WO2021169212A1

WO2021169212A1 - Risk control model construction method and apparatus, risk control checking method and apparatus, and computer device

Info

Publication number: WO2021169212A1
Application number: PCT/CN2020/111948
Authority: WO
Inventors: 万忠伟; 甘丽婷
Original assignee: 平安科技（深圳）有限公司
Priority date: 2020-02-26
Filing date: 2020-08-28
Publication date: 2021-09-02
Also published as: CN111260257A

Abstract

A risk control model construction method and apparatus, a risk control checking method and apparatus, a computer device, and a computer-readable storage medium, which belong to the technical field of artificial intelligence. The risk control model construction method comprises: when a preset operation for constructing a risk control model is received, displaying a preset interface coding table, a query result coding table and a risk control result coding table by means of a list page (S101); receiving, by means of the list page, a selection instruction used for determining a target interface code from the interface coding table, determining a target query result from the query result coding table and determining a target risk control result from the risk control result coding table, so as to obtain a correlation between the target interface code, the target query result and the target risk control result (S102); and on the basis of the correlation, respectively configuring the target interface code, the target query result and the target risk control result to be on nodes of a preset decision tree, so as to generate a risk control model based on the decision tree (S103).

Description

Risk control model construction method, risk control review method, device and computer equipment

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on February 26, 2020, the application number is 202010120287.7, and the invention title is "risk control model construction method, risk control review method, device and computer equipment", all of which The content is incorporated in this application by reference.

Technical field

This application relates to the field of artificial intelligence technology, and in particular to a method for building a risk control model based on a decision tree, a method, device, device, computer equipment, and computer-readable storage medium for risk control review based on a decision tree.

Background technique

In the risk control business, there is a type of business that requires users to pass a risk control assessment before they can be processed. Generally, a risk control model is used for the user's risk control assessment, and different business needs to use different risk control models. For example, in the mobile phone installment business, customers can only transact business after they have passed the system risk control review. Since different operators and different cities have different business conditions, it is obviously not feasible to use the same set of risk control models for all. Yes, for example, there are many colleges and universities in City A, and many college students are blank credit accounts. It is necessary to establish a set of risk control models suitable for college students, which is different from ordinary salary groups. Under normal circumstances, different risk control models need to be developed accordingly. When there are multiple businesses, multiple risk control models need to be developed, but the development of multiple risk control models will cause time-consuming and labor-intensive problems.

In order to support business development, the system needs to support configurable risk control model functions. In traditional technology, there is a configurable risk control system, but it cannot meet various business scenarios. Many risk control models for specific business scenarios still need to be developed by code to achieve. It takes a lot of manpower to construct or adjust, which reduces the update efficiency of the risk control model.

Summary of the invention

This application provides a method for constructing a risk control model based on a decision tree, a method for reviewing a risk control based on a decision tree, a device, computer equipment, and a computer-readable storage medium, which can solve the update efficiency of the risk control model in the traditional technology, thereby improving the risk Control the accuracy of the audit.

In the first aspect, this application provides a method for constructing a risk control model based on a decision tree. The method includes: when a preset operation for constructing a risk control model is received, displaying a preset interface code table and query through a list page A result code table and a risk control result code table; wherein the interface code table includes multiple interface codes, the query result code table includes multiple query results, and the risk control result code table includes multiple risk control results ; Receiving through the list page a selection instruction to determine a target interface code from the interface code table, determine a target query result from the query result code table, and determine a target risk control result from the risk control result code table , Obtain the corresponding relationship between the target interface code, the target query result, and the target risk control result; based on the corresponding relationship, the target interface code, the target query result, and the target risk control The results are respectively configured at the nodes of the preset decision tree to generate a risk control model based on the decision tree.

In a second aspect, this application provides a risk control review method based on a decision tree, the method comprising: receiving user data of a user, the user data including preset information used to determine the user's business type; The preset information retrieves a risk control model based on a decision tree from a preset storage file; the risk control review is performed based on the user profile through the risk control model based on the decision tree.

In the third aspect, the present application also provides a decision tree-based risk control model construction device, including: a display unit for displaying the preset interface code through the list page when the preset operation of constructing the risk control model is received Table, query result code table, and risk control result code table; wherein, the interface code table includes multiple interface codes, the query result code table includes multiple query results, and the risk control result code table includes multiple A risk control result; a first receiving unit configured to receive, through the list page, the target interface code determined from the interface code table, the target query result determined from the query result code table, and the risk control result code The selection instruction for determining the target risk control result in the table is used to obtain the corresponding relationship between the target interface code, the target query result, and the target risk control result; The target interface code, the target query result, and the target risk control result are respectively configured at nodes of a preset decision tree to generate a risk control model based on the decision tree.

In a fourth aspect, this application also provides a risk control review device based on a decision tree, including: a second receiving unit for receiving user data of a user, and the user data includes information for determining the user's business type Preset information; a retrieval unit for retrieving a risk control model based on a decision tree from a preset storage file according to the preset information; an auditing unit for retrieving a risk control model based on a decision tree based on the decision tree The user information is reviewed for risk control.

In a fifth aspect, the present application also provides a computer device that includes a memory and a processor connected to the memory; the memory is used to store a computer program; the processor is used to run the storage in the memory The computer program to achieve the following steps: when receiving the preset operation of constructing the risk control model, display the preset interface code table, the query result code table and the risk control result code table through the list page; wherein, the interface code The table contains multiple interface codes, the query result code table contains multiple query results, and the risk control result code table includes multiple risk control results; the target is determined from the interface code table by receiving through the list page Interface code, determine the target query result from the query result code table, and determine the selection instruction of the target risk control result from the risk control result code table to obtain the target interface code, the target query result, and the The corresponding relationship between the target risk control results; based on the corresponding relationship, the target interface code, the target query result, and the target risk control result are respectively configured at the nodes of the preset decision tree to generate a decision tree-based Risk control model.

In a sixth aspect, the present application also provides a computer device, the computer device including a memory and a processor connected to the memory; the memory is used to store a computer program; the processor is used to run the memory stored in the memory The computer program to achieve the following steps: receiving user information of the user, the user information including preset information used to determine the type of business of the user; according to the preset information, retrieve from a preset storage file based on A risk control model of the decision tree; the risk control review is performed according to the user data through the risk control model based on the decision tree.

In a seventh aspect, the present application also provides a computer-readable storage medium that stores a computer program, and when the computer program is executed by a processor, the following steps are implemented: During the preset operation, the preset interface code table, query result code table, and risk control result code table are displayed on the list page; wherein, the interface code table includes multiple interface codes, and the query result code table includes multiple Query results, the risk control result code table includes a plurality of risk control results; receive through the list page the target interface code determined from the interface code table, and the target query result determined from the query result code table, and The selection instruction of the target risk control result is determined from the risk control result code table, and the corresponding relationship between the target interface code, the target query result, and the target risk control result is obtained; based on the corresponding relationship, The target interface code, the target query result, and the target risk control result are respectively configured at nodes of a preset decision tree to generate a risk control model based on the decision tree.

In an eighth aspect, the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented: receiving user information from a user, The user profile includes preset information used to determine the user's business type; according to the preset information, a risk control model based on a decision tree is retrieved from a preset storage file; through the risk control model based on a decision tree Conduct a risk control review based on the user profile.

Description of the drawings

In order to explain the technical solutions of the embodiments of the present application more clearly, the following will briefly introduce the drawings used in the description of the embodiments. Obviously, the drawings in the following description are some embodiments of the present application. Ordinary technicians can obtain other drawings based on these drawings without creative work.

FIG. 1 is a schematic flowchart of a method for constructing a risk control model based on a decision tree provided by an embodiment of the application;

2 is a schematic diagram of the relationship between three code tables involved in the risk control model based on a decision tree provided by an embodiment of the application;

FIG. 3 is a schematic diagram of a decision tree architecture provided by an embodiment of the application;

FIG. 4 is a schematic flowchart of a risk control review method based on a decision tree provided by an embodiment of the application;

FIG. 5 is a schematic diagram of another decision tree architecture provided by an embodiment of this application;

FIG. 6 is a schematic diagram of a sub-process of a risk control review method based on a decision tree provided by an embodiment of the application;

FIG. 7 is a schematic block diagram of a device for constructing a risk control model based on a decision tree provided by an embodiment of the application;

FIG. 8 is a schematic block diagram of a risk control review device based on a decision tree provided by an embodiment of the application; and

FIG. 9 is a schematic block diagram of a computer device provided by an embodiment of the application.

Detailed ways

The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all of them. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of this application.

Please refer to Figure 1 to Figure 3. FIG. 1 is a schematic flowchart of a method for constructing a risk control model based on a decision tree provided in an embodiment of the application, and FIG. 2 is a relationship between three code tables involved in a risk control model based on a decision tree provided in an embodiment of the application Figure 3 is a schematic diagram of a decision tree architecture provided by an embodiment of the application. As shown in Figure 1, the method includes the following steps S101-S103:

S101. When a preset operation for constructing a risk control model is received, display a preset interface code table, query result code table, and risk control result code table through a list page; wherein the interface code table includes multiple interface codes, The query result code table includes multiple query results, and the risk control result code table includes multiple risk control results.

Specifically, an interface code table is preset, the interface code table includes multiple interface codes, each of the interface codes corresponds to a query result code table, and each query result code table contains multiple query results, each The query result corresponds to a risk control result coding table, and each risk control result coding table includes multiple risk control results. When the risk control model needs to be constructed, the risk control model manager starts to build the risk control model, and the computer equipment receives the preset operation of the risk control model manager to construct the risk control model. When the preset operation to construct the risk control model is received, the list is passed The page displays the preset interface code table, query result code table, and risk control result code table. For example, the computer device receives the operation of constructing the risk control model, and according to the operation, displays the preset interface code table and the query result code corresponding to each interface code contained in the interface code table in a preset manner through the list page according to the operation. Table, and a risk control result code table corresponding to each query result contained in each query result code table. Further, the display modes of the three code tables of the interface code table, the query result code table, and the risk control result code table can include the following two ways: 1) The preset interface code table and each of the interface code tables The query result code table corresponding to each interface code, and the risk control result code table corresponding to each query result contained in the query result code table are displayed on the list page at the same time, so that the risk control model managers can learn from three Select one code table one by one. 2) First display the preset interface code table, according to the target interface code selected by the risk control model manager from the interface code table, and then display the query result code table corresponding to the target interface code, and manage according to the risk control model The personnel determines the target query result from the query result code table, and then displays the risk control result code table corresponding to the target query result, so that the risk control model manager can customize the target corresponding to the target query result according to business needs Risk control results.

It should be noted that the display mode of the above three code tables can be implemented in a variety of specific forms according to the page design of the list page. No matter how the display order and display mode of the three code tables change, they all reflect the display shown in Figure 2. The association relationship of the three code tables, that is, each interface code contained in the three code tables, the query result code table corresponding to each of the interface codes, and the risk control result corresponding to each query result The corresponding relationship.

S102. Receive, through the list page, information about determining a target interface code from the interface code table, determining a target query result from the target query result code table, and determining a target risk control result from the risk control result code table. Select the instruction to obtain the correspondence between the target interface code, the target query result, and the target risk control result.

Specifically, the risk control model manager selects the required interface code from the interface code table according to business needs to obtain the target interface code, and then selects the required target query result from the query result code table associated with each target interface code. Finally, the target risk control result corresponding to each target query result is selected from the risk control result code table corresponding to each target query result, and the computer equipment receives the risk control model management personnel from the list page in turn. The target interface code is determined from the interface code table, the target query result corresponding to the target interface code is determined from the query result code table, and the target query result corresponding to the target query result is determined from the risk control result code table The target risk control result selection instruction is based on the preset target interface code, the target query result and the association relationship between the target risk control result to obtain the target interface code, the target query result, and the target risk control result. The corresponding relationship between the target risk control results. Please continue to refer to Figure 2. The interface code 2 corresponding to node 2 of the decision tree shown in Figure 2 corresponds to query result 1 and query result 2 and their respective query results. Interface code 2 can also be customized as Corresponding to the query result 3 and its risk control result, a corresponding relationship between the target interface code, each target query result, and the target risk control result corresponding to each target query result is formed. Among them, the relationship between interface coding, query results, and risk control results can be implemented based on one-to-one, one-to-many, and many-to-one mapping relationships in the design of database tables. For example, in SQL databases, one For multiple mapping relationships, foreign keys are generally added to the table on the multiple side. The one-to-many mapping relationship in JAVA can be achieved by defining classes and arrays. In the embodiment of the present application, the association relationship between the interface code, the query result, and the risk control result may specifically include the following two ways:

1) The mapping relationship between the interface code and the query result adopts a one-to-many mapping relationship. 2) The mapping relationship between the query result and the risk control result can be a one-to-one mapping relationship, or a many-to-one mapping relationship. For example, if all query results correspond to the same risk control results, each query result corresponds to a one-to-one mapping relationship of a risk control result code table, which will increase the number of risk control result code tables, thereby increasing database resources For repeated use, the query result and the risk control result code table can adopt a many-to-one mapping relationship, that is, when each query result sets its own corresponding risk control result, the same risk control result code table is called to save the risk control result. The resource of the code table improves the efficiency of the use of the code table of risk control results.

S103. Based on the corresponding relationship, the target interface code, the target query result, and the target risk control result are respectively configured at nodes of a preset decision tree to generate a risk control model based on the decision tree.

Specifically, each node of the decision tree corresponds to an interface code, each interface code is associated with one or more query results, each query result corresponds to a risk control result, and the corresponding relationship corresponding to each interface code is A node corresponding to the decision tree is set in advance to generate a risk control model based on the decision tree, that is, the corresponding relationship between the target interface code, the target query result, and the target risk control result is coded through the interface Correspondingly set with the nodes of the decision tree to generate a risk control model based on the decision tree.

The embodiment of this application is based on the design of three code tables of interface coding table, query result coding table and risk control result coding table. Three code tables of interface coding table, query result coding table and risk control result coding table are established in advance to realize decision-based When the tree's risk control model is constructed, it receives the preset operation of constructing the risk control model. When the preset operation of constructing the risk control model is received, the preset interface code table, query result code table and risk control result are displayed on the list page. A code table is received through the list page to determine a target interface code from the interface code table, determine a target query result from the target query result code table, and determine a target risk control result from the risk control result code table The corresponding relationship between the target interface code, the target query result, and the target risk control result is obtained, and the corresponding relationship corresponding to the interface code is preset to correspond to the node of the decision tree In order to generate a risk control model based on the decision tree, the risk control model manager can customize the corresponding to each node contained in the risk control model based on the decision tree from the three code tables according to the business needs corresponding to the business type. The target interface code of the target interface code, the target query results corresponding to each target interface code, and the target risk control result corresponding to each target query result, to generate a risk control model based on a decision tree to realize custom decision making Tree, a decision tree corresponds to a final risk control model, so as to meet the flexible configuration requirements of the risk control business for the risk control model, and can realize the personalized configuration of the risk control model in different scenarios according to different business needs. Since neither the configuration of the risk control model nor the modification of the risk control model need to develop code, and will take effect in real time after the risk control mode is configured, compared with the configurable risk control system in the traditional technology, due to the configurable risk control in the traditional technology The system defines a batch of fixed risk control conditions in advance, and then makes some simple choices and combinations. Therefore, it is unable to achieve very detailed configuration, resulting in the risk control model configuration is not flexible enough to meet various business scenarios The risk control model implements the refined configuration of the interface level. Therefore, the risk control model of many specific business scenarios in the traditional technology still needs to be developed to realize the risk control model. Each adjustment requires a lot of manpower. The embodiment of this application is based on decision-making The construction and update of the tree's risk control model are simple and efficient. Therefore, the efficiency of the construction and update of the risk control model is improved, and the timeliness of the construction and update of the risk control model is improved.

In an embodiment, the receiving through the list page determines the target interface code from the interface code table, determines the target query result from the target query result code table, and receives from the risk control result code table The step of determining the selection instruction of the target risk control result, and obtaining the corresponding relationship between the target interface code, the target query result, and the target risk control result includes: receiving the target interface code determined from the interface code table Select an instruction to obtain a target interface code; display the query result code table associated with the target interface code through the list page; receive a selection instruction for determining a target query result from the query result code table to obtain the target interface The target query result corresponding to the code; display the risk control result code table associated with the target query result through the list page; receive a selection instruction to determine the target risk control result from the risk control result code table to obtain the result The target risk control result corresponding to the target query result; the target interface code, the target query result, and the target risk control result are associated through a preset association method to obtain the target interface code, the target Correspondence between the query result and the target risk control result.

Specifically, the risk control model manager starts the configuration of the risk control model, and the computer device receives the operation to start the configuration of the risk control model, and the preset interface code table is displayed through the list page, and the risk control model manager selects from the interface code table Target interface code, the computer device receives the selection of the interface code contained in the interface code table to obtain the target interface code, and after receiving the target interface code selected by the risk control model manager, it displays each interface code through the list page. The query result code table associated with the target interface code itself, and the computer equipment then receives the risk control model manager’s selection of the query results contained in the query result code table to obtain the corresponding code of the target interface code. The target query result, the query result is used to describe the preset classification corresponding to the user profile, for example, the age interface code corresponds to the division of age groups, and the computer device further displays the target query result through the list page according to the target query result. The risk control result code table associated with the target query result itself, and the computer device then receives the selection of the risk control result contained in the risk control result code table to obtain the target risk control result corresponding to the target query result, thereby Obtain the corresponding relationship between the target interface code selected by the risk control model manager, the target query result corresponding to the target interface code, and the risk control result corresponding to the target query result, and the target interface code, the target query result The target query result and the target risk control result are related by a preset association method, for example, the interface code table and the query result are through a one-to-many relationship, and the query result and the risk control result adopt a one-to-one association Relationship to obtain the corresponding relationship between the target interface code, the target query result, and the target risk control result. In the embodiment of the application, the interface code table, the query result code table corresponding to each target interface code, and the risk control result code table corresponding to each target query result are displayed in sequence according to the selection of the model manager , The risk control model manager selects the target interface code from the three code tables, several target query results corresponding to each target interface code, and the target risk control result corresponding to each target query result, which can make a visualized page The design is hierarchical and a rational page layout is constructed to save page resources and cache resources of computer equipment, and improve the display efficiency of computer equipment, especially when the computer equipment is multi-tasking, which can improve the resource utilization of computer equipment.

In one embodiment, each query result contained in the query result code table is correspondingly set with a query result code; said receiving a selection instruction for determining a target risk control result from the risk control result code table, to The step of obtaining the target risk control result corresponding to the target query result includes: obtaining the target query result code corresponding to the target query result; displaying the risk control result code associated with the target query result code through the list page Table; receiving a selection instruction for determining a target risk control result from the risk control result code table to obtain the target risk control result corresponding to the target query result code, thereby obtaining the target risk control corresponding to the target query result result.

Specifically, by setting a corresponding query result code in the query result coding table for each query result, the established query result code table includes the query result, the query result code corresponding to the query result, and the query result judgment condition. Wherein, the query result is used to describe the corresponding classification category of the data specified in advance by the user according to the preset classification condition satisfied by the data. Since the query result, the query result code and the query result judgment condition can be set separately, the query result, the query result code and the query result judgment condition have the flexibility of their respective settings. For example, please refer to Table 1. Table 1 is an example of the query result code table of the age control interface. The interface code return result of the age control interface is age, which can be configured as the following query results shown in Table 1. When receiving When it comes to the user's age information contained in the user profile in a risk control review, the age of the user is combined with the age judgment condition to obtain the user's age query result and the query result code corresponding to the age query result. Further, the query results in Table 1 can be further subdivided according to business needs. For example, the query result judgment conditions can be divided into 0-17 years old, 18-30 years old, 31-50 years old, 50-65 years old, and The corresponding query results and query result codes are also adjusted to five types for the classification of five conditions, including those over 65 years old.

Table 1

Furthermore, it can be seen from Table 1 that in this embodiment of the application, since only different query result judgment conditions need to be customized, the user condition control interface corresponding to an interface code can return various query results according to the customized query result judgment conditions. Different corresponding query results, so as to achieve the same user condition control interface to realize the customization of different conditions. The more detailed the definition of the judgment condition of the customized query result, the more detailed the query result returned by the interface code, so that the wind can be realized. The refined configuration of the control model. For example, since the age control interface can return the 4 kinds of query results shown in Table 1, it can also return the 5 kinds of query results mentioned above according to business needs, or return the control age control interface to other settings according to business needs. 3 kinds of query results, so as to realize the risk control model manager can realize the fine configuration of the risk control model through the visual risk control model configuration page configuration in the background system, thereby improving the flexibility and accuracy of the query results returned by the interface code to achieve Accuracy, flexibility and refined configuration of wind control model settings.

Further, please refer to Table 2. Table 2 is an example of a risk control result coding table. In Table 2, according to the risk control requirements of different business needs, the risk control results of "pass" and "reject" are established respectively, and the results of each category can support the configuration of multiple specific sub-categories. . Among them, the risk control results can be set according to business needs. For example, a risk control result can be uniquely determined by combining a custom credit line with a query result code according to business needs. Among them, a custom credit line refers to a credit line that meets preset conditions. The requirements corresponding to the preset conditions, for example, for age conditions, the risk control result corresponding to the age under 18 is rejected, or it can be set to the risk control result corresponding to the age under 22 as reject, where the age is over 18 Or 22 years of age or older is a custom credit line, which is self-defined according to business needs.

Form 2

Please continue to refer to Form 2. If Form 2 is for the telecommunications (campus) card business of the organization, its business processing requirements allow customers aged 18 to 70 years old to apply. The risk control results can be selected: 1) The customer type is "Age" 18 years old (customer_class002)", set the risk control result to "pass". 2) The customer type is "Over 70 years old (customer_class004)", and the risk control result is "Rejected".

In this embodiment of the application, a corresponding query result code is set in the query result coding table through each query result, and the risk control result corresponding to the query result is coded by the query result corresponding to the query result For association, after obtaining the target query result selected by the risk control model manager, the computer device obtains the target query result code corresponding to the target query result according to the target query result, and then displays the target query result code through the list page The risk control result code table associated with the target query result code, the computer equipment then receives the risk control model manager’s selection of the risk control results contained in the risk control result code table to obtain the corresponding target query result code The target risk control result, so as to obtain the target risk control result corresponding to the target query result, for multiple nodes of the decision tree, repeat the generation process of the corresponding relationship corresponding to one of the nodes until the model manager according to the business requirements The setting of the risk control review model is completed to generate the risk control review model based on the decision tree. The query result is associated with the risk control result by setting a query result code between the query result and the risk control result. Since there is no direct correlation between the query result and the risk control result, there is no need to directly modify the query result or the risk control result. Modifying the code of the risk control results and query results reduces the workload of the development model. The configuration of the risk control model is simple and clear, which reduces the probability of errors in the configuration of complex risk control models, and reduces the use of risk control model managers. Threshold makes the configuration and operation of the risk control model as simple and efficient as possible, improves the generation efficiency of the risk control model, and improves the timeliness of the update of the risk control model.

Please refer to FIG. 4, which is a schematic flowchart of a risk control review method based on a decision tree provided by an embodiment of the application. As shown in Figure 4, the method includes the following steps S401-S403:

S401. Receive user data of a user, where the user data includes preset information used to determine a service type of the user.

Among them, the business type refers to the type of business with different target groups as the target objects obtained by classifying the business according to the business goal. For example, in the risk control review, there is a type of risk control review business for college students. Since the conditions of the college student group and the salary group are different, the respective risk control reviews for the college student group and the salary group are carried out through different conditions. , Will correspond to the type of risk control review business for college students and the type of risk control review business for salary groups. Among them, the preset information includes the user's address, the user's age, the business name, and the ID number. The preset information is mainly used to judge and identify the type of business handled by the user.

Specifically, when a user handles a business that requires a risk control review, the user will be required to provide user information through a computer device, and the computer device will receive the user information of the user. The user information includes a preset for determining the user’s business type Information, through the preset information to determine the type of business the user handles. For example, when a user handles a consumer installment service, the user information that needs to be provided includes personal information and the type of package processed, and the user's business type is determined by the information such as the personal information and the type of package.

S402: According to the preset information, retrieve the risk control model based on the decision tree from the preset storage file.

Among them, the decision tree is Decision Tree in English. The decision tree is a tree structure in which each internal node represents an attribute, each node represents a branch, and each branch represents an attribute output, and each node includes at least one leaf Nodes, each leaf node represents a risk control result, and the nodes of the decision tree support serial and parallel relationships, AND/or relationships, and multi-level structures.

Specifically, in order to achieve different business scenarios, for example, because different regions and cities, different packages and other business types need to set up different risk control models, the risk control model in this embodiment of the application is based on a decision tree description. The risk control model It is constructed based on the decision tree-based risk control model construction method of each of the foregoing embodiments. When conducting a risk control review on a user, according to the preset information, a pre-built risk control model based on the decision tree of the business type corresponding to the preset information is retrieved from a preset storage file.

Further, the preset information includes the user address and the target business name. According to the preset information, the step of retrieving a risk control model based on a decision tree from a preset storage file includes: according to the user address and the target business name, retrieving the pre- and The risk control model based on the decision tree of the business type corresponding to the user address and the target business name.

Specifically, since in the risk control business, corresponding risk control models will be set up according to different groups in different regions, during the risk control review, the risk control review will be conducted according to the user's address and the user's target business name. If the preset information includes the user address and the target business name, it will retrieve the pre-built relationship with the user address and the target business name from the preset storage file according to the user address and the target business name. A risk control model based on a decision tree corresponding to the business type. For example, in the risk control review business, because different business operators have different business conditions in different regions and cities, the corresponding business types are generally formulated according to the regions, cities and target groups. Therefore, the user address and target business name can be pre-defined. It is assumed that the information determines the risk control model corresponding to the user's risk control review, and the risk control model is only carried out by the risk control model manager in the background of the system through the visual risk control model configuration page according to the actual business needs of different cities The configuration can be obtained without the need to develop the code of the risk control model, so as to realize a simple, flexible and refined configuration of the risk control model to improve the accuracy and efficiency of the risk control review process.

S403: Perform a risk control review based on the user profile through the risk control model based on the decision tree.

Specifically, since the target classification can be achieved through a decision tree (also called a classification tree), in the embodiment of the present application, the risk control review target can be audited through the decision tree, so that the risk control review target can be reviewed according to the risk control model. Obtain the results of the risk control audit, for example, get the audit results of two categories: "approved" and "failed". Further, the risk control review based on the user data through the risk control model based on the decision tree is based on the structure of the preset decision tree and the risk control results of each node in the decision tree, and the user's risk control can be obtained. Control audit results. For example, according to the user profile, the target query result corresponding to each interface code in the risk control model is obtained through each of the interface codes, and the target query result corresponding to the target query result is obtained according to the target query result. The target risk control result to obtain the target risk control result corresponding to the interface code, so as to obtain the target risk control result corresponding to each node included in the risk control model, so as to obtain a risk control review according to the risk control model the result of.

Furthermore, the structure of the decision tree and the relationship between the nodes can be set according to business needs. For example, the nodes of the decision tree can include an "and" relationship and an "or" relationship, and the "and" relationship is used to describe the The risk control model needs to satisfy that the risk control results of two nodes are both "passed", and the "or" relationship is used to describe that the risk control model needs to satisfy that at least one of the risk control results of the two nodes is "passed". Taking the decision tree-based risk control model including the "and" relationship and the "or" relationship between the nodes as an example, the step of obtaining the result of the risk control review according to the risk control model includes: judging the risk control Whether the risk control results corresponding to the interface codes corresponding to the nodes in the "and" relationship in the model are all "passed"; if the risk control model corresponds to the nodes in the "and" relationship The risk control results corresponding to the interface code are all "passed", and it is determined that the user's risk control review is passed; if the risk control model in the risk control model corresponds to the interface code corresponding to the node in the "and" relationship If the risk control results are not all "passed", it is determined that the user's risk control review has not passed. For example, in a risk control model based on a decision tree, the decision tree includes multiple nodes, and the multiple nodes may have an "and" relationship or an "or" relationship, etc., please continue to refer to Figure 3, as shown in Figure 3 , The decision tree corresponding to the risk control model shown in Figure 3 contains nodes: A, A1, A2, A3, A11, A12, A31, and A32. Each node represents an interface code. These nodes can be "and "Or" relationship, the relationship between nodes is "and", which means that the conditions of two nodes need to be met at the same time, and the relationship between nodes is "or", which means that the condition of one of the two nodes is satisfied. Through the "and" relationship or "or" relationship between nodes, and the risk control result corresponding to each node, the result of the decision tree is obtained, and the result of the risk control review is thus obtained.

When the embodiment of this application implements decision tree-based risk control review, the risk control model constructed by the decision tree-based risk control model construction method in each of the above embodiments, because the risk control model can be refined at the interface level Customize the configuration, so that every construction and update of the risk control model can be tailored to the actual business needs, and the risk control model will take effect immediately after configuration, enabling the risk control model to match the risk control review business in a refined and efficient manner, and improve The efficiency and accuracy of the risk control review process.

In one embodiment, each query result contained in the query result coding table is correspondingly set with a query result code; the step of performing risk control review according to the user data through the decision tree-based risk control model includes : According to the user profile, the target query result corresponding to each interface code is obtained through the interface code corresponding to each node included in the risk control model; each target query result is obtained according to each target query result. The target query result code corresponding to each of the target query results; according to each target query result code, the target risk control result corresponding to each target query result code is obtained to obtain each node Each corresponding risk control result; according to the risk control result corresponding to each node, the risk control audit of the user is performed through the risk control model based on the decision tree.

Specifically, the computer device retrieves the risk control model of the business type corresponding to the preset information according to the preset information in the user profile, and compares the received user profile with each of the risk control model based on the decision tree. Nodes are matched. For example, the user’s age information is matched with the interface code through the node corresponding to the age interface code in the risk control model, and the user’s income information is matched through the node and interface code corresponding to the income interface code in the risk control model. Matching, etc., obtain the target query result corresponding to each of the interface codes in the risk control model through each of the interface codes, and obtain the corresponding corresponding to each target query result according to each of the target query results According to the target query result code of each target query result code, the target risk control result corresponding to each target query result code is obtained to obtain the risk control result corresponding to each node, according to The risk control result corresponding to each node is combined with the tree structure of the decision tree to conduct a risk control review on the user.

In one embodiment, before the step of obtaining the target risk control result corresponding to each of the target query result codes, the method further includes: pre-querying the node information contained in the risk control model. Pre-query the corresponding risk control results.

Among them, pre-query refers to pre-query the risk control results of the credit reporting and risk control interface. The pre-query method includes pre-query according to the interface type of the credit reporting and risk control interface, and time-consuming pre-query according to the interface query of the credit reporting and risk control interface. Query and combine the above two methods to pre-query the credit information and risk control interface involved in the risk control model.

Specifically, by pre-querying the risk control results of the interface codes corresponding to the nodes included in the risk control model, the response time of the risk control review system can be shortened, and the efficiency of the computer equipment in processing the risk control review can be improved.

Further, taking the pre-query according to the interface type of the credit information and risk control interface as an example, in one embodiment, the interface code includes multiple, and each of the interface codes corresponds to a credit information and risk control query interface. The decision tree includes multiple levels, and the step of pre-querying the risk control results corresponding to the nodes included in the risk control model by pre-querying The credit information and risk control query interface performs concurrent queries, and each query result corresponding to the first type of credit information and risk control query interface is stored in the cache. The first type of credit and risk control query interface includes free credit information. According to the current level to which the node of the decision tree belongs, according to the current level to which the node of the executed decision tree belongs, request in advance for the child nodes included in the first preset level below the current level The respective query results of the second type of credit information and risk control query interface are stored in the cache for subsequent use by the decision tree, and the second type of credit and risk control query interface The credit report and risk control query interface includes a query interface for charging for each query.

Specifically, in the risk control review, in order to more accurately identify the risks of customers, it is generally necessary to obtain user data from multiple credit data sources to establish a risk control model, such as a third-party identity verification interface, a blacklist interface, Interfaces such as the multi-head lending interface and the credit score query interface involve obtaining verification results from multiple third-party interfaces, leading to more and more time-consuming risk control reviews. In order to ensure the efficiency of risk control review in the case of multiple data sources, the risk control review data of multiple data sources can be processed separately through the following steps:

1) When the risk control review is initialized, perform concurrent queries on the first type of credit information and risk control query interface, and store each query result corresponding to the first type of credit information and risk control query interface in the cache. The first type of credit information and risk control query interface includes the free credit information and risk control query interface and the credit information and risk control query interface for fee packages. 2) According to the current level to which the node of the executed decision tree belongs, request in advance the respective query results of the second type of credit information and risk control query interface corresponding to the first preset number of sub-nodes below the current level, and The respective query results of the second type of credit information and risk control query interface are stored in the cache for subsequent call by the decision tree, and the second type of credit information and risk control query interface includes billing for each query Query interface.

Since the most time-consuming operation of the risk control model is the waiting time for external data interface query, and external data query is charged, it is not possible to query all interfaces at the beginning, otherwise there will be users who fail to pass the risk control review Query waste. So you can take the following two steps to query separately. For example, 1) When the user initializes the letter review, he will perform concurrent queries on all the free interfaces and the interface of the fee package, such as the interface of the fee package, and store the Redis cache and other caches. In, for subsequent decision tree calls. 2) For charging interfaces, according to the current decision tree execution level, request in advance the first preset number of layers below the current level, for example, the interface data of the two sub-nodes below the current level, and store it in the Redis cache and other caches , For the subsequent decision tree to call. Please refer to FIG. 5. FIG. 5 is a schematic diagram of another decision tree architecture in the decision tree-based risk control review method provided by an embodiment of the application. As shown in Figure 5, if judgment A is currently being executed, the data of six interfaces B1, B2, C1, C2, C3, and C4 can be queried in advance. If the risk control result can be obtained by judgment A, there is no need to query D1, D2 The data of eight interfaces such as D3, D4, D5, D6, D7 and D8, especially when D1-D8 interfaces include interfaces that need to be charged, it can avoid query waste and shorten the response time of the credit review system; if B2 is currently implemented To judge, it is necessary to query the data of six interfaces such as C3, C4, D5, D6, D7 and D8 in advance. If the risk control result can be obtained through the B2 judgment, there is no need to query the data of the interface corresponding to the node after the D5 to D8 level. Especially when the subsequent interfaces of D5-D8 include interfaces that need to be charged, query waste can be avoided, and the response time of the risk control audit system can also be shortened. Among them, the three logics between the three-level nodes contained in each of Figure 4 are "or" relationships.

The embodiment of this application uses the pre-query mechanism of the credit investigation and risk control interface in the above-mentioned risk control review, within the controllable range of the credit inquiry fee, by shortening the response time of the credit review system, and improving the performance of the computer equipment in processing the credit review. Improve the efficiency of credit review.

Further, taking the time-consuming pre-query according to the interface query of the credit information risk control interface as an example, please refer to FIG. 6. FIG. 6 is a schematic diagram of a sub-process of the risk control review method based on a decision tree provided by an embodiment of the application. As shown in FIG. 6, in the embodiment corresponding to the sub-process shown in FIG. 6, the decision tree includes multiple levels, and the nodes included in the risk control model are corresponding to the nodes included in the risk control model through a pre-query. The steps of pre-querying the results of risk control include: S601. The query time-consuming historical data for querying according to each credit information and risk control query interface is calculated, and the query consumption of each of the credit information and risk control query interfaces in the preset time period is counted. Time data; S602. Sort the query time-consuming data in a preset order to obtain a preset number of the credit information and risk control query interfaces obtained according to the query time-consuming order in descending order, to form a query interface Information and risk control query interface set; S603. Determine whether the query interface corresponding to the child node contained in the second preset number below the current level to which the decision tree execution node belongs includes the information and risk control query interface set Credit information and risk control query interface; S604. If the query interface corresponding to the child nodes included in the second preset number below the current level to which the decision tree execution node belongs includes the credit information and risk control query The credit information and risk control query interface in the centralized interface calls the credit information and risk control query interface in the credit information and risk control query interface set, and stores the query results corresponding to the query interface in the credit and risk control query interface set in the cache S605, if the current level to which the decision tree execution node belongs, the query interface corresponding to the child node included in the second preset number of levels below does not include the The credit information and risk control query interface in the centralized credit information and risk control query interface does not call the credit information and risk control query interface in the centralized credit and risk control query interface.

Specifically, in a risk control review, user data needs to be obtained from multiple credit reporting data sources to build a risk control model. The query time-consuming historical data can be queried according to the various credit reporting and risk control query interfaces, before statistics The query time-consuming data of each of the credit investigation risk control query interfaces within a preset time period is sorted in a preset order to obtain the query time-consuming data in descending order according to the query time A preset number of the credit information and risk control query interfaces are formed to form a credit information and risk control query interface set, and the query interfaces corresponding to the child nodes included in the second preset number below the current level to which the decision tree execution node belongs are determined Whether it contains the credit information and risk control query interface in the credit information and risk control query interface set, if the current level to which the decision tree execution node belongs, the sub-nodes included in the second preset number of levels below correspond to the The query interface includes the credit report and risk control query interface in the credit report and risk control query interface collection, calls the credit report and risk control query interface in the credit report and risk control query interface collection, and centralizes the credit report and risk control query interface The query result corresponding to the query interface of the query interface is stored in the cache for the subsequent call of the decision tree. If the current level to which the decision tree execution node belongs corresponds to the child node contained in the second preset level The query interface does not include the credit information and risk control query interface in the credit information and risk control query interface set, and does not call the credit information and risk control query interface in the credit information and risk control query interface set. For example, according to the time-consuming monitoring of each interface in the previous hour in the history, the most time-consuming interface is counted. If this interface appears in the lower five sub-nodes, it will be called in advance and stored in the Redis cache and other caches for subsequent decision tree calls. , So as to improve query efficiency, can shorten the response time of the credit review system, improve the performance of computer equipment to process credit review, so as to improve the efficiency of credit review.

Furthermore, in one embodiment, the pre-query according to the interface type of the credit reporting risk control interface and the time-consuming pre-query according to the interface query of the credit reporting risk control interface are combined to compare the risks involved in the risk control model. Pre-query through the credit report and risk control interface.

Specifically, the pre-query can be performed according to the interface type of the credit reporting and risk control interface, and then the time-consuming pre-query can be performed according to the interface query of the credit reporting and risk control interface, or the pre-query can be performed first according to the interface query of the credit reporting and risk control interface. Query, and then pre-query according to the interface type of the credit information risk control interface. Please continue to refer to Figure 5. You can pre-query according to the interface type of the credit information risk control interface, that is, perform A judgment at the current time, and then query the data of the six interfaces B1, B2, C1, C2, C3 and C4 in advance, and then When querying the data of the above six interfaces, if the C2 query takes the longest time according to the statistics, then perform the pre-query according to the interface query of the credit risk control interface. First query C2, and then query B1, B2, C1, C3 And C4 five interface data, in order to realize the combination of the above two methods. Alternatively, if the C2 query takes the longest time according to the statistics, the pre-query can be performed according to the interface query time-consuming of the credit information risk control interface. In the current execution of A judgment, the data of the five interfaces B1, B2, C1, C3 and C4 are then queried in advance to realize the combination of the above two methods.

It should be noted that in the decision tree-based risk control model construction method and the decision tree-based risk control review method described in the above embodiments, the technical features included in the different embodiments can be recombined as needed to obtain The combined implementation schemes are all within the scope of protection required by this application.

Please refer to FIG. 7, which is a schematic block diagram of a device for constructing a risk control model based on a decision tree provided by an embodiment of the application. Corresponding to the above-mentioned method for building a risk control model based on a decision tree, an embodiment of the present application also provides a device for building a risk control model based on a decision tree. As shown in FIG. 7, the decision tree-based risk control model construction device includes a unit for executing the above-mentioned decision tree-based risk control model construction method, and the device can be configured in a computer device such as a server. Specifically, referring to FIG. 7, the device 700 for constructing a risk control model based on a decision tree includes a display unit 701, a first receiving unit 702 and a generating unit 703.

Wherein, the display unit 701 is configured to display the preset interface code table, query result code table, and risk control result code table through the list page when the preset operation of constructing the risk control model is received; wherein, the interface code table Contains multiple interface codes, the query result code table contains multiple query results, and the risk control result code table includes multiple risk control results; the first receiving unit 702 is configured to receive data from all sources through the list page. The target interface code is determined in the interface code table, the target query result is determined from the query result code table, and the selection instruction of the target risk control result is determined from the risk control result code table to obtain the target interface code, The corresponding relationship between the target query result and the target risk control result; a generating unit 703, configured to configure the target interface code, the target query result, and the target risk control result separately based on the corresponding relationship At the nodes of the preset decision tree, a risk control model based on the decision tree is generated.

In one embodiment, the first receiving unit 702 includes: a first receiving subunit, configured to receive a selection instruction for determining a target interface code from the interface code table to obtain the target interface code; a first display subunit, It is used to display the query result code table associated with the target interface code through the list page; the second receiving subunit is used to receive a selection instruction for determining the target query result from the query result code table to obtain the The target query result corresponding to the target interface code; the second display sub-unit is used to display the risk control result code table associated with the target query result through the list page; the third receiving sub-unit is used to receive from the The selection instruction for determining the target risk control result in the risk control result coding table to obtain the target risk control result corresponding to the target query result; the correlation subunit is used to encode the target interface code, the target query result and all The target risk control results are associated through a preset association method to obtain the corresponding relationship between the target interface code, the target query result, and the target risk control result.

In one embodiment, each query result contained in the query result code table corresponds to a query result code; the third receiving subunit includes: an obtaining subunit for obtaining the target query result The corresponding target query result code; the third display subunit is used to display the risk control result code table associated with the target query result code through the list page; the fourth receiving subunit is used to receive data from the risk control The selection instruction of the target risk control result is determined in the result code table to obtain the target risk control result corresponding to the target query result code, so as to obtain the target risk control result corresponding to the target query result.

Please refer to FIG. 8, which is a schematic block diagram of a risk control review device based on a decision tree provided by an embodiment of the application. Corresponding to the above-mentioned risk control review method based on the decision tree, an embodiment of the present application also provides a risk control review device based on the decision tree. As shown in FIG. 8, the decision tree-based risk control review device includes a unit for executing the above-mentioned decision tree-based risk control review method, and the device can be configured in a computer device such as a terminal. Specifically, referring to FIG. 8, the risk control review device 800 based on a decision tree includes a second receiving unit 801, a retrieval unit 802 and a review unit 803.

Wherein, the second receiving unit 801 is configured to receive user data of the user, and the user data includes preset information used to determine the service type of the user; the calling unit 802 is configured to obtain information from A risk control model based on a decision tree is retrieved from a preset storage file; the review unit 803 is configured to conduct a risk control review based on the user profile through the risk control model based on the decision tree.

In one embodiment, each query result contained in the query result code table is correspondingly set with a query result code; the review unit 803 includes: a first obtaining subunit, which is configured to pass the The interface code corresponding to each node included in the risk control model obtains the target query result corresponding to each of the interface codes; the second obtaining subunit is used to obtain each target query result according to each target query result. The target query result code corresponding to each of the target query results; the audit subunit is used to perform risk control on the user through the risk control model based on the decision tree according to the risk control result corresponding to each of the nodes Audit.

In one embodiment, the decision tree-based risk control review device 800 further includes: a pre-query unit for performing a pre-query method on the risk control results corresponding to the nodes included in the risk control model. Pre-query.

It should be noted that those skilled in the art can clearly understand that the specific implementation process of the above-mentioned decision tree-based risk control model construction device and each unit is different from the above-mentioned decision tree-based risk control review device and the specific implementation process of each unit. For the implementation process, please refer to the corresponding descriptions in the foregoing method embodiments respectively. For the convenience and brevity of the description, details are not repeated here.

At the same time, the above-mentioned decision tree-based risk control model construction device and the decision tree-based risk control review device are only used as examples for the division and connection of each unit. In other embodiments, the decision tree-based risk control model The construction device and the decision tree-based risk control review device are divided into different units as needed. The units in the decision tree-based risk control model construction device and the decision tree-based risk control review device can also be connected in different connection sequences. And methods to respectively complete all or part of the functions of the above-mentioned decision tree-based risk control model construction device and the decision tree-based risk control review device.

The aforementioned decision tree-based risk control model construction device and the decision tree-based risk control review device can be implemented in the form of a computer program, and the computer program can run on the computer device as shown in FIG. 9 respectively.

Please refer to FIG. 9, which is a schematic block diagram of a computer device according to an embodiment of the present application. The computer device 900 may be a computer device such as a desktop computer or a server, and may also be a component or component in other devices.

Referring to FIG. 9, the computer device 900 includes a processor 902, a memory, and a network interface 905 connected through a system bus 901. The memory may include a non-volatile storage medium 903 and an internal memory 904, and the memory may also be volatile storage. medium.

The non-volatile storage medium 903 can store an operating system 9031 and a computer program 9032. When the computer program 9032 is executed, it can make the processor 902 execute the aforementioned decision tree-based risk control model construction method, or can make the processor 902 execute the aforementioned decision tree-based risk control review method.

The processor 902 is used to provide calculation and control capabilities to support the operation of the entire computer device 900.

The internal memory 904 provides an environment for the operation of the computer program 9032 in the non-volatile storage medium 903. When the computer program 9032 is executed by the processor 902, the processor 902 can execute a risk control model construction based on the decision tree described above. Alternatively, the processor 902 may be caused to execute the above-mentioned risk control review method based on the decision tree.

The network interface 905 is used for network communication with other devices. Those skilled in the art can understand that the structure shown in FIG. 9 is only a block diagram of part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device 900 to which the solution of the present application is applied. The specific computer device 900 may include more or fewer parts than shown in the figure, or combine certain parts, or have a different arrangement of parts. For example, in some embodiments, the computer device may only include a memory and a processor. In such embodiments, the structures and functions of the memory and the processor are consistent with the embodiment shown in FIG. 9 and will not be repeated here.

Taking the foregoing processor 902 executing the foregoing decision tree-based risk control model construction method as an example, the processor 902 is configured to run a computer program 9032 stored in the memory to implement the risk control model construction method of the embodiment of the present application. Or risk control review method.

It should be understood that, in the embodiment of the present application, the processor 902 may be a central processing unit (Central Processing Unit, CPU), and the processor 902 may also be other general-purpose processors, digital signal processors (Digital Signal Processors, DSPs), and special purpose processors. Integrated circuit (Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor.

Those of ordinary skill in the art can understand that all or part of the processes in the methods of the foregoing embodiments can be implemented by a computer program, and the computer program can be stored in a computer-readable storage medium. The computer program is executed by at least one processor in the computer system to implement the steps of the embodiment of the above-mentioned risk control model construction method or risk control review method.

Therefore, the embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium may be a non-volatile computer-readable storage medium, or may be a volatile storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the processor Perform the steps of the decision tree-based risk control model construction method described in the above embodiments, or make the computer execute the steps of the decision tree-based risk control review method described in the above embodiments.

The storage medium is a physical, non-transitory storage medium, such as a U disk, a mobile hard disk, a read-only memory (Read-Only Memory, ROM), a magnetic disk or an optical disk, etc., which can store computer programs. medium.

A person of ordinary skill in the art may be aware that the units and algorithm steps of the examples described in the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of both, in order to clearly illustrate the hardware and software Interchangeability, in the above description, the composition and steps of each example have been generally described in accordance with the function. Whether these functions are executed by hardware or software depends on the specific application and design constraint conditions of the technical solution. Professionals and technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered beyond the scope of this application.

The above are only specific implementations of this application, but the scope of protection stated in this application is not limited to this. Any person skilled in the art can easily think of various equivalents within the technical scope disclosed in this application. Modifications or replacements, these modifications or replacements should be covered within the scope of protection of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims.

Claims

A method for building a risk control model based on a decision tree, including:

When the preset operation of constructing the risk control model is received, the preset interface code table, query result code table, and risk control result code table are displayed through the list page; wherein, the interface code table includes multiple interface codes, and The query result code table includes multiple query results, and the risk control result code table includes multiple risk control results;

Receiving, through the list page, a selection instruction for determining a target interface code from the interface code table, determining a target query result from the query result code table, and determining a target risk control result from the risk control result code table, Obtain the correspondence between the target interface code, the target query result, and the target risk control result;

Based on the corresponding relationship, the target interface code, the target query result, and the target risk control result are respectively configured at nodes of a preset decision tree to generate a risk control model based on the decision tree.
The method for constructing a risk control model based on a decision tree according to claim 1, wherein said receiving through the list page determines a target interface code from the interface code table, and determines a target query result from the query result code table , And determining a selection instruction for a target risk control result from the risk control result code table, and obtaining the corresponding relationship between the target interface code, the target query result, and the target risk control result includes:

Receiving a selection instruction for determining a target interface code from the interface code table to obtain the target interface code;

Displaying the query result code table associated with the target interface code through the list page;

Receiving a selection instruction for determining a target query result from the query result code table to obtain a target query result corresponding to the target interface code;

Displaying the risk control result code table associated with the target query result through the list page;

Receiving a selection instruction for determining a target risk control result from the risk control result coding table to obtain the target risk control result corresponding to the target query result;

The target interface code, the target query result, and the target risk control result are associated through a preset association method to obtain the relationship between the target interface code, the target query result, and the target risk control result Correspondence.
The method for constructing a risk control model based on a decision tree according to claim 2, wherein each query result contained in the query result code table is correspondingly set with a query result code;

The step of receiving a selection instruction for determining a target risk control result from the risk control result coding table to obtain the target risk control result corresponding to the target query result includes:

Acquiring the target query result code corresponding to the target query result;

Displaying the risk control result code table associated with the target query result code through the list page;

A selection instruction for determining a target risk control result from the risk control result code table is received to obtain the target risk control result corresponding to the target query result code, so as to obtain the target risk control result corresponding to the target query result.
A risk control review method based on a decision tree, wherein the method includes:

Receiving user data of a user, where the user data includes preset information used to determine a service type of the user;

According to the preset information, retrieve the risk control model based on the decision tree from the preset storage file;

Perform a risk control review based on the user profile through the risk control model based on the decision tree.
The risk control review method based on a decision tree according to claim 4, wherein each query result contained in the query result code table is correspondingly set with a query result code;

The step of conducting a risk control review according to the user profile through the decision tree-based risk control model includes:

According to the user information, obtain the target query result corresponding to each interface code through the interface code corresponding to each node included in the risk control model;

Obtaining the target query result code corresponding to each target query result according to each target query result;

According to each target query result code, obtain the target risk control result corresponding to each target query result code, so as to obtain the risk control result corresponding to each node;

According to the risk control result corresponding to each of the nodes, the risk control audit of the user is performed through the risk control model based on the decision tree.
The risk control review method based on a decision tree according to claim 5, wherein, before the step of obtaining the target risk control result corresponding to each of the target query result codes, the method comprises:

Pre-query the risk control result corresponding to the node included in the risk control model through a pre-query method.
The risk control review method based on a decision tree according to claim 6, wherein the step of pre-querying the risk control results corresponding to the nodes included in the risk control model in a pre-query manner comprises:

Pre-query the risk control results corresponding to the nodes included in the risk control model according to the interface type corresponding to the credit reporting risk control interface.
The risk control review method based on a decision tree according to claim 4, wherein the preset information includes a user address and a target business name, and the decision tree-based review is retrieved from a preset storage file according to the preset information The steps of the risk control model include:

According to the user address and the target business name, a risk control model based on the decision tree of the business type corresponding to the user address and the target business name in advance is retrieved from a preset storage file.
A risk control model construction device based on a decision tree, which includes:

The display unit is used to display the preset interface code table, query result code table, and risk control result code table through the list page when the preset operation of constructing the risk control model is received; wherein, the interface code table includes multiple Interface code, the query result code table includes multiple query results, and the risk control result code table includes multiple risk control results;

The first receiving unit is configured to receive, through the list page, a target interface code determined from the interface code table, a target query result determined from the query result code table, and a target determined from the risk control result code table The selection instruction of the risk control result obtains the correspondence between the target interface code, the target query result, and the target risk control result;

The generating unit is configured to configure the target interface code, the target query result, and the target risk control result in nodes of a preset decision tree based on the corresponding relationship, and generate a risk control model based on the decision tree.
A risk control review device based on a decision tree, which includes:

The second receiving unit is configured to receive user data of the user, where the user data includes preset information used to determine the service type of the user;

The retrieval unit is configured to retrieve the risk control model based on the decision tree from the preset storage file according to the preset information;

The review unit is configured to conduct a risk control review based on the user profile through the risk control model based on the decision tree.
A computer device, wherein the computer device includes a memory and a processor connected to the memory; the memory is used to store a computer program; the processor is used to run the computer program stored in the memory to achieve the following step:

When the preset operation of constructing the risk control model is received, the preset interface code table, query result code table, and risk control result code table are displayed through the list page; wherein, the interface code table includes multiple interface codes, and The query result code table includes multiple query results, and the risk control result code table includes multiple risk control results;

Receiving, through the list page, a selection instruction for determining a target interface code from the interface code table, determining a target query result from the query result code table, and determining a target risk control result from the risk control result code table, Obtain the correspondence between the target interface code, the target query result, and the target risk control result;

Based on the corresponding relationship, the target interface code, the target query result, and the target risk control result are respectively configured at nodes of a preset decision tree to generate a risk control model based on the decision tree.
11. The computer device according to claim 11, wherein said receiving, through said list page, determining a target interface code from said interface code table, determining a target query result from said query result coding table, and receiving from said risk control The step of determining the selection instruction of the target risk control result in the result code table, and obtaining the corresponding relationship between the target interface code, the target query result, and the target risk control result includes:

Receiving a selection instruction for determining a target interface code from the interface code table to obtain the target interface code;

Displaying the query result code table associated with the target interface code through the list page;

Receiving a selection instruction for determining a target query result from the query result code table to obtain a target query result corresponding to the target interface code;

Displaying the risk control result code table associated with the target query result through the list page;

Receiving a selection instruction for determining a target risk control result from the risk control result coding table to obtain the target risk control result corresponding to the target query result;

The target interface code, the target query result, and the target risk control result are associated through a preset association method to obtain the relationship between the target interface code, the target query result, and the target risk control result Correspondence.
The computer device according to claim 12, wherein each query result contained in the query result code table corresponds to a query result code;

The step of receiving a selection instruction for determining a target risk control result from the risk control result coding table to obtain the target risk control result corresponding to the target query result includes:

Acquiring the target query result code corresponding to the target query result;

Displaying the risk control result code table associated with the target query result code through the list page;

A selection instruction for determining a target risk control result from the risk control result code table is received to obtain the target risk control result corresponding to the target query result code, so as to obtain the target risk control result corresponding to the target query result.
A computer device, wherein the computer device includes a memory and a processor connected to the memory; the memory is used to store a computer program; the processor is used to run the computer program stored in the memory to achieve the following step:

Receiving user data of a user, where the user data includes preset information used to determine a service type of the user;

According to the preset information, retrieve the risk control model based on the decision tree from the preset storage file;

Perform a risk control review based on the user profile through the risk control model based on the decision tree.
14. The computer device according to claim 14, wherein each query result contained in the query result code table is correspondingly set with a query result code;

The step of conducting a risk control review according to the user profile through the decision tree-based risk control model includes:

According to the user information, obtain the target query result corresponding to each interface code through the interface code corresponding to each node included in the risk control model;

Obtaining the target query result code corresponding to each target query result according to each target query result;

According to each target query result code, obtain the target risk control result corresponding to each target query result code, so as to obtain the risk control result corresponding to each node;

According to the risk control result corresponding to each of the nodes, the risk control audit of the user is performed through the risk control model based on the decision tree.
15. The computer device according to claim 15, wherein before the step of obtaining the target risk control result corresponding to each of the target query result codes, the step comprises:

Pre-query the risk control result corresponding to the node included in the risk control model through a pre-query method.
The risk control review method based on a decision tree according to claim 16, wherein the step of pre-querying the risk control results corresponding to the nodes included in the risk control model in a pre-query manner comprises:

Pre-query the risk control results corresponding to the nodes included in the risk control model according to the interface type corresponding to the credit reporting risk control interface.
The risk control review method based on a decision tree according to claim 14, wherein the preset information includes a user address and a target business name, and according to the preset information, the decision tree is retrieved from a preset storage file. The steps of the risk control model include:

According to the user address and the target business name, a risk control model based on the decision tree of the business type corresponding to the user address and the target business name in advance is retrieved from a preset storage file.
A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

When a preset operation of constructing a risk control model is received, the preset interface code table, query result code table, and risk control result code table are displayed through the list page; wherein, the interface code table includes multiple interface codes, and The query result code table includes multiple query results, and the risk control result code table includes multiple risk control results;

Receiving, through the list page, a selection instruction for determining a target interface code from the interface code table, determining a target query result from the query result code table, and determining a target risk control result from the risk control result code table, Obtain the correspondence between the target interface code, the target query result, and the target risk control result;

Based on the corresponding relationship, the target interface code, the target query result, and the target risk control result are respectively configured at nodes of a preset decision tree to generate a risk control model based on the decision tree.
A computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following steps are implemented:

Receiving user data of a user, where the user data includes preset information used to determine a service type of the user;

According to the preset information, retrieve the risk control model based on the decision tree from the preset storage file;

Perform a risk control review based on the user profile through the risk control model based on the decision tree.