater than 10 And less than 100;
secondly, performing segmentation processing on the credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one credit business risk control rule.
Based on this, by predetermining the first decision tree parameter and the second decision tree parameter, the segmentation processing of the sample information can be limited to ensure the rejection rate (i.e. the ratio of the sample information of the credit service for which the audit is not passed). That is, the specific values of the first and second decision tree parameters may be determined in conjunction with the amount of credit service sample information and rejection rate.
It will be appreciated that in an alternative example, the credit business sample information in the training sample set may be segmented based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm based on the following steps:
firstly, performing first segmentation processing on credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree to obtain at least one initial credit business risk control rule;
secondly, determining whether the interpretability of the at least one initial credit business risk control rule reaches a target degree, and adjusting the first decision tree parameter and the second decision tree parameter to obtain a new first decision tree parameter and a new second decision tree parameter when the interpretability does not reach the target degree;
and then, performing second segmentation processing on the credit business sample information in the training sample set based on the new first decision tree parameters, the new second decision tree parameters and the target decision tree to obtain at least one credit business risk control rule.
It is to be understood that, in the above steps, the first and second slicing processes may refer to two slicing processes performed one after another based on different parameters. Whether the interpretability reaches the target degree may be in response to an operational determination by a corresponding training person.
It is to be understood that, in the above steps, when the first decision tree parameter and the second decision tree parameter are adjusted, the first decision tree parameter and the second decision tree parameter may be reduced, that is, the new first decision tree parameter may be smaller than the first decision tree parameter, and the new second decision tree parameter may be smaller than the second decision tree parameter. And the new first decision tree is larger than 0.001 and smaller than 0.01, and the new second decision tree parameter is also larger than 10 and smaller than 100.
It is to be understood that, in the above example, when the step S130 is executed, the verification process may be performed on the at least one credit business risk control rule based on the following steps:
firstly, verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain a verification result;
secondly, judging whether the verification result meets a target condition;
and then, if the verification result meets the target condition, determining the at least one credit business risk control rule as a target credit business risk control rule.
It is understood that, in the above steps, performing the verification process may refer to performing a prediction process on the credit business sample information in the test sample set based on the at least one credit business risk control rule, and then comparing a prediction result (such as that the audit is passed or that the audit is not passed) with the corresponding tag information to obtain a predicted accuracy, so that when the accuracy is greater than or equal to a certain threshold, it may be determined that the target condition is met, and when the accuracy is less than the threshold, it may be determined that the target condition is not met.
It is understood that, in the above example, when step S130 is executed, the target credit business risk control rule may also be obtained based on the following steps:
firstly, if the verification result does not meet the target condition, screening a plurality of sample features included in at least one piece of credit business sample information in the training sample set to obtain a new training sample set, wherein the number of credit business sample information included in the new training sample set is the same as the number of credit business sample information included in the training sample set;
secondly, performing segmentation processing on the credit business sample information in the new training sample set based on the target decision tree, and using the obtained at least one credit business risk control rule as at least one target credit business risk control rule meeting the target condition.
It is understood that, in the above step, the screening process on the sample features may refer to discarding some sample features of a plurality of sample features included in a piece of credit business sample information, such as some sample features whose discrimination is unstable or which are easily over-fitted. The sample characteristics may refer to specific information included in the credit service sample information, such as gender, marriage, age, credit, and the like, as described above.
It will be appreciated that in an alternative example, the target credit business risk control rule may be that credit applications for credit customers older than the target age are not passed, or credit applications for credit customers whose annual income is less than the target amount are not passed, or credit applications for credit customers whose debt amount is greater than the target debt amount are not passed, etc.
With reference to fig. 3, an embodiment of the present application further provides a risk control rule determining apparatus 100 applicable to the electronic device 10. The risk control rule determination apparatus 100 may include a sample information obtaining module 110, a sample information dividing module 120, a sample information dividing module 130, and a control rule verifying module 140.
The sample information obtaining module 110 is configured to obtain multiple pieces of credit service sample information, where each piece of credit service sample information is generated based on each history after credit service is processed, and each piece of credit service sample information has tag information, where the tag information is used to represent whether a credit customer is overdue. In this embodiment, the sample information obtaining module 110 may be configured to perform step S110 shown in fig. 2, and reference may be made to the foregoing description of step S110 regarding relevant contents of the sample information obtaining module 110.
The sample information segmentation module 120 is configured to segment the multiple pieces of credit service sample information to form a training sample set and a test sample set, where the training sample set includes the multiple pieces of credit service sample information, and the test sample set includes the multiple pieces of credit service sample information. In this embodiment, the sample information segmentation module 120 may be configured to perform step S120 shown in fig. 2, and reference may be made to the foregoing description of step S120 for relevant contents of the sample information segmentation module 120.
The sample information segmentation module 130 is configured to segment the credit service sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit service risk control rule. In this embodiment, the sample information segmentation module 130 may be configured to perform step S130 shown in fig. 2, and reference may be made to the foregoing description of step S130 for relevant contents of the sample information segmentation module 130.
The control rule verification module 140 is configured to perform verification processing on the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting a target condition, where the at least one target credit business risk control rule is used to determine whether an overdue probability of a target credit business reaches a preset probability, and when the overdue probability is greater than the preset probability, determine that the target credit business is not approved. In this embodiment, the control rule verification module 140 may be configured to execute step S140 shown in fig. 2, and the relevant content of the control rule verification module 140 may refer to the foregoing description of step S140.
It will be appreciated that in an alternative example, the sample information segmentation module 130 includes a decision tree parameter acquisition sub-module and a sample information segmentation sub-module.
In detail, the decision tree parameter obtaining sub-module is configured to obtain a first decision tree parameter and a second decision tree parameter that are predetermined, where the first decision tree parameter is used to represent a minimum impure degree of node division of the decision tree during the splitting process, the second decision tree parameter is used to represent a minimum sample number required by the node division of the decision tree during the splitting process, the first decision tree parameter is less than 0.001 and greater than 0.01, and the second decision tree parameter is less than 10 and greater than 100. And the sample information segmentation submodule is used for carrying out segmentation processing on the credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one credit business risk control rule.
It will be appreciated that in an alternative example, the control rule validation module 140 includes a control rule validation sub-module, a validation result determination sub-module, and a control rule determination sub-module.
In detail, the control rule validation submodule is configured to perform validation processing on the at least one credit business risk control rule based on the credit business sample information in the test sample set, so as to obtain a validation result. And the verification result judgment submodule is used for judging whether the verification result meets the target condition. The control rule determining submodule is used for determining the at least one credit business risk control rule as a target credit business risk control rule when the verification result meets the target condition.
In summary, according to the risk control rule determining method, the risk control rule determining device and the electronic device provided by the application, on one hand, the credit business sample information is segmented through the decision tree, so that the credit business risk control rule can be obtained, and therefore, compared with a conventional scheme based on an artificial experience setting rule, the reliability and the accuracy of the credit business risk control rule can be improved. On the other hand, the credit business sample information is divided into the training sample set and the testing sample set, so that the obtained credit business risk control rule can be verified based on the testing sample set, and the target credit business risk control rule is obtained.
In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus and method embodiments described above are illustrative only, as the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
In addition, functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.
The functions, if implemented in the form of software functional modules and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, an electronic device, or a network device) to perform all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes. It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The above description is only a preferred embodiment of the present application and is not intended to limit the present application, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims (10)
1. A risk control rule determination method is applied to electronic equipment, and comprises the following steps:
obtaining a plurality of pieces of credit business sample information, wherein each piece of credit business sample information is generated based on each history after credit business is processed, each piece of credit business sample information has label information, and the label information is used for representing whether a credit customer is overdue or not;
performing segmentation processing on the multiple pieces of credit business sample information to form a training sample set and a test sample set, wherein the training sample set comprises the multiple pieces of credit business sample information, and the test sample set comprises the multiple pieces of credit business sample information;
performing segmentation processing on the credit business sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit business risk control rule;
and verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting target conditions, wherein the at least one target credit business risk control rule is used for judging whether the overdue probability of the target credit business reaches a preset probability, and when the overdue probability is greater than the preset probability, determining that the target credit business is not approved.
2. The risk control rule determination method according to claim 1, wherein the step of performing segmentation processing on the plurality of pieces of credit business sample information to form a training sample set and a testing sample set comprises:
acquiring predetermined target proportion information;
and performing segmentation processing on the plurality of pieces of credit business sample information based on the target proportion information to form a training sample set comprising a first number of pieces of credit business sample information and a testing sample set comprising a second number of pieces of credit business sample information, wherein the ratio between the first number and the second number is equal to the target proportion information, and the first number is larger than the second number.
3. The method for determining the risk control rule according to claim 1, wherein the step of performing segmentation processing on the credit business sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit business risk control rule comprises:
obtaining a first predetermined decision tree parameter and a second predetermined decision tree parameter, wherein the first decision tree parameter is used for representing the minimum purity of node division of the decision tree during the segmentation processing, the second decision tree parameter is used for representing the minimum sample number required by the node division of the decision tree during the segmentation processing, the first decision tree parameter is less than 0.001 and greater than 0.01, and the second decision tree parameter is less than 10 and greater than 100;
and segmenting the credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one credit business risk control rule.
4. The risk control rule determination method according to claim 3, wherein the step of performing segmentation processing on the credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one credit business risk control rule comprises:
performing first segmentation processing on credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one initial credit business risk control rule;
determining whether the interpretability of the at least one initial credit business risk control rule reaches a target degree, and adjusting the first decision tree parameter and the second decision tree parameter to obtain a new first decision tree parameter and a new second decision tree parameter when the interpretability does not reach the target degree;
and performing second segmentation processing on the credit business sample information in the training sample set based on the new first decision tree parameters, the new second decision tree parameters and a target decision tree algorithm to obtain at least one credit business risk control rule.
5. The risk control rule determination method according to claim 1, wherein the step of performing a verification process on the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting a target condition comprises:
verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain a verification result;
judging whether the verification result meets a target condition;
and if the verification result meets the target condition, determining the at least one credit business risk control rule as a target credit business risk control rule.
6. The risk control rule determination method of claim 5, wherein the step of validating the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain at least one target credit business risk control rule meeting a target condition further comprises:
if the verification result does not meet the target condition, screening a plurality of sample features included in at least one piece of credit service sample information in the training sample set to obtain a new training sample set, wherein the number of the credit service sample information included in the new training sample set is the same as the number of the credit service sample information included in the training sample set;
and segmenting the credit business sample information in the new training sample set based on the target decision tree algorithm, and taking the obtained at least one credit business risk control rule as at least one target credit business risk control rule meeting the target condition.
7. A risk control rule determination device applied to an electronic device, the risk control rule determination device comprising:
the credit business system comprises a sample information acquisition module, a credit business analysis module and a credit business analysis module, wherein the sample information acquisition module is used for acquiring a plurality of pieces of credit business sample information, each piece of credit business sample information is generated based on each history after credit business is processed, each piece of credit business sample information has label information, and the label information is used for representing whether a credit customer is overdue or not;
the sample information segmentation module is used for segmenting the plurality of pieces of credit business sample information to form a training sample set and a test sample set, wherein the training sample set comprises the plurality of pieces of credit business sample information, and the test sample set comprises the plurality of pieces of credit business sample information;
the sample information segmentation module is used for carrying out segmentation processing on the credit business sample information in the training sample set based on a target decision tree algorithm to obtain at least one credit business risk control rule;
and the control rule verification module is used for verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set so as to obtain at least one target credit business risk control rule meeting target conditions, wherein the at least one target credit business risk control rule is used for judging whether the overdue probability of the target credit business reaches a preset probability, and when the overdue probability is greater than the preset probability, the target credit business is determined not to pass the audit.
8. The risk control rule determination method of claim 7, wherein the sample information segmentation module comprises:
a decision tree parameter obtaining submodule, configured to obtain a first decision tree parameter and a second decision tree parameter that are predetermined, where the first decision tree parameter is used to represent a minimum degree of uncertainty of node partitioning of a decision tree during segmentation processing, the second decision tree parameter is used to represent a minimum number of samples required by node partitioning of the decision tree during segmentation processing, the first decision tree parameter is less than 0.001 and greater than 0.01, and the second decision tree parameter is less than 10 and greater than 100;
and the sample information segmentation submodule is used for carrying out segmentation processing on the credit business sample information in the training sample set based on the first decision tree parameter, the second decision tree parameter and a target decision tree algorithm to obtain at least one credit business risk control rule.
9. The risk control rule determination method of claim 7, wherein the control rule validation module comprises:
the control rule verification submodule is used for verifying the at least one credit business risk control rule based on the credit business sample information in the test sample set to obtain a verification result;
the verification result judgment submodule is used for judging whether the verification result meets the target condition;
and the control rule determining submodule is used for determining the at least one credit business risk control rule as a target credit business risk control rule when the verification result meets the target condition.
10. An electronic device, comprising:
a memory for storing a computer program;
a processor coupled to the memory for executing a computer program stored by the memory to implement the risk control rule determination method of any of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110621370.7A CN113256402A (en) | 2021-06-03 | 2021-06-03 | Risk control rule determination method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110621370.7A CN113256402A (en) | 2021-06-03 | 2021-06-03 | Risk control rule determination method and device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113256402A true CN113256402A (en) | 2021-08-13 |
Family
ID=77186388
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110621370.7A Pending CN113256402A (en) | 2021-06-03 | 2021-06-03 | Risk control rule determination method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113256402A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114625747A (en) * | 2022-05-13 | 2022-06-14 | 杭银消费金融股份有限公司 | Wind control updating method and system based on information security |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978680A (en) * | 2019-03-18 | 2019-07-05 | 杭州绿度信息技术有限公司 | A kind of air control method and system segmenting objective group's credit operation air control differentiation price |
| CN112163642A (en) * | 2020-10-30 | 2021-01-01 | 北京云从科技有限公司 | Wind control rule obtaining method, device, medium and equipment |
| CN112329874A (en) * | 2020-11-12 | 2021-02-05 | 京东数字科技控股股份有限公司 | Data service decision method and device, electronic equipment and storage medium |
| CN112613977A (en) * | 2020-12-17 | 2021-04-06 | 中国建设银行股份有限公司 | Personal credit loan admission credit granting method and system based on government affair data |
| CN112785086A (en) * | 2021-02-10 | 2021-05-11 | 中国工商银行股份有限公司 | Credit overdue risk prediction method and device |
-
2021
- 2021-06-03 CN CN202110621370.7A patent/CN113256402A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109978680A (en) * | 2019-03-18 | 2019-07-05 | 杭州绿度信息技术有限公司 | A kind of air control method and system segmenting objective group's credit operation air control differentiation price |
| CN112163642A (en) * | 2020-10-30 | 2021-01-01 | 北京云从科技有限公司 | Wind control rule obtaining method, device, medium and equipment |
| CN112329874A (en) * | 2020-11-12 | 2021-02-05 | 京东数字科技控股股份有限公司 | Data service decision method and device, electronic equipment and storage medium |
| CN112613977A (en) * | 2020-12-17 | 2021-04-06 | 中国建设银行股份有限公司 | Personal credit loan admission credit granting method and system based on government affair data |
| CN112785086A (en) * | 2021-02-10 | 2021-05-11 | 中国工商银行股份有限公司 | Credit overdue risk prediction method and device |
Non-Patent Citations (1)
| Title |
|---|
| 杨游云: "《Python广告数据挖掘与分析实战》", 机械工业出版社, pages: 122 - 123 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114625747A (en) * | 2022-05-13 | 2022-06-14 | 杭银消费金融股份有限公司 | Wind control updating method and system based on information security |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107040397B (en) | Service parameter acquisition method and device | |
| CN113138982B (en) | Big data cleaning method | |
| CN106611023B (en) | Method and device for detecting website access abnormality | |
| US20210374582A1 (en) | Enhanced Techniques For Bias Analysis | |
| CN108197795B (en) | Malicious group account identification method, device, terminal and storage medium | |
| CN113177700B (en) | Risk assessment method, system, electronic equipment and storage medium | |
| CN111652315A (en) | Model training method, object classification method, model training device, object classification device, electronic equipment and storage medium | |
| CN112036476A (en) | Data feature selection method and device based on two-classification service and computer equipment | |
| CN112348685A (en) | Credit scoring method, device, equipment and storage medium | |
| CN113256402A (en) | Risk control rule determination method and device and electronic equipment | |
| CN114519519A (en) | Method, device and medium for assessing enterprise default risk based on GBDT algorithm and logistic regression model | |
| CN112529319A (en) | Grading method and device based on multi-dimensional features, computer equipment and storage medium | |
| CN113255806A (en) | Sample feature determination method, sample feature determination device and electronic equipment | |
| CN115439079A (en) | Item classification method and device | |
| CN114581219A (en) | Anti-telecommunication network fraud early warning method and system | |
| CN114971887A (en) | Credit system operation monitoring method and device, storage medium and electronic equipment | |
| CN113239031A (en) | Big data denoising processing method | |
| CN113239381A (en) | Data security encryption method | |
| CN112835952A (en) | Stock index-based calculation analysis method and system | |
| CN113516165B (en) | Customer satisfaction judging method based on image pyramid matching posterior | |
| CN112907306B (en) | Customer satisfaction judging method and device | |
| CN113837863B (en) | Business prediction model creation method and device and computer readable storage medium | |
| CN113723522B (en) | Abnormal user identification method and device, electronic equipment and storage medium | |
| Enzinger et al. | Use Case—Fraud Detection Using Machine Learning Techniques | |
| US20240160696A1 (en) | Method for Automatic Detection of Pair-Wise Interaction Effects Among Large Number of Variables |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |