Method, device, terminal and storage medium for generating target feature cluster

Abstract

The present application discloses a method, device, terminal and storage medium for generating a target feature cluster. The method includes: receiving a credit report, determining the feature information of credit risk control based on the credit report, and then determining the target atomic function corresponding to the feature information of credit risk control based on the feature information of credit risk control and a preset normalized feature generation model, and then generating a target feature cluster corresponding to the feature information of credit risk control based on the target atomic function and a preset normalized feature combination method. The present invention classifies the feature information of credit risk control through a preset normalized feature generation model to obtain different types of target atomic functions, and then combines different types of target atomic functions through a preset normalized feature combination method, and normalizes the output of large-scale, high-coverage credit features, which not only improves the generation efficiency of credit features, but also broadens the dimension and coverage of target feature clusters.

Description

Technical Field

[0001] The present application relates to the field of financial risk control technology, and in particular to a method, device, terminal and storage medium for generating a target feature cluster. Background Art

[0002] In the field of financial risk control, personal credit reports contain data that characterize personal credit status, such as credit card and loan information, credit inquiry record information, personal basic information, etc., and are also essential factors in all aspects of credit risk control. Therefore, how to quickly generate target feature clusters (i.e., features of various risk exposure directions of credit risk control) through personal credit reports has become an urgent problem to be solved.

[0003] At present, the target feature clusters are mainly constructed manually based on expert experience, that is, based on the expert experience of the banking system, the credit features in the credit report of the People's Bank of China are processed to form the target feature clusters.

[0004] However, the target feature cluster formed based on the above method has the problem of small feature dimension. Summary of the invention

[0005] The main purpose of the present application is to provide a method, device, terminal and storage medium for generating a target feature cluster to solve the problem that the target feature cluster formed in the related art has a small feature dimension.

[0006] In order to achieve the above objectives, in a first aspect, the present application provides a method for generating a target feature cluster, comprising:

[0007] Receive credit report;

[0008] Determine the characteristic information of credit risk control based on the credit report;

[0009] Based on the characteristic information of credit risk control and the preset normalized characteristic generation model, determine the target atomic function corresponding to the characteristic information of credit risk control;

[0010] Based on the target atomic function and the preset normalized feature combination method, a target feature cluster corresponding to the feature information of credit risk control is generated.

[0011] In a possible implementation, based on the characteristic information of credit risk control and a preset normalized characteristic generation model, a target atomic function corresponding to the characteristic information of credit risk control is determined, including:

[0012] Establish a preset normalized feature generation model;

[0013] Input the characteristic information of credit risk control into the preset normalized feature generation model to generate the target atomic function.

[0014] In a possible implementation, the target atomic function includes at least a first target atomic function, a second target atomic function, and a third target atomic function;

[0015] Input the characteristic information of credit risk control into the preset normalized feature generation model to generate the target atomic function, including:

[0016] The characteristic information of credit risk control is input into a preset normalized feature generation model to generate a first target atomic function, a second target atomic function and a third target atomic function, wherein the first target atomic function is used to store condition information corresponding to the characteristic information of credit risk control, the second target atomic function is used to store feature cluster direction information corresponding to the characteristic information of credit risk control, and the third target atomic function is used to store calculation method information corresponding to the characteristic information of credit risk control.

[0017] In a possible implementation, the preset normalized feature combination mode includes at least a first normalized feature combination mode and a second normalized feature combination mode, and the target feature cluster includes at least a first target feature cluster and a second target feature cluster;

[0018] Based on the target atomic function and the preset normalized feature combination method, the target feature cluster corresponding to the feature information of credit risk control is generated, including:

[0019] Combining the first target atomic function, the second target atomic function and the third target atomic function based on the first normalized feature combination method to obtain a first target feature cluster;

[0020] The first target atomic function, the second target atomic function and the third target atomic function are combined based on the second normalized feature combination method to obtain a second target feature cluster.

[0021] In a possible implementation, the condition information includes at least first condition information, the calculation method information includes at least first calculation method information, and the first target feature cluster includes a plurality of sub-first target feature clusters;

[0022] The first target atomic function, the second target atomic function and the third target atomic function are combined based on the first normalized feature combination method to obtain a first target feature cluster, including:

[0023] Selecting any feature cluster direction information in the second target atomic function as the first feature cluster direction information;

[0024] Selecting first condition information and first calculation method information matching the first feature cluster direction information from the first target atomic function and the third target atomic function respectively;

[0025] The first feature cluster direction information, the first condition information and the first calculation method information are combined to obtain a plurality of sub-first target feature clusters.

[0026] In a possible implementation, the condition information includes at least second condition information, the calculation method information includes at least second calculation method information, and the second target feature cluster includes a plurality of sub-second target feature clusters;

[0027] The first target atomic function, the second target atomic function and the third target atomic function are combined based on the second normalized feature combination method to obtain a second target feature cluster, including:

[0028] Selecting at least two feature cluster direction information in the second target atomic function as second feature cluster direction information;

[0029] Selecting second condition information and second calculation method information matching the second feature cluster direction information from the first target atomic function and the third target atomic function respectively;

[0030] The second feature cluster direction information, the second condition information and the second calculation method information are combined to obtain a plurality of sub-second target feature clusters.

[0031] In a possible implementation, based on the credit report, characteristic information of credit risk control is determined, including:

[0032] Read data from credit reports;

[0033] The data in the credit report is subjected to feature analysis, feature cleaning, code mapping and feature verification in turn to obtain feature information for credit risk control.

[0034] In a second aspect, an embodiment of the present invention provides a device for generating a target feature cluster, comprising:

[0035] A receiving module, used for receiving a credit report;

[0036] Data processing module, used to determine the characteristic information of credit risk control based on the credit report;

[0037] A function generation module, used to determine the target atomic function corresponding to the characteristic information of credit risk control based on the characteristic information of credit risk control and a preset normalized characteristic generation model;

[0038] The feature cluster generation module is used to generate a target feature cluster corresponding to the feature information of credit risk control based on the target atomic function and the preset normalized feature combination method.

[0039] In a third aspect, an embodiment of the present invention provides a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any of the above methods for generating a target feature cluster when executing the computer program.

[0040] In a fourth aspect, an embodiment of the present invention provides 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 steps of any of the above methods for generating a target feature cluster are implemented.

[0041] The embodiment of the present invention provides a method, device, terminal and storage medium for generating a target feature cluster, including: receiving a credit report, determining feature information of credit risk control based on the credit report, then determining the target atomic function corresponding to the feature information of credit risk control based on the feature information of credit risk control and a preset normalized feature generation model, and then generating a target feature cluster corresponding to the feature information of credit risk control based on the target atomic function and a preset normalized feature combination method. The present invention classifies the feature information of credit risk control through a preset normalized feature generation model to obtain different types of target atomic functions, and then combines different types of target atomic functions through a preset normalized feature combination method to normalize and output large quantities of credit features with high coverage, which not only improves the generation efficiency of credit features, but also broadens the dimension and coverage of the target feature cluster. In addition, it also provides a feature basis for the risk control model to improve the evaluation effect of the risk control model. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The drawings constituting a part of this application are used to provide a further understanding of this application, so that other features, purposes and advantages of this application become more obvious. The schematic embodiment drawings and their descriptions of this application are used to explain this application and do not constitute an improper limitation on this application. In the drawings:

[0043] Figure 1 is a flow chart of a method for generating a target feature cluster provided by an embodiment of the present invention;

[0044] Figure 2 is a flow chart of an implementation of a method for generating a target feature cluster provided by another embodiment of the present invention;

[0045] Figure 3 is a schematic diagram of the structure of the target atomic function provided by an embodiment of the present invention;

[0046] Figure 4 It is a structural schematic diagram of a device for generating a target feature cluster provided by an embodiment of the present invention;

[0047] Figure 5 is a schematic diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein.

[0050] It should be understood that in various embodiments of the present invention, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0051] It should be understood that in the present invention, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0052] It should be understood that in the present invention, "plurality" refers to two or more than two. "And / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "Contains A, B and C", "Contains A, B, C" means that A, B, and C are all included, "Contains A, B or C" means that one of A, B, and C is included, and "Contains A, B and / or C" means that any one, any two, or any three of A, B, and C are included.

[0053] It should be understood that in the present invention, "B corresponding to A", "B corresponding to A", "A corresponds to B" or "B corresponds to A" means that B is associated with A and B can be determined based on A. Determining B based on A does not mean determining B based only on A, but B can also be determined based on A and / or other information. A and B match when the similarity between A and B is greater than or equal to a preset threshold.

[0054] Depending on the context, "if" as used herein may be interpreted as "when" or "when" or "in response to determining" or "in response to detecting."

[0055] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0056] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0057] In one embodiment, Figure 1 As shown, a method for generating a target feature cluster is provided, comprising the following steps:

[0058] Step S101: receiving a credit report;

[0059] Step S102: Determine characteristic information of credit risk control based on the credit report.

[0060] In one embodiment, in combination Figure 2 , after receiving the credit report, it is necessary to first read the data in the credit report, where the data in the credit report can be in XML or JSON format. Taking the data in the credit report in XML format as an example, the XML format data is sequentially subjected to feature parsing, feature cleaning, coding mapping and feature verification (which can be first- and second-generation feature consistency verification) to obtain the characteristic information of credit risk control. Among them, the characteristic information of credit risk control refers to the characteristic information used for credit risk control evaluation, which can be query information, credit product attribute information, etc. In this application, after the characteristic information of credit risk control is obtained, the characteristic information of credit risk control is stored in the data middle platform for subsequent services.

[0061] Step S103: Based on the characteristic information of credit risk control and a preset normalized characteristic generation model, determining a target atomic function corresponding to the characteristic information of credit risk control;

[0062] Step S104: Based on the target atomic function and the preset normalized feature combination method, a target feature cluster corresponding to the feature information of credit risk control is generated.

[0063] Specifically, when establishing a scientific and standardized database, some specifications need to be met to optimize the data storage method. Therefore, paradigms are widely used. Paradigms refer to the specifications followed when the database stores data. The preset normalized feature generation model in this application sets specific data storage specifications according to specific embodiments to meet the storage of characteristic information of credit risk control. Among them, the storage specifications in the preset normalized feature generation model can be adjusted based on different information, and are not specifically limited here.

[0064] On the basis of the above-mentioned embodiments, the characteristic information of the credit risk control in the present application is generated through a preset normalized feature generation model to generate different types of target atomic functions (i.e., atomic functions), wherein the target atomic function is a feature set for storing specific attribute information corresponding to the storage specification generated based on the storage specification in the preset normalized feature generation model. Afterwards, different types of target atomic functions are combined in a preset normalized feature combination method (i.e., normalized feature strategy combination) to generate a target feature cluster. Among them, the target feature cluster refers to a feature set formed by combining multiple features, and the target feature cluster includes multiple feature clusters, such as feature cluster 1, feature cluster 2, feature cluster 3...feature cluster N, where N is a positive integer greater than 3.

[0065] The embodiment of the present invention provides a method for generating a target feature cluster, including: receiving a credit report, determining feature information of credit risk control based on the credit report, then determining a target atomic function corresponding to the feature information of credit risk control based on the feature information of credit risk control and a preset normalized feature generation model, and then generating a target feature cluster corresponding to the feature information of credit risk control based on the target atomic function and a preset normalized feature combination method. The present invention classifies the feature information of credit risk control through a preset normalized feature generation model to obtain different types of target atomic functions, and then combines different types of target atomic functions through a preset normalized feature combination method, and normalizes the output of large-volume, high-coverage credit features, which not only improves the efficiency of generating credit features, but also broadens the dimension and coverage of the target feature cluster. In addition, it also provides a feature basis for the risk control model to improve the evaluation effect of the risk control model.

[0066] In one embodiment, step S103 includes:

[0067] Step S201: establishing a preset normalized feature generation model;

[0068] Step S202: input the characteristic information of credit risk control into a preset normalized feature generation model to generate a target atomic function.

[0069] Among them, the target atomic function includes at least a first target atomic function, a second target atomic function and a third target atomic function, that is, the present application is not limited to the above three types of target atomic functions, and the target atomic function may also include a fourth target atomic function, a fifth target atomic function, etc., which varies according to the storage specifications set in the preset normalized feature generation model.

[0070] Further, the target atomic function includes the first target atomic function, the second target atomic function and the third target atomic function, and the preset normalized feature generation model is the CTF normalized feature generation framework as an example to illustrate the process of generating the target atomic function. Specifically, the CTF normalized feature generation framework includes three types of data storage modules, namely, condition-feature generation condition module, target-feature cluster direction module and function-feature calculation method module. The feature information of credit risk control is input into the CTF normalized feature generation framework, and the first target atomic function corresponding to the condition-feature generation condition module, the second target atomic function corresponding to the target-feature cluster direction module, and the third target atomic function corresponding to the function-feature calculation method module are generated respectively. Among them, the first target atomic function is used to store the condition information corresponding to the feature information of credit risk control, and the condition information includes the first condition information and the second condition information; the second target atomic function is used to store the feature cluster direction information corresponding to the feature information of credit risk control; the third target atomic function is used to store the calculation method information corresponding to the feature information of credit risk control, and the calculation method information includes the first calculation method information and the second condition information.

[0071] The following takes the characteristic information of credit risk control as the attribute information of credit products as an example to illustrate the steps of generating the target atomic function, which are as follows:

[0072] 1. After receiving the credit report, read the XML format data in the credit report, and perform feature parsing, feature cleaning, code mapping and feature verification on the XML format data in turn to obtain the credit product attribute information shown in Table 1.

[0073] Table 1 Credit product attribute information

[0074]

[0075] Among them, the credit product attribute information is classified by loan product type, and the parameter indicators in each type of loan product information are the same, including the lending institution type, loan type, term, credit amount, and performance term.

[0076] 2. Input the credit product attribute information into the CTF normalized feature generation framework to generate the first target atomic function, the second target atomic function and the third target atomic function.

[0077] Specific, combined Figure 3 The first target atomic function stores feature generation condition data, that is, the prerequisites that feature generation depends on, single or multiple combinations, such as loan status, loan type, lending institution type, etc.; the second target atomic function stores feature cluster direction data, that is, the direction of feature calculation, such as: loan amount, balance, number of periods, etc.; the third target atomic function stores feature calculation method data, that is, enumerating the numerical size of each feature in different representation dimensions, such as sum (and calculation), max (maximum value calculation), etc.

[0078] In one embodiment, step S104 includes:

[0079] Step S301: combining the first target atomic function, the second target atomic function and the third target atomic function based on the first normalized feature combination method to obtain a first target feature cluster;

[0080] Step S302: combining the first target atomic function, the second target atomic function and the third target atomic function based on the second normalized feature combination method to obtain a second target feature cluster.

[0081] Among them, the preset normalized feature combination method includes at least the first normalized feature combination method and the second normalized feature combination method, and the target feature cluster includes at least the first target feature cluster and the second target feature cluster, that is, the preset normalized feature combination method and the target feature cluster can be multiple categories, not limited to the above two categories, and can be set according to specific circumstances.

[0082] The following takes the preset normalized feature combination mode including the first normalized feature combination mode and the second normalized feature combination mode, the target feature cluster including the first target feature cluster and the second target feature cluster, and the first target feature cluster including multiple sub-first target feature clusters, and the second target feature cluster including multiple sub-second target feature clusters as an example to illustrate the generation process of the first target feature cluster and the second target feature cluster, as follows:

[0083] When the first normalized feature combination method is adopted, it is necessary to first select any feature cluster direction information in the second target atomic function as the first feature cluster direction information, and then select the first condition information and the first calculation method information matching the first feature cluster direction information in the first target atomic function and the third target atomic function respectively, and then combine the first feature cluster direction information, the first condition information and the first calculation method information to obtain multiple sub-first target feature clusters.

[0084] In a specific embodiment, Figure 3 Taking the first atomic function, the second atomic function and the third atomic function corresponding to the credit product attribute information as an example, the steps of generating multiple sub-first target feature clusters by using the first normalized feature combination method are described:

[0085] 1. Select the loan amount as the feature cluster direction information in the second target atomic function;

[0086] 2. Select the first condition information and the first calculation method information that match the loan amount, which are the loan type and the calculated maximum value (sum), respectively, where the loan type includes type A, type B, and type C;

[0087] 3. By combining the loan amount, loan type and calculated maximum value (sum) (which can be any combination including the loan amount), multiple sub-first target feature clusters can be obtained, such as loan amount, type A and calculated maximum value (sum), loan amount, type B and calculated maximum value (sum), loan amount, type C and calculated maximum value (sum), loan amount, type B, type C and calculated maximum value (sum), etc.

[0088] When the second normalized feature combination method is adopted, it is necessary to first select at least two feature cluster direction information in the second target atomic function as the second feature cluster direction information, and then select the second condition information and the second calculation method information matching the second feature cluster direction information in the first target atomic function and the third target atomic function respectively, and then combine the second feature cluster direction information, the second condition information and the second calculation method information to obtain multiple sub-second target feature clusters.

[0089] In a specific embodiment, Figure 3 Taking the first atomic function, the second atomic function and the third atomic function corresponding to the credit product attribute information as an example, the steps of generating multiple sub-second target feature clusters by using the second normalized feature combination method are described:

[0090] 1. Select the loan amount and loan period as the feature cluster direction information in the second target atomic function;

[0091] 2. Select the second condition information and the second calculation method information that match the loan amount and the number of loan terms, which are the loan status and the calculation minimum value (min), respectively. The loan status includes loan status A and loan status B;

[0092] 3. By combining the loan amount, loan term, loan status and calculated minimum value (min) (which can be any combination of loan amount and loan term), multiple sub-second target feature clusters can be obtained, such as loan amount, loan term, loan status B and calculated minimum value (min), loan amount, loan term, loan status A, loan status B and calculated minimum value (min), etc.

[0093] It should be noted here that within the combination rule range of the first normalized feature combination method and the second normalized feature combination method, the first atomic function, the second atomic function and the third atomic function can be arbitrarily combined, and the above two specific embodiments are only examples.

[0094] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0095] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0096] Figure 4 The schematic diagram of the structure of a target feature cluster generation device provided by an embodiment of the present invention is shown. For the convenience of explanation, only the part related to the embodiment of the present invention is shown. The target feature cluster generation device includes a receiving module 41, a data processing module 42, a function generation module 43 and a feature cluster generation module 44, which are as follows:

[0097] A receiving module 41, used for receiving a credit report;

[0098] A data processing module 42 is used to determine characteristic information of credit risk control based on the credit report;

[0099] The function generation module 43 is used to determine the target atomic function corresponding to the characteristic information of the credit risk control based on the characteristic information of the credit risk control and the preset normalized characteristic generation model;

[0100] The feature cluster generation module 44 is used to generate a target feature cluster corresponding to the feature information of credit risk control based on the target atomic function and a preset normalized feature combination method.

[0101] In a possible implementation, the function generation module 43 includes:

[0102] The model building submodule is used to build a preset normalized feature generation model;

[0103] The function generation submodule is used to input the characteristic information of credit risk control into the preset normalized feature generation model to generate the target atomic function.

[0104] In a possible implementation, the target atomic function includes at least a first target atomic function, a second target atomic function, and a third target atomic function;

[0105] The function generation submodules include:

[0106] A function generation unit is used to input the characteristic information of credit risk control into a preset normalized feature generation model to generate a first target atomic function, a second target atomic function and a third target atomic function, wherein the first target atomic function is used to store condition information corresponding to the characteristic information of credit risk control, the second target atomic function is used to store feature cluster direction information corresponding to the characteristic information of credit risk control, and the third target atomic function is used to store calculation method information corresponding to the characteristic information of credit risk control.

[0107] In a possible implementation, the preset normalized feature combination mode includes at least a first normalized feature combination mode and a second normalized feature combination mode, and the target feature cluster includes at least a first target feature cluster and a second target feature cluster;

[0108] The feature cluster generation module 44 includes:

[0109] A first feature cluster generation submodule is used to combine the first target atomic function, the second target atomic function and the third target atomic function based on a first normalized feature combination method to obtain a first target feature cluster;

[0110] The second feature cluster generation submodule is used to combine the first target atomic function, the second target atomic function and the third target atomic function based on the second normalized feature combination method to obtain a second target feature cluster.

[0111] In a possible implementation, the condition information includes at least first condition information, the calculation method information includes at least first calculation method information, and the first target feature cluster includes a plurality of sub-first target feature clusters;

[0112] The first feature cluster generation submodule includes:

[0113] A first selection unit is used to select any feature cluster direction information in the second target atomic function as the first feature cluster direction information;

[0114] A first matching unit is used to select first condition information and first calculation method information that match the first feature cluster direction information from the first target atomic function and the third target atomic function respectively;

[0115] The first combining unit is used to combine the first feature cluster direction information, the first condition information and the first calculation method information to obtain a plurality of sub-first target feature clusters.

[0116] In a possible implementation, the condition information includes at least second condition information, the calculation method information includes at least second calculation method information, and the second target feature cluster includes a plurality of sub-second target feature clusters;

[0117] The second feature cluster generation submodule includes:

[0118] A second selection unit is used to select at least two feature cluster direction information in the second target atomic function as second feature cluster direction information;

[0119] A second matching unit is used to select second condition information and second calculation method information that match the second feature cluster direction information from the first target atomic function and the third target atomic function respectively;

[0120] The second combining unit is used to combine the second feature cluster direction information, the second condition information and the second calculation method information to obtain a plurality of sub-second target feature clusters.

[0121] In a possible implementation, the data processing module 42 includes:

[0122] The data reading submodule is used to read the data in the credit report;

[0123] The data processing submodule is used to perform feature analysis, feature cleaning, code mapping and feature verification on the data in the credit report in sequence to obtain feature information for credit risk control.

[0124] Figure 5 is a schematic diagram of a terminal provided by an embodiment of the present invention. Figure 5 As shown, the terminal 5 of this embodiment includes: a processor 50, a memory 51, and a computer program 52 stored in the memory 51 and executable on the processor 50. When the processor 50 executes the computer program 52, the steps in the above-mentioned generation method of each target feature cluster are implemented, for example Figure 1 Alternatively, when the processor 50 executes the computer program 52, the functions of each module / unit in the above-mentioned target feature cluster generation device embodiment are realized, for example Figure 4 Functionality of the modules / units 41 to 44 shown.

[0125] The present invention also provides a readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, it is used to implement the methods provided in the above-mentioned various embodiments.

[0126] Among them, the readable storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application-specific integrated circuit (Application Specific Integrated Circuits, abbreviated as: ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the readable storage medium can also exist in a communication device as discrete components. The readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0127] The present invention also provides a program product, which includes an execution instruction, which is stored in a readable storage medium. At least one processor of a device can read the execution instruction from the readable storage medium, and at least one processor executes the execution instruction so that the device implements the methods provided in the above various embodiments.

[0128] In the embodiments of the above-mentioned devices, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the present invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0129] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A method for generating a target feature cluster, characterized in that: include: Receive credit report; Determine characteristic information of credit risk control based on the credit report; Based on the characteristic information of the credit risk control and a preset normalized characteristic generation model, determining a target atomic function corresponding to the characteristic information of the credit risk control; Based on the target atomic function and the preset normalized feature combination method, a target feature cluster corresponding to the feature information of the credit risk control is generated; The step of determining the target atomic function corresponding to the characteristic information of the credit risk control based on the characteristic information of the credit risk control and a preset normalized characteristic generation model includes: Establishing the preset normalized feature generation model, wherein the preset normalized feature generation model sets the data storage specification to meet the storage of feature information of credit risk control; Inputting the characteristic information of the credit risk control into the preset normalized feature generation model to generate the target atomic function; The target atomic function at least includes a first target atomic function, a second target atomic function and a third target atomic function; The step of inputting the characteristic information of the credit risk control into the preset normalized characteristic generation model to generate the target atomic function includes: The characteristic information of the credit risk control is input into the preset normalized feature generation model to generate the first target atomic function, the second target atomic function and the third target atomic function, wherein the first target atomic function is used to store the condition information corresponding to the characteristic information of the credit risk control, the second target atomic function is used to store the feature cluster direction information corresponding to the characteristic information of the credit risk control, and the third target atomic function is used to store the calculation method information corresponding to the characteristic information of the credit risk control.

2. The method for generating a target feature cluster according to claim 1, characterized in that: The preset normalized feature combination mode includes at least a first normalized feature combination mode and a second normalized feature combination mode, and the target feature cluster includes at least a first target feature cluster and a second target feature cluster; The step of generating a target feature cluster corresponding to the feature information of the credit risk control based on the target atomic function and a preset normalized feature combination method includes: Combining the first target atomic function, the second target atomic function and the third target atomic function based on the first normalized feature combination method to obtain the first target feature cluster; The first target atomic function, the second target atomic function and the third target atomic function are combined based on the second normalized feature combination method to obtain the second target feature cluster.

3. The method for generating a target feature cluster as claimed in claim 2, characterized in that: The condition information includes at least first condition information, the calculation method information includes at least first calculation method information, and the first target feature cluster includes a plurality of sub-first target feature clusters; The first target atomic function, the second target atomic function and the third target atomic function are combined based on the first normalized feature combination mode to obtain the first target feature cluster, including: Selecting any feature cluster direction information in the second target atomic function as the first feature cluster direction information; Selecting first condition information and first calculation method information matching the first feature cluster direction information from the first target atomic function and the third target atomic function respectively; The first feature cluster direction information, the first condition information and the first calculation method information are combined to obtain the multiple sub-first target feature clusters.

4. The method for generating a target feature cluster as claimed in claim 2, characterized in that: The condition information includes at least second condition information, the calculation method information includes at least second calculation method information, and the second target feature cluster includes a plurality of sub-second target feature clusters; The combining of the first target atomic function, the second target atomic function and the third target atomic function based on the second normalized feature combination method to obtain the second target feature cluster includes: Selecting at least two feature cluster direction information in the second target atomic function as second feature cluster direction information; Selecting second condition information and second calculation method information matching the second feature cluster direction information from the first target atomic function and the third target atomic function respectively; The second feature cluster direction information, the second condition information and the second calculation method information are combined to obtain the plurality of sub-second target feature clusters.

5. The method for generating a target feature cluster according to any one of claims 1 to 4, characterized in that: Determining the characteristic information of credit risk control based on the credit report includes: Reading data from the credit report; The data in the credit report is subjected to feature analysis, feature cleaning, code mapping and feature verification in sequence to obtain feature information of the credit risk control.

6. A device for generating a target feature cluster, characterized in that: include: A receiving module, used for receiving a credit report; A data processing module, used to determine characteristic information of credit risk control based on the credit report; A function generation module, used to determine the target atomic function corresponding to the characteristic information of the credit risk control based on the characteristic information of the credit risk control and a preset normalized characteristic generation model; A feature cluster generation module, used to generate a target feature cluster corresponding to the feature information of the credit risk control based on the target atomic function and a preset normalized feature combination method; The step of determining the target atomic function corresponding to the characteristic information of the credit risk control based on the characteristic information of the credit risk control and a preset normalized characteristic generation model includes: Establishing the preset normalized feature generation model, wherein the preset normalized feature generation model sets the data storage specification to meet the storage of feature information of credit risk control; Inputting the characteristic information of the credit risk control into the preset normalized feature generation model to generate the target atomic function; The target atomic function at least includes a first target atomic function, a second target atomic function and a third target atomic function; The step of inputting the characteristic information of the credit risk control into the preset normalized characteristic generation model to generate the target atomic function includes: The characteristic information of the credit risk control is input into the preset normalized feature generation model to generate the first target atomic function, the second target atomic function and the third target atomic function, wherein the first target atomic function is used to store the condition information corresponding to the characteristic information of the credit risk control, the second target atomic function is used to store the feature cluster direction information corresponding to the characteristic information of the credit risk control, and the third target atomic function is used to store the calculation method information corresponding to the characteristic information of the credit risk control.

7. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method for generating a target feature cluster according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method for generating a target feature cluster according to any one of claims 1 to 5 are implemented.

Images