Method and apparatus for generating a set of risk control rules

Abstract

The embodiments of the present specification provide a method and device for generating a risk control rule set. In the method, a comprehensive risk value of each risk control rule for each risk type is calculated according to the weight of each risk type; the risk control rule with a comprehensive risk value greater than a first risk threshold is classified into a high-risk rule set; the risk control rule with a comprehensive risk value not greater than the first risk threshold is classified into a low-risk rule set; and the credit limit corresponding to the low-risk rule set is determined according to a specified risk indicator of the low-risk rule set and an inverse proportional relationship between the specified risk indicator and the credit limit.

Description

Technical Field

[0001] The embodiments in this specification relate to the field of computer technology, and more specifically, to a method and apparatus for generating risk control rule sets. Background Technology

[0002] In the financial sector, lending is a crucial asset business for financial institutions. Profits are generated by lending and recovering principal and interest, making it a primary means of revenue generation for these institutions. Because lending is outside the control of financial institutions, the risk of failing to recover principal and interest on time is significant. Therefore, financial institutions employ rigorous risk control management. Through risk control, they assess the risk level of each lending transaction. If the risk level is high, the applied-for loan can be blocked to avoid losses.

[0003] Currently, financial institutions implement risk control management by using risk control rule sets. These rule sets contain several risk control rules, each detecting only one type of risk. When a rule in the risk control rule set is matched, it indicates that the loan transaction is risky and can be blocked. Summary of the Invention

[0004] In view of the above, embodiments of this specification provide a method and apparatus for generating risk control rule sets. Through the technical solutions of these embodiments, multiple risk types are introduced using a comprehensive risk value, enabling the high-risk rule set to simultaneously manage multiple risk types. Furthermore, the medium- and low-risk rule sets obtained based on the comprehensive risk value correspond to appropriate credit limits, and the credit limits determined based on an inverse relationship can reduce the risk caused by customers matching the medium- and low-risk rule sets.

[0005] According to one aspect of the embodiments of this specification, a method for generating a risk control rule set is provided, comprising: calculating a comprehensive risk value for each risk control rule for each risk type based on the weight of each risk type, wherein each risk control rule is used for at least one risk type; classifying risk control rules with a comprehensive risk value greater than a first risk threshold into a high-risk rule set, the high-risk rule set being used to intercept customers with high risk; classifying risk control rules with a comprehensive risk value not greater than the first risk threshold into a medium-low risk rule set; and determining a credit limit corresponding to the medium-low risk rule set based on a specified risk indicator of the medium-low risk rule set and an inverse relationship between the specified risk indicator and the credit limit, wherein the specified risk indicator includes a comprehensive risk value and / or an improvement degree.

[0006] According to another aspect of the embodiments of this specification, an apparatus for generating a risk control rule set is also provided, comprising: a risk value calculation unit configured to calculate a comprehensive risk value for each risk control rule for each risk type according to the weight of each risk type, wherein each risk control rule is used for at least one risk type; a first risk control rule classification unit configured to classify risk control rules with a comprehensive risk value greater than a first risk threshold into a high-risk rule set, the high-risk rule set being used to intercept customers with high risk; a second risk control rule classification unit configured to classify risk control rules with a comprehensive risk value not greater than the first risk threshold into a medium-low risk rule set; and a credit limit determination unit configured to determine the credit limit corresponding to the medium-low risk rule set according to a specified risk indicator of the medium-low risk rule set and an inverse relationship between the specified risk indicator and the credit limit, wherein the specified risk indicator includes a comprehensive risk value and / or an improvement degree.

[0007] According to another aspect of the embodiments of this specification, an electronic device is also provided, comprising: at least one processor, a memory coupled to the at least one processor, and a computer program stored on the memory, wherein the at least one processor executes the computer program to implement the method for generating a risk control rule set as described above.

[0008] According to another aspect of the embodiments of this specification, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the method for generating a risk control rule set as described above.

[0009] According to another aspect of the embodiments of this specification, a computer program product is also provided, including a computer program that, when executed by a processor, implements the method for generating a risk control rule set as described above. Attached Figure Description

[0010] A further understanding of the nature and advantages of the embodiments described in this specification can be achieved by referring to the following accompanying drawings. In the drawings, similar components or features may have the same reference numerals.

[0011] Figure 1 A schematic diagram illustrating an application scenario of the risk control rule set generation method according to an embodiment of this specification is shown.

[0012] Figure 2 A flowchart illustrating an example of a method for generating a risk control rule set according to an embodiment of this specification is shown.

[0013] Figure 3 A flowchart illustrating another example of a method for generating a risk control rule set according to an embodiment of this specification is shown.

[0014] Figure 4 A flowchart illustrating another example of a method for generating a risk control rule set according to an embodiment of this specification is shown.

[0015] Figure 5 A block diagram of an example apparatus for generating a risk control rule set according to an embodiment of this specification is shown.

[0016] Figure 6 A block diagram of another example of an apparatus for generating a risk control rule set according to an embodiment of this specification is shown.

[0017] Figure 7 A block diagram of another example of an apparatus for generating a risk control rule set according to an embodiment of this specification is shown.

[0018] Figure 8 A block diagram of an electronic device for implementing a risk control rule set generation method according to an embodiment of this specification is shown. Detailed Implementation

[0019] The subject matter described herein will be discussed below with reference to exemplary embodiments. It should be understood that these embodiments are discussed merely to enable those skilled in the art to better understand and implement the subject matter described herein, and are not intended to limit the scope, applicability, or examples set forth in the claims. The function and arrangement of the elements discussed may be changed without departing from the scope of the embodiments described herein. Various processes or components may be omitted, substituted, or added as needed in the various examples. Furthermore, features described in some examples may be combined in other examples.

[0020] As used herein, the term "comprising" and its variations are open terms meaning "including but not limited to". The term "based on" means "at least partially based on". The terms "one embodiment" and "an embodiment" mean "at least one embodiment". The term "another embodiment" means "at least one other embodiment". The terms "first", "second", etc., may refer to different or the same objects. Other definitions, whether explicit or implicit, may be included below. Unless explicitly indicated by the context, the definition of a term shall remain consistent throughout the specification.

[0021] In this specification, the term "business model" refers to a machine learning model applied in a business scenario to provide business prediction services, such as machine learning models used for classification prediction, business risk prediction, etc. Examples of machine learning models may include, but are not limited to: linear regression models, logistic regression models, neural network models, decision tree models, support vector machines, etc. Examples of neural network models may include, but are not limited to: deep neural network (DNN) models, convolutional neural network (CNN) models, backpropagation (BP) neural networks, etc.

[0022] The specific implementation of a business model depends on the applicable business scenario. For example, if the business model is used to classify users, it is implemented as a user classification model. Accordingly, user classification prediction can be performed based on the user characteristic data of the users to be classified. If the business model is used to predict business risks associated with business transactions occurring on a business system, it is implemented as a business risk prediction model. Accordingly, business risk prediction can be performed based on the business transaction characteristic data of that transaction.

[0023] With the development of artificial intelligence technology, machine learning technology is widely used as a business model in various business application scenarios to provide various business prediction services, such as classification prediction and business risk prediction. For example, business models have wide applications in fields such as financial fraud, recommendation systems, and image recognition. To achieve better model performance, more training data is needed to train the business model. In application areas such as healthcare and finance, different enterprises or institutions have different data samples. Once these data are jointly trained, the accuracy of the business model will be greatly improved, thereby bringing huge economic benefits to enterprises.

[0024] In the financial sector, lending is a crucial asset business for financial institutions. Profits are generated by lending and recovering principal and interest, making it a primary means of revenue generation for these institutions. Because lending is outside the control of financial institutions, the risk of failing to recover principal and interest on time is significant. Therefore, financial institutions employ rigorous risk control management. Through risk control, they assess the risk level of each lending transaction. If the risk level is high, the applied-for loan can be blocked to avoid losses.

[0025] Currently, financial institutions implement risk control management by using risk control rule sets. These rule sets contain several risk control rules, each detecting only one type of risk. When a rule in the risk control rule set is matched, it indicates that the loan transaction is risky and can be blocked.

[0026] However, in current risk control management, each rule set applied only covers one type of risk, and all risk rules are simply divided into high-risk rules and other risk rules. Credit transactions that match high-risk rules are blocked, while credit transactions that do not match high-risk rules are approved. This risk control approach cannot take into account multiple risk types and cannot perform risk control more precisely.

[0027] In view of the above, embodiments of this specification provide a method and apparatus for generating risk control rule sets. In this method, a comprehensive risk value for each risk control rule is calculated based on the weight of each risk type; risk control rules with a comprehensive risk value greater than a first risk threshold are classified into a high-risk rule set; risk control rules with a comprehensive risk value not greater than the first risk threshold are classified into a medium-low risk rule set; and the credit limit corresponding to the medium-low risk rule set is determined based on a specified risk indicator of the medium-low risk rule set and the inverse relationship between the specified risk indicator and the credit limit. Through the technical solution of the embodiments of this specification, multiple risk types are introduced using comprehensive risk values, so that the high-risk rule set can simultaneously perform risk control for multiple risk types. Furthermore, the medium-low risk rule set obtained based on the comprehensive risk value corresponds to an appropriate credit limit, and the credit limit determined based on the inverse relationship can reduce the risk caused by customers matching the medium-low risk rule set.

[0028] Figure 1 A schematic diagram of an application scenario 100 of the risk control rule set generation method according to an embodiment of this specification is shown.

[0029] like Figure 1 As shown, users can apply to financial institutions for relevant financial services, such as credit services. After receiving the user's application, the financial institution needs to conduct risk control management on the user and the applied financial service. At this time, the user becomes the customer group for risk control by the financial institution. One way for financial institutions to perform risk control management is by using risk control devices.

[0030] Financial institutions can send customer information and business information to the risk control device. Customer information can include the customer's identity information, credit information, asset information and historical business information, while business information includes loan amount, loan term, etc.

[0031] The risk control device uses a risk control rule base to manage customer and business information. This rule base stores various risk control rules. These rules can be stored as sets, such as a high-risk rule set including high-risk rules, and a medium-to-low-risk rule set including medium-to-low-risk rules. During risk management, customer and business information are matched against rules in the rule base. When a customer or business information matches a risk control rule, that rule is considered a match. If a rule is matched, the processing method corresponding to the rule set to which it belongs is executed. For example, if the matched rule belongs to the high-risk rule set, the corresponding processing method includes blocking, i.e., blocking the customer group and the applied business.

[0032] After each risk control management operation, the risk control device generates corresponding historical risk control data. This data includes customer and business information, the triggered risk control rules, and the corresponding processing methods. The generated historical risk control data is stored in a database, which can also store sample data and other data. Both the historical and sample data in the database can be used as samples to iteratively optimize the risk control rules.

[0033] The risk control rule set generation device can optimize risk control rules in the risk control rule base to generate an optimized risk control rule set. The optimized rule set is stored in the risk control rule base for easy access by the risk control device, thereby improving the accuracy of risk control when the device performs risk management. Furthermore, the risk control rule set generation device can also retrieve relevant historical risk control data and sample data from the database, allowing it to optimize the risk control rules in the risk control rule base using the data from the database.

[0034] Figure 2 A flowchart of an example 200 of a method for generating a risk control rule set according to an embodiment of this specification is shown.

[0035] like Figure 2 As shown in Figure 210, the comprehensive risk value of each risk control rule for each risk type can be calculated based on the weight of each risk type.

[0036] In the embodiments described in this specification, the various risk control rules used to generate the risk control rule set can be applied to the same financial business, such as risk control rules applied to credit business. The following explanation uses credit business as an example. In credit business, the types of risks that exist can include fraud risk and credit risk, etc.

[0037] Each risk control rule can be applied to at least one risk type; that is, each risk control rule can detect risks of at least one risk type. For example, when a risk control rule is applied to fraud risk, then that risk control rule can be used to detect the existing fraud risk.

[0038] In the embodiments of this specification, each risk type has a corresponding weight, which represents the degree of risk control attention paid to the corresponding risk type. The larger the weight of a risk type, the higher the degree of risk control attention paid to that risk type, and thus the higher the probability of it being subject to risk control; the smaller the weight of a risk type, the lower the degree of risk control attention paid to that risk type, and thus the lower the probability of it being subject to risk control.

[0039] In one example, the weight of each risk type can be determined based on at least one of the following factors: frequency of occurrence, severity of harm, and amount of loss. The frequency of occurrence and amount of loss for each risk type can be obtained statistically from historical risk data. The higher the severity of harm caused by a risk type, the greater the weight of that risk type; the higher the frequency of occurrence of a risk type, the greater the weight of that risk type; and the greater the amount of loss caused by a risk type, the greater the weight of that risk type.

[0040] In one method of calculating the overall risk value, for each risk control rule, the overall risk value of that rule can be calculated based on the weights corresponding to the various risk types to which the rule applies. In one example, the sum of the weights corresponding to the various risk types to which the rule applies can be used to determine the overall risk value of that rule.

[0041] The overall risk value of each risk control rule is composed of the weights of the risk types applied to that rule. When a risk control rule applies to multiple risk types, the corresponding overall risk value integrates these multiple risk types, thus reflecting the integrated risk values. Furthermore, the overall risk value of each risk control rule also reflects the degree to which that rule is subject to risk control; the higher the overall risk value, the higher the degree to which the corresponding risk control rule is subject to risk control.

[0042] In the embodiments of this specification, the magnitude of the overall risk value of each risk control rule is related to the number of risk types applied to that rule. The more risk types applied by a risk control rule, the more weights are used to calculate the overall risk value, resulting in a larger overall risk value; conversely, the fewer risk types applied, the smaller the overall risk value. Furthermore, the magnitude of the overall risk value of each risk control rule is also related to the magnitude of its weights. The larger the weights used to calculate the overall risk value, the larger the overall risk value; conversely, the smaller the weights used, the smaller the overall risk value. Therefore, the overall risk value of each risk control rule, while taking into account multiple risk types, also reflects the degree of risk control applied to the rule.

[0043] In 220, risk control rules with a comprehensive risk value greater than the first risk threshold can be classified into the high-risk rule set.

[0044] In the embodiments described in this specification, a first risk threshold can be specified. A high-risk rule set is used to intercept customer groups with high risk, and each high-risk rule in the high-risk rule set is used to detect risks of high-risk types. When a customer group in a credit business matches any high-risk rule in the high-risk rule set, that customer group can be determined as a high-risk customer group, and thus, that customer group can be intercepted.

[0045] In 230, risk control rules with a comprehensive risk value not exceeding the first risk threshold can be classified into the low-to-medium risk rule set.

[0046] In the embodiments described in this specification, the risks detected by each medium-to-low risk rule in the set of medium-to-low risk rules are medium-to-low risk. Therefore, when any medium-to-low risk rule in the set of medium-to-low risk rules is hit, it indicates the existence of medium-to-low risk. In credit business, for customers with medium-to-low risk, they can be allowed to pass through without being blocked.

[0047] It should be noted that the execution order of operations 220 and 230 is not restricted. Figure 2 The execution order shown is for illustrative purposes only.

[0048] It should also be noted that the above-mentioned high-risk rule sets and medium-to-low-risk rule sets can be used as business models.

[0049] In 240, the credit limit corresponding to the medium- and low-risk rule set can be determined based on the specified risk indicators of the medium- and low-risk rule set and the inverse relationship between the specified risk indicators and the credit limit.

[0050] In the embodiments described in this specification, the specified risk indicators may include a comprehensive risk value and / or lift. Lift (i.e., the ability of a risk rule or set of risk rules to predict a target is compared to the ratio of a target randomly selected without utilizing that risk rule or set of risk rules.) Lift can be calculated using the following formula:

[0051]

[0052] Where Lift represents the lift, TP represents a true positive, FP represents a false positive, TN represents a true negative, and FN represents a false negative.

[0053] Specifying risk indicators can be used to indicate the degree of risk; the higher the value of the specified risk indicator, the higher the degree of risk. Therefore, the specified risk indicator can have an inverse relationship with the credit limit: a higher specified risk indicator value results in a lower credit limit, and vice versa. In one example, when the specified risk indicator is an increase in risk level, the increase in risk level and the credit limit are inversely proportional. In another example, when the specified risk indicator is a comprehensive risk value, the comprehensive risk value and the credit limit are inversely proportional.

[0054] The specified risk indicator for a low-to-medium risk rule set can be applied to the entire low-to-medium risk rule set. When the specified risk indicator is lift, it can be the lift of the entire low-to-medium risk rule set. When the specified risk indicator is a composite risk value, it is the composite risk value of the entire low-to-medium risk rule set. The composite risk value of a low-to-medium risk rule set can be obtained from the composite risk values ​​corresponding to each low-to-medium risk rule in the set. In one example, the average of the composite risk values ​​corresponding to each low-to-medium risk rule in the set can be used as the composite risk value for the entire low-to-medium risk rule set.

[0055] Once the credit limit corresponding to the set of medium- and low-risk rules is determined, the credit limit that can be provided for a credit business that matches any medium- or low-risk rule in the set of medium- and low-risk rules is the determined credit limit.

[0056] In one example, the low-to-medium risk rules within the low-to-medium risk rule set can be tiered to obtain multiple subsets of low-to-medium risk rules at different tiers. In this example, the specified risk indicator used can include the specified risk indicator corresponding to each subset of low-to-medium risk rules, and the determined credit limit can include the credit limit for each subset of low-to-medium risk rules. For this example, see below. Figure 3 The examples will be explained in detail below.

[0057] In the embodiments described in this specification, the high-risk rule set and the medium-to-low-risk rule set with credit limits can be used as the generated risk control rule set. The generated risk control rule set can be stored in the rule base for easy access by the risk control device to manage the credit business of various customer groups.

[0058] The technical solutions provided in the embodiments of this specification can generate high-risk rule sets and medium-to-low-risk rule sets with credit limits. The high-risk rule sets can block high-risk customers and credit transactions, while the medium-to-low-risk rule sets can allow low-limit approval for medium-to-low-risk customers and credit transactions; that is, they allow these customers and credit transactions to proceed, but with relatively low credit limits. This can reduce losses caused by medium-to-low-risk transactions to a certain extent.

[0059] Figure 3 A flowchart of another example 300 of a method for generating a risk control rule set according to an embodiment of this specification is shown.

[0060] like Figure 3 As shown in Figure 310, the comprehensive risk value of each risk control rule for each risk type can be calculated based on the weight of each risk type.

[0061] In 320, risk control rules with a comprehensive risk value greater than the first risk threshold can be classified into the high-risk rule set.

[0062] In 330, risk control rules with a comprehensive risk value not exceeding the first risk threshold can be classified into the low-to-medium risk rule set.

[0063] It should be noted that the operations from 310 to 330 are related to... Figure 2 The operations for 210 to 230 in the example are the same; you can refer to the operation instructions for 210 to 230.

[0064] In 340, each medium- and low-risk rule can be categorized based on its comprehensive risk value within the medium- and low-risk rule set to obtain subsets of medium- and low-risk rules at different levels.

[0065] In this example, the number of risk tiers can be specified, as can the range of the comprehensive risk value corresponding to each tier. In one example, the number of risk tiers and the range of the comprehensive risk value corresponding to each tier can be determined based on the various low-to-medium risk rules to be tiered, to ensure that each tier includes low-to-medium risk rules, and also to ensure that the number of low-to-medium risk rules in each tier is relatively even.

[0066] The risk tiers can be divided according to the overall risk value, either from highest to lowest or from lowest to highest. This results in different risk value ranges for different tiers, and these ranges do not overlap. Furthermore, each tier corresponds to a subset of low-to-medium risk rules, and the low-to-medium risk rules within that subset are then further subdivided into their respective subsets.

[0067] In 350, the credit limit corresponding to each subset of medium- and low-risk rules can be determined based on the specified risk indicators corresponding to each subset of medium- and low-risk rules and the inverse relationship between the specified risk indicators and the credit limit.

[0068] In this example, a corresponding credit limit can be determined separately for each subset of medium- and low-risk rules. The process of determining the credit limit for different subsets of medium- and low-risk rules does not affect each other.

[0069] For each subset of medium- and low-risk rules, the designated risk indicator corresponding to the subset can be determined first. Then, based on the inverse relationship between the designated risk indicator and the credit limit, the credit limit corresponding to the subset of medium- and low-risk rules can be determined.

[0070] In one example, when the inverse relationship is a function expression, the credit limit is the dependent variable and the specified risk indicator is the independent variable. Therefore, the determined specified risk indicator can be substituted into the function expression to calculate the credit limit. The calculated credit limit is the credit limit corresponding to the subset of medium and low risk rules.

[0071] In another example, when the inverse relationship is a direct correspondence between two factors, each specified risk indicator range corresponds to a credit limit. Different specified risk indicator ranges correspond to different credit limits, and the larger the specified risk indicator range, the smaller the corresponding credit limit. In this example, firstly, the specified risk indicator range to which the specified risk indicator belongs is determined. Then, the credit limit corresponding to that specified risk indicator range is determined in the correspondence relationship. This credit limit is the credit limit corresponding to the subset of low-to-medium risk rules.

[0072] In one example, the specified risk indicator includes the lift level. For each subset of low-to-medium risk rules, the lift level corresponding to that subset can be determined using data samples. Then, based on the lift level corresponding to that subset of low-to-medium risk rules and the inverse relationship between the lift level and the credit limit, the corresponding credit limit can be determined. The data samples used can include historical risk control data based on credit business, risk control data generated in other similar application scenarios, etc., and each sample included in the data sample can be labeled.

[0073] In another example, the specified risk indicator includes a comprehensive risk value. For each subset of low-to-medium risk rules, the comprehensive risk value for the rule set corresponding to that subset can be calculated based on the comprehensive risk values ​​corresponding to each low-to-medium risk rule within that subset. In one example, the sum of the comprehensive risk values ​​corresponding to each low-to-medium risk rule can be used to determine the corresponding comprehensive risk value for the rule set. In another example, the average of the comprehensive risk values ​​corresponding to each low-to-medium risk rule can be calculated, and the resulting mean can be used to determine the corresponding comprehensive risk value for the rule set.

[0074] Then, based on the comprehensive risk value of the rule set corresponding to the low-to-medium risk rule subset and the inverse relationship between the comprehensive risk value and the credit limit, the credit limit corresponding to the low-to-medium risk rule subset is determined.

[0075] The technical solution described above categorizes low-to-medium risk rules into tiers, assigns corresponding credit limits to subsets of low-to-medium risk rules within each tier, and provides different credit limits to customer groups at different risk levels within the low-to-medium risk range, thereby enabling more precise risk control.

[0076] Figure 4 A flowchart of another example 400 of a method for generating a risk control rule set according to an embodiment of this specification is shown.

[0077] like Figure 4 As shown in Figure 410, the comprehensive risk value of each risk control rule for each risk type can be calculated based on the weight of each risk type.

[0078] It should be noted that the operation of 410 is related to... Figure 2 The operation for 210 in the example is the same; you can refer to the operation instructions for 210.

[0079] In 420, risk control rules can be divided into medium-high risk control rules and low-risk risk control rules based on the lift level corresponding to each rule.

[0080] In this example, data samples can be used to test each risk control rule individually to determine the lift level corresponding to each rule. In one example, a first lift level threshold can be set, which can be specified. Risk control rules with a lift level greater than the first lift level threshold are identified as medium-to-high risk rules, while those with a lift level not greater than the first lift level threshold are identified as low-risk rules.

[0081] In 430, risk control rules whose comprehensive risk value is greater than the first risk threshold can be classified into the high-risk rule set.

[0082] In 440, medium-risk risk control rules can be selected from the medium- and high-risk risk control rules, where the comprehensive risk value is not greater than the first risk threshold.

[0083] By using operations 430 and 440, the medium-to-high risk control rules can be further divided into high-risk rules and medium-risk rules.

[0084] In 450, the selected medium-risk and low-risk risk control rules can be categorized into the medium-low risk rule set.

[0085] In 460, the credit limit corresponding to the medium- and low-risk rule set can be determined based on the specified risk indicators of the medium- and low-risk rule set and the inverse relationship between the specified risk indicators and the credit limit.

[0086] It should be noted that the operation of 460 is different from... Figure 2 The operation is the same as in example 240; you can refer to the operation instructions for 240.

[0087] In one example, after dividing the various risk control rules into medium-to-high risk control rules and low-risk risk control rules, the low-risk rules can be further categorized according to their comprehensive risk values ​​to obtain subsets of low-risk rules at different levels.

[0088] The grading method in this example can be referenced. Figure 3 The tiering method for operation 340 in the example. The number of tiers and the corresponding comprehensive risk value range for low-risk rules can be the same as or different from the number of tiers and the corresponding comprehensive risk value range for medium-to-low-risk rules in operation 340.

[0089] In this example, the number of tiers for low-risk rules and the corresponding comprehensive risk value range can be determined based on the various low-risk rules to be tiered, so as to ensure that each tier includes low-risk rules and that the number of low-risk rules in each tier is relatively even.

[0090] In the process of incorporating rules into the low-to-medium risk rule set, the various low-risk rule subsets and the medium-risk risk control rule subset composed of medium-risk rules can be merged into a low-to-medium risk rule set. Then, based on the designated risk indicators corresponding to each low-risk rule subset and medium-risk risk control rule subset within the low-to-medium risk rule set, and the inverse relationship between the designated risk indicators and the credit limit, the corresponding credit limit for each low-risk rule subset and medium-risk risk control rule subset is determined.

[0091] In this example, the specified risk indicators include the overall risk value and / or the enhancement level. When the specified risk indicator includes the overall risk value, for each low-risk rule subset, the first subset overall risk value corresponding to that low-risk rule subset can be calculated based on the overall risk values ​​corresponding to each low-risk rule in that subset. For the medium-risk risk control rule subset, the second subset overall risk value corresponding to that medium-risk rule subset can be calculated based on the overall risk values ​​corresponding to each medium-risk rule in that subset. Then, based on the first subset overall risk value corresponding to each low-risk rule subset, the second subset overall risk value corresponding to each medium-risk rule subset, and the inverse relationship between the overall risk value and the credit limit, the credit limit corresponding to each low-risk rule subset and the medium-risk risk control rule subset is determined.

[0092] When the specified risk indicator includes lift, the first lift corresponding to each low-risk rule subset can be determined using data samples, and the second lift corresponding to each medium-risk risk control rule subset can be determined using data samples. Then, based on the first lift corresponding to each low-risk rule subset, the second lift corresponding to each medium-risk rule subset, and the inverse relationship between lift and credit limit, the credit limit corresponding to each low-risk rule subset and medium-risk risk control rule subset can be determined.

[0093] The scheme in this example categorizes low-risk rules into tiers and assigns corresponding credit limits to the subsets of low-risk and medium-risk rules within each tier. This allows for more precise risk control by providing different credit limits to customer groups at different levels of low-risk risk.

[0094] Figure 5 A block diagram of an example of an apparatus for generating a risk control rule set (hereinafter referred to as risk control rule set generation apparatus 500) according to an embodiment of this specification is shown.

[0095] like Figure 5 As shown, the risk control rule set generation device 500 includes: a risk value calculation unit 510, a first risk control rule classification unit 520, a second risk control rule classification unit 530, and a credit limit determination unit 540.

[0096] The risk value calculation unit 510 can be configured to calculate the comprehensive risk value of each risk control rule for each risk type according to the weight of each risk type, wherein each risk control rule is used for at least one risk type.

[0097] In one example, the weight of each risk type is determined based on at least one of the following factors: frequency of occurrence, degree of harm caused, and amount of loss.

[0098] The first risk control rule classification unit 520 can be configured to classify risk control rules with a comprehensive risk value greater than a first risk threshold into a high-risk rule set, which is used to intercept customers with high risk.

[0099] The second risk control rule classification unit 530 can be configured to categorize risk control rules with a comprehensive risk value not exceeding the first risk threshold into the low-to-medium risk rule set. It should be noted that the first risk control rule classification unit 520 and the second risk control rule classification unit 530 can be the same unit or different units.

[0100] The credit limit determination unit 540 can be configured to determine the credit limit corresponding to the medium- and low-risk rule set based on the specified risk indicators of the medium- and low-risk rule set and the inverse relationship between the specified risk indicators and the credit limit. The specified risk indicators include the comprehensive risk value and / or the enhancement degree.

[0101] Figure 6 A block diagram of another example of an apparatus for generating a risk control rule set (hereinafter referred to as risk control rule set generation apparatus 600) according to an embodiment of this specification is shown.

[0102] like Figure 6 As shown, the risk control rule set generation device 600 includes: a risk value calculation unit 510, a first risk control rule classification unit 520, a second risk control rule classification unit 530, a credit limit determination unit 540, and a third risk control rule classification unit 550.

[0103] The risk value calculation unit 510 and the credit limit determination unit 540 can be referenced. Figure 5 The operating instructions are in the middle.

[0104] The third risk control rule classification unit 550 can be configured to divide each risk control rule into medium-high risk control rules and low risk control rules according to the lift degree corresponding to each risk control rule.

[0105] The first risk control rule classification unit 520 can also be configured to classify risk control rules with a comprehensive risk value greater than the first risk threshold in the medium- and high-risk risk control rules into the high-risk risk rule set.

[0106] The second risk control rule classification unit 530 can also be configured to select medium-risk risk control rules from the medium- and high-risk risk control rules whose comprehensive risk value is not greater than the first risk threshold; and to classify the selected medium-risk risk control rules and low-risk risk control rules into the medium- and low-risk risk rule set.

[0107] In one example, the risk control rule set generation device 500 may further include a risk control rule grading unit, which can be configured to grade the low-risk rules according to the comprehensive risk value of each low-risk rule to obtain subsets of low-risk rules at different tiers.

[0108] The second risk control rule classification unit 530 can also be configured to combine the obtained subsets of low-risk rules and the subset of medium-risk rules composed of medium-risk rules into a set of medium-low-risk rules.

[0109] The credit limit determination unit 540 can also be configured to determine the credit limit corresponding to each low-risk rule subset and medium-risk risk control rule subset based on the specified risk indicators corresponding to each low-risk rule subset and medium-risk risk control rule subset in the low- and medium-risk rule set, as well as the inverse relationship between the specified risk indicators and the credit limit.

[0110] Figure 7 A block diagram of another example of an apparatus for generating a risk control rule set (hereinafter referred to as risk control rule set generation apparatus 700) according to an embodiment of this specification is shown.

[0111] like Figure 7 As shown, the risk control rule set generation device 700 includes: a risk value calculation unit 510, a first risk control rule classification unit 520, a second risk control rule classification unit 530, a credit limit determination unit 540, and a risk control rule grading unit 560.

[0112] The risk value calculation unit 510, the first risk control rule classification unit 520, and the second risk control rule classification unit 530 can be referred to Figure 5 The operating instructions are in the middle.

[0113] The risk control rule grading unit 560 can be configured to grade each low-risk rule according to the comprehensive risk value of each low-risk rule in the low-risk rule set, so as to obtain a subset of low-risk rules at different grading levels.

[0114] The credit limit determination unit 540 can also be configured to determine the credit limit corresponding to each subset of medium- and low-risk rules based on the specified risk indicators corresponding to each subset of medium- and low-risk rules and the inverse relationship between the specified risk indicators and the credit limit.

[0115] In one example, the specified risk indicator includes the lift level. The credit limit determination unit 540 can also be configured to: determine the lift level corresponding to each subset of medium- and low-risk rules using data samples; and determine the credit limit corresponding to each subset of medium- and low-risk rules based on the lift level corresponding to each subset of medium- and low-risk rules and the inverse relationship between the lift level and the credit limit.

[0116] In one example, the specified risk indicator includes a comprehensive risk value. The credit limit determination unit 540 can also be configured to: calculate a comprehensive risk value for each subset of medium- and low-risk rules based on the comprehensive risk value corresponding to each medium- and low-risk rule in that subset; and determine the credit limit corresponding to each subset of medium- and low-risk rules based on the comprehensive risk value of the rule set corresponding to each subset of medium- and low-risk rules and the inverse relationship between the comprehensive risk value and the credit limit.

[0117] Reference above Figures 1 to 7 Embodiments of the method and apparatus for generating risk control rule sets according to the embodiments of this specification have been described.

[0118] The apparatus for generating risk control rule sets in the embodiments of this specification can be implemented in hardware, software, or a combination of hardware and software. Taking software implementation as an example, as a logical device, it is formed by the processor of its host device reading the corresponding computer program instructions from the memory into memory and executing them. In the embodiments of this specification, the apparatus for generating risk control rule sets can, for example, be implemented using electronic devices.

[0119] Figure 8 A block diagram of an electronic device 800 for implementing a risk control rule set generation method according to an embodiment of this specification is shown.

[0120] like Figure 8 As shown, the electronic device 800 may include at least one processor 810, a memory (e.g., non-volatile memory) 820, a RAM 830, and a communication interface 840, and the at least one processor 810, memory 820, RAM 830, and communication interface 840 are connected together via a bus 850. The at least one processor 810 executes at least one computer-readable instruction (i.e., the elements implemented in software above) stored or encoded in the memory.

[0121] In one embodiment, computer-executable instructions are stored in a memory, which, when executed, cause at least one processor 810 to: calculate the comprehensive risk value of each risk control rule for each risk type according to the weight of each risk type; classify risk control rules with a comprehensive risk value greater than a first risk threshold into a high-risk rule set; classify risk control rules with a comprehensive risk value not greater than the first risk threshold into a medium-low-risk rule set; and determine the credit limit corresponding to the medium-low-risk rule set according to a specified risk indicator of the medium-low-risk rule set and the inverse relationship between the specified risk indicator and the credit limit.

[0122] It should be understood that the computer-executable instructions stored in memory, when executed, cause at least one processor 810 to perform the above-described combinations in the various embodiments of this specification. Figure 1-7 The description includes various operations and functions.

[0123] According to one embodiment, a program product, such as a machine-readable medium, is provided. The machine-readable medium may have instructions (i.e., the elements implemented in software as described above), which, when executed by a machine, cause the machine to perform the above-described combinations of the various embodiments of this specification. Figure 1-7 The description includes various operations and functions.

[0124] Specifically, a system or apparatus equipped with a readable storage medium may be provided, on which software program code implementing the functions of any of the embodiments described above is stored, and the computer or processor of the system or apparatus can read and execute the instructions stored in the readable storage medium.

[0125] In this case, the program code itself, which can be read from a readable medium, can perform the functions of any of the above embodiments. Therefore, the machine-readable code and the readable storage medium storing the machine-readable code constitute a part of the present invention.

[0126] The computer program code required for the operation of each part of this manual can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB, .NET, and Python; conventional procedural programming languages ​​such as C, Visual Basic 2003, Perl, COBOL 2002, PHP, and ABAP; dynamic programming languages ​​such as Python, Ruby, and Groovy; or other programming languages. This program code can run on the user's computer, or as a standalone software package on the user's computer, or partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer via any network, such as a local area network (LAN) or wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service, such as Software as a Service (SaaS).

[0127] Examples of readable storage media include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD-RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer or the cloud via a communication network.

[0128] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0129] Not all steps and units in the above process and system structure diagrams are mandatory; some steps or units can be omitted as needed. The execution order of each step is not fixed and can be determined as required. The device structure described in the above embodiments can be a physical structure or a logical structure. That is, some units may be implemented by the same physical entity, or some units may be implemented by multiple physical entities, or they may be jointly implemented by certain components in multiple independent devices.

[0130] The term "exemplary" as used throughout this specification means "serving as an example, instance, or illustration" and does not imply that it is "preferred" or "advantageous" over other embodiments. Detailed descriptions are included for the purpose of providing an understanding of the described techniques. However, these techniques may be practiced without these detailed descriptions. In some instances, well-known structures and apparatuses are shown in block diagram form to avoid obscuring the concepts of the described embodiments.

[0131] The optional embodiments of the present specification have been described in detail above with reference to the accompanying drawings. However, the embodiments of the present specification are not limited to the specific details in the above embodiments. Within the scope of the technical concept of the embodiments of the present specification, various simple modifications can be made to the technical solutions of the embodiments of the present specification, and these simple modifications all fall within the protection scope of the embodiments of the present specification.

[0132] The foregoing description of this specification is provided to enable any person skilled in the art to implement or use the content of this specification. Various modifications to the content of this specification will be apparent to those skilled in the art, and the general principles defined herein can be applied to other variations without departing from the scope of protection of this specification. Therefore, this specification is not limited to the examples and designs described herein, but is consistent with the widest scope of the principles and novel features disclosed herein.

Claims

1. A method for generating a risk control rule set, comprising: The comprehensive risk value of each risk control rule for each risk type is calculated based on the weight of each risk type, wherein each risk control rule is used for at least one risk type; Based on the lift degree corresponding to each risk control rule, the risk control rules are divided into medium-high risk control rules and low risk control rules. Risk control rules with a comprehensive risk value greater than the first risk threshold in the medium-to-high-risk risk control rules are classified into the high-risk risk rule set; the high-risk risk rule set is used to intercept customer groups with high risk. From the medium-to-high-risk risk control rules, medium-risk risk control rules with a comprehensive risk value not exceeding the first risk threshold are selected, and these selected medium-risk risk control rules are grouped with the low-risk risk control rules into a medium-to-low-risk risk rule set; and The credit limit corresponding to the medium- and low-risk rule set is determined based on the specified risk indicators of the medium- and low-risk rule set and the inverse relationship between the specified risk indicators and the credit limit. The specified risk indicators include the comprehensive risk value and / or the enhancement degree.

2. The method of claim 1, further comprising: The low-risk risk control rules are categorized into different tiers based on their comprehensive risk values ​​to obtain subsets of low-risk rules at different tiers. The selected medium-risk risk control rules and the low-risk risk control rules are grouped into the medium-low risk rule set, including: The obtained subsets of low-risk rules and the subset of medium-risk rules composed of the medium-risk rules are combined into a set of medium-low-risk rules; and The credit limit corresponding to the low-to-medium risk rule set is determined based on the specified risk indicators of the low-to-medium risk rule set and the inverse relationship between the specified risk indicators and the credit limit, including: Based on the designated risk indicators corresponding to each low-risk rule subset and the medium-risk risk control rule subset in the low-to-medium risk rule set, and the inverse relationship between the designated risk indicators and the credit limit, the credit limit corresponding to each low-risk rule subset and the medium-risk risk control rule subset is determined.

3. The method of claim 1, further comprising: The low- and medium-risk rules are categorized into different tiers based on their comprehensive risk values ​​within the low- and medium-risk rule set. as well as The credit limit corresponding to the low-to-medium risk rule set is determined based on the specified risk indicators of the low-to-medium risk rule set and the inverse relationship between the specified risk indicators and the credit limit, including: The credit limit corresponding to each subset of medium- and low-risk rules is determined based on the specified risk indicators corresponding to each subset of medium- and low-risk rules and the inverse relationship between the specified risk indicators and the credit limit.

4. The method of claim 3, wherein, The specified risk indicators include the degree of improvement. The credit limit corresponding to each subset of medium- and low-risk rules is determined based on the specified risk indicators corresponding to each subset of medium- and low-risk rules and the inverse relationship between the specified risk indicators and the credit limit, including: The lift of each subset of low-to-medium risk rules is determined using data samples; and Based on the lift degree corresponding to each subset of medium- and low-risk rules and the inverse relationship between the lift degree and the credit limit, the credit limit corresponding to each subset of medium- and low-risk rules is determined.

5. The method of claim 3, wherein, The specified risk indicators include the comprehensive risk value. The credit limit corresponding to each subset of medium- and low-risk rules is determined based on the specified risk indicators corresponding to each subset of medium- and low-risk rules and the inverse relationship between the specified risk indicators and the credit limit, including: For each subset of medium- and low-risk rules, calculate the comprehensive risk value of the rule set corresponding to that subset based on the comprehensive risk value corresponding to each medium- and low-risk rule in that subset; and The credit limit corresponding to each subset of medium- and low-risk rules is determined based on the comprehensive risk value of the rule set corresponding to each subset of medium- and low-risk rules and the inverse relationship between the comprehensive risk value and the credit limit.

6. The method of claim 1, wherein, The weights of each risk type are determined based on at least one of the following factors: frequency of occurrence, degree of harm caused, and amount of loss.

7. An apparatus for generating a risk control rule set, comprising: The risk value calculation unit is configured to calculate the comprehensive risk value of each risk control rule for each risk type according to the weight of each risk type, wherein each risk control rule is used for at least one risk type; The third risk control rule classification unit is configured to divide each risk control rule into medium-high risk risk control rules and low risk risk control rules according to the lift degree of each risk control rule. The first risk control rule classification unit is configured to classify risk control rules whose comprehensive risk value is greater than the first risk threshold into the high-risk rule set; the high-risk rule set is used to intercept customers with high risk. The second risk control rule classification unit is configured to filter out medium-risk risk control rules whose comprehensive risk value is not greater than the first risk threshold from the medium-high risk risk control rules, and to classify the filtered medium-risk risk control rules and the low-risk risk control rules into a medium-low risk rule set; and The credit limit determination unit is configured to determine the credit limit corresponding to the medium-low risk rule set based on the specified risk indicators of the medium-low risk rule set and the inverse relationship between the specified risk indicators and the credit limit, wherein the specified risk indicators include a comprehensive risk value and / or an improvement degree.

8. The apparatus of claim 7 further includes a risk rule classification unit. The risk rule grading unit is configured to grade each low-risk rule according to the comprehensive risk value of each low-risk rule in the low-risk rule set, so as to obtain a subset of low-risk rules at different levels. The credit limit determination unit is further configured to determine the credit limit corresponding to each subset of medium- and low-risk rules based on the specified risk indicators corresponding to each subset of medium- and low-risk rules and the inverse relationship between the specified risk indicators and the credit limit.

9. An electronic device, comprising: At least one processor, a memory coupled to said at least one processor, and a computer program stored on said memory, said at least one processor executing said computer program to implement the method as described in any one of claims 1-6.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as described in any one of claims 1-6.

11. A computer program product comprising a computer program that, when executed by a processor, implements the method as described in any one of claims 1-6.

Images