Credit Scoring Mapping Method, Device and Storage Medium Based on Cumulative Bad Debt Rate

Abstract

The present invention provides a credit score mapping method, device and storage medium based on the cumulative bad debt rate, including: respectively obtaining the first type of third-party score data and the second type of third-party score data, where the first type of third-party score data and the second type of third-party score data respectively include the score data of multiple users; respectively calculating the cumulative default rate of users from the first type of third-party score data and the second type of third-party score data; connecting the first type of third-party score data and the second type of third-party score data according to the cumulative default rate to obtain the mapping from the second type of third-party score data to the first type of third-party score data; taking the second type of third-party score data as the input and the first type of third-party score data as the output to obtain the corresponding functional relationship. The technical solution provided by the present invention can realize the mapping between scores, that is, determine the other third-party scores of the current score within the maximum and minimum values of the current score, improve the accuracy, and make the online model more stable.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a credit score mapping method, device, and storage medium based on the cumulative bad debt rate. Background Art

[0002] A credit risk scoring model is a traditional credit risk quantification model. The model is trained based on the basic information, behavioral information, authorized information, and externally connected information of borrowers, through feature transformation and using statistical methods such as logistic regression algorithms or machine learning algorithms such as LightBGM and XGBoost, to obtain a probability or score that can measure the credit risk of borrowers. For different third-party data providers, various third-party data scoring products are developed based on the nature of the product itself, its own unique data, and unique training data. For users, although different third-party products have differences in precision rates for the current used samples, they can still be used to evaluate the credit risk of borrowers.

[0003] Due to risks such as data supervision, data source changes, and supplier development changes, the currently used third-party scores are invalid, or cannot be used continuously due to cost reasons, procurement restrictions, etc. However, the online production environment still requires the existence of this score as part of the risk interception strategy. Therefore, there is an urgent need for a technical solution that can achieve mapping between scores, that is, find other third-party scores that can replace the current score within the maximum and minimum values of the current score, and at the same time, try not to reduce the accuracy after replacement, making the online model more stable. Summary of the Invention

[0004] Embodiments of the present invention provide a credit score mapping method, device, and storage medium based on the cumulative bad debt rate, which can achieve mapping between scores, that is, determine other third-party scores of the current score within the maximum and minimum values of the current score, improve the accuracy, and make the online model more stable.

[0005] In a first aspect of the embodiments of the present invention, a credit score mapping method based on the cumulative bad debt rate is provided, including:

[0006] Obtain first-party third-party score data and second-party third-party score data respectively, where the first-party third-party score data and the second-party third-party score data respectively include score data of multiple users;

[0007] Calculate the cumulative default rate of users for the first-party third-party score data and the second-party third-party score data respectively;

[0008] Connect the first-party third-party score data and the second-party third-party score data according to the cumulative default rate to obtain a mapping from the second-party third-party score data to the first-party third-party score data;

[0009] Taking the second type of third-party scoring data as input and the first type of third-party scoring data as output to obtain the corresponding functional relationship.

[0010] Optionally, in a possible implementation manner of the first aspect, in the step of calculating the cumulative default rate of the user from the first type of third-party scoring data and the second type of third-party scoring data respectively, it further includes:

[0011] Sorting the scoring data of the users in the first type of third-party scoring data and the second type of third-party scoring data respectively to obtain the first type of third-party scoring data and the second type of third-party scoring data with a preset order;

[0012] Calculating the corresponding cumulative default rates from the first type of third-party scoring data and the second type of third-party scoring data with a preset order respectively.

[0013] Optionally, in a possible implementation manner of the first aspect, in the step of respectively obtaining the first type of third-party scoring data and the second type of third-party scoring data, where the first type of third-party scoring data and the second type of third-party scoring data respectively include the scoring data of multiple users, it includes:

[0014] The first type of third-party scoring data is The second type of third-party scoring data is Wherein, is the scoring data of the nth user in the first type of third-party scoring data, is the scoring data of the Nth user in the first type of third-party scoring data, is the scoring data of the nth user in the second type of third-party scoring data, is the scoring data of the Nth user in the second type of third-party scoring data.

[0015] Optionally, in a possible implementation manner of the first aspect, in the step of sorting the scoring data of the users in the first type of third-party scoring data and the second type of third-party scoring data respectively to obtain the first type of third-party scoring data and the second type of third-party scoring data with a preset order, it includes:

[0016] Respectively obtaining the user label sequences of the users in the first type of third-party scoring data and the second type of third-party scoring data. The user label sequence is Y = [y1, y2,..., y n ,..., y N , where 0 is the first preset type of customer, 1 is the second preset type of customer, and y n is the user label of the nth user, y N is the user label of the Nth user, and y n ∈{0, 1};

[0017] Sort the first type of third-party scoring data and the second type of third-party scoring data based on the user tag sequence to obtain the sorted first type of third-party scoring data The second type of third-party scoring data Among them, is the scoring data of the nth user in the updated first type of third-party scoring data, is the scoring data of the Nth user in the updated first type of third-party scoring data, is the scoring data of the nth user in the updated second type of third-party scoring data, is the scoring data of the Nth user in the updated second type of third-party scoring data.

[0018] Optionally, in a possible implementation manner of the first aspect, in the step of calculating the corresponding cumulative default rates for the first type of third-party scoring data and the second type of third-party scoring data with a preset order respectively, it includes:

[0019] For the first type of third-party scoring data and the second type of third-party scoring data Calculate respectively to obtain the cumulative default rate sequences of the first type of third-party scoring data and the second type of third-party scoring data and Among them is the proportion of bad customers with a third-party score S1 below Among them is the proportion of bad customers with a third-party score S1 below is the proportion of bad customers with a third-party score S2 below is the proportion of bad customers with a third-party score S2 below

[0020]

[0021] Optionally, in a possible implementation manner of the first aspect, in the step of connecting the first type of third-party scoring data and the second type of third-party scoring data according to the cumulative default rate to obtain the mapping from the second type of third-party scoring data to the first type of third-party scoring data, it includes:

[0021] Use the sequence of the said cumulative default rate as the connection key to connect the first type of third-party scoring data and the second type of third-party scoring data, thereby forming a mapping from the third-party score S2 sequence to the third-party score S1 sequence.

[0022] Optionally, in a possible implementation manner of the first aspect, in the step of obtaining the corresponding functional relationship with the second type of third-party scoring data as the input and the first type of third-party scoring data as the output, it includes:

[0023] Based on the mapping from the three - party score S2 sequence to the three - party score S1 sequence, using the three - party score S2 sequence as the input and the three - party score S1 sequence as the output, a polynomial regression algorithm is used for fitting, and finally the functional relationship f(x) from the three - party score S2 to the three - party score S1 is obtained, that is, S1 = f(S2).

[0024] Optionally, in a possible implementation manner of the first aspect, where N is the number of the same users in the first three - party score data and the second three - party score data.

[0025] In the second aspect of the embodiments of the present invention, a credit score mapping device based on the cumulative bad debt rate is provided, including:

[0026] An acquisition module, configured to respectively acquire the first three - party score data and the second three - party score data, where the first three - party score data and the second three - party score data respectively include the score data of multiple users;

[0027] A calculation module, configured to respectively calculate the cumulative default rate of users from the first three - party score data and the second three - party score data;

[0028] A connection module, configured to connect the first three - party score data and the second three - party score data according to the cumulative default rate to obtain the mapping from the second three - party score data to the first three - party score data;

[0029] A function obtaining module, configured to obtain the corresponding functional relationship with the second three - party score data as the input and the first three - party score data as the output.

[0030] In the third aspect of the embodiments of the present invention, a storage medium is provided, in which a computer program is stored, and when the computer program is executed by a processor, it is used to implement the method described in the first aspect and various possible designs of the first aspect of the present invention.

[0031] A credit score mapping method, device and storage medium based on the cumulative bad debt rate provided by the present invention can realize the mapping between scores, that is, determine other three - party scores of the current score within the maximum and minimum values of the current score, improve the accuracy, and make the online model more stable. And by using the method of cumulative calculation of the bad debt rate to realize the effective mapping between different three - party scores, the method is simple and efficient, and does not affect the current credit score usage threshold when switching scores, and is unaware of the switching of the current score in terms of strategy usage. In actual usage scenarios, the present invention can make the score switching process more stable and does not affect the normal operation of online services. Description of the Drawings

[0032] Figure 1Flow chart of the first implementation manner of the credit score mapping method based on the cumulative bad debt rate;

[0033] Figure 2 Flow chart of the second implementation manner of the credit score mapping method based on the cumulative bad debt rate;

[0034] Figure 3 Structural diagram of the first implementation manner of the credit score mapping device based on the cumulative bad debt rate. Detailed implementation manner

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the scope of protection of the present invention.

[0036] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein.

[0037] It should be understood that in various embodiments of the present invention, the magnitudes of the sequence numbers of the various processes do not mean the order of execution, and the order of execution of the various processes should be determined by their functions and internal logics, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0038] It should be understood that in the present invention, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0039] It should be understood that in the present invention, "a plurality of" means two or more. "And / or" is merely a correlative relationship describing associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "Including A, B, and C" and "including A, B, C" mean that all of A, B, and C are included. "Including A, B, or C" means including any one of A, B, and C. "Including A, B, and / or C" means including any one, any two, or all three of A, B, and C.

[0040] It should be understood that in the present invention, "B corresponding to A", "B corresponding to A relatively", "A corresponding to B relatively", or "B corresponding to A relatively" means that B is associated with A, and B can be determined according to A. Determining B according to A does not mean determining B only according to A, and B can also be determined according to A and / or other information. The matching of A and B means that the similarity between A and B is greater than or equal to a preset threshold.

[0041] Depending on the context, as used herein, "if" can be interpreted as "when", "while", "in response to determination", or "in response to detection".

[0042] The technical solution of the present invention will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0043] The present invention provides a credit score mapping method based on the cumulative bad debt rate, as Figure 1 and Figure 2 shown, including:

[0044] Step S110, respectively obtain the first type of third-party score data and the second type of third-party score data, where the first type of third-party score data and the second type of third-party score data respectively include the score data of multiple users.

[0045] The technical solution provided by the present invention, in step S110, includes:

[0046] The first type of third-party score data is The second type of third-party score data is wherein, is the score data of the nth user in the first type of third-party score data, is the score data of the Nth user in the first type of third-party score data, is the score data of the nth user in the second type of third-party score data, It is the rating data of the Nth user in the second type of third-party rating data. The present invention represents the first type of third-party rating data and the second type of third-party rating data in the form of a set, where the first type of third-party rating data and the second type of third-party rating data can be from different systems.

[0047] In a possible implementation, assume that there are two types of third-party ratings S1 and S2, and there are a certain number of the same users between them. These two types of third-party rating sequences are respectively represented as and where N represents the number of cross-samples of the two types of third-party ratings.

[0048] In a possible implementation, where N is the number of the same users in the first type of third-party rating data and the second type of third-party rating data.

[0049] Step S120: Calculate the cumulative default rate of users for the first type of third-party rating data and the second type of third-party rating data respectively.

[0050] In step S120, it further includes:

[0051] Sort the rating data of users in the first type of third-party rating data and the second type of third-party rating data respectively to obtain the first type of third-party rating data and the second type of third-party rating data with a preset order. The technical solution provided by the present invention will sort the rating data of users in the first type of third-party rating data and the second type of third-party rating data respectively.

[0052] In the step of sorting the rating data of users in the first type of third-party rating data and the second type of third-party rating data respectively to obtain the first type of third-party rating data and the second type of third-party rating data with a preset order, it includes:

[0053] Respectively obtain the user label sequences of users in the first type of third-party rating data and the second type of third-party rating data. The user label sequence is Y = [y1, y2,..., y n ,..., y N , where 0 is the first preset type of customer, 1 is the second preset type of customer, y n is the user label of the nth user, y N is the user label of the Nth user, and y n ∈{0, 1}. The first preset type of customer can be a good customer (a customer with preset positive behaviors), and the second preset type of customer can be a bad customer (a customer with preset negative behaviors).

[0054] In a possible implementation, use the third-party rating S2 to replace the third-party rating S1. The user label sequence of the third-party rating S1 is Y = [y1, y2,..., yn ,..., y N , y n ∈ {0, 1}, where 0 represents a good customer and 1 represents a bad customer.

[0055] Sort the first type of third-party scoring data and the second type of third-party scoring data based on the user label sequence to obtain the sorted first type of third-party scoring data The second type of third-party scoring data Among them, is the scoring data of the nth user in the updated first type of third-party scoring data, is the scoring data of the Nth user in the updated first type of third-party scoring data, is the scoring data of the nth user in the updated second type of third-party scoring data, is the scoring data of the Nth user in the updated second type of third-party scoring data.

[0056] In a possible implementation, the two types of third-party scores are sorted separately in the order from good to bad, and the sorted score sequences are respectively represented as and

[0057] Calculate the corresponding cumulative default rates for the first type of third-party scoring data and the second type of third-party scoring data with a preset order.

[0058] In a possible implementation, in the step of calculating the corresponding cumulative default rates for the first type of third-party scoring data and the second type of third-party scoring data with a preset order, it includes:

[0059] For the first type of third-party scoring data and the second type of third-party scoring data Calculate respectively to obtain the cumulative default rate sequences of the first type of third-party scoring data and the second type of third-party scoring data and Among them is the proportion of bad customers of the third-party score S1 below where is the proportion of bad customers of the third-party score S1 below The proportion of bad customers below is the proportion of bad customers of the third-party score S2 below The proportion of bad customers below is the proportion of bad customers of the third-party score S2 below The proportion of bad customers below.

[0060] Calculate the cumulative default rate sequences corresponding to the two types of third-party score sequences respectively, which are respectively and wherein represents the proportion of bad customers where the tripartite score S1 is below, Similarly, we can obtain the mapping relationship between the tripartite score sequence and the cumulative default rate sequence.

[0061] Step S130: Connect the first type of tripartite score data and the second type of tripartite score data according to the cumulative default rate to obtain the mapping from the second type of tripartite score data to the first type of tripartite score data.

[0062] In the step of connecting the first type of tripartite score data and the second type of tripartite score data according to the cumulative default rate to obtain the mapping from the second type of tripartite score data to the first type of tripartite score data, it includes:

[0063] Using the sequence of the cumulative default rate as the connection key, connect the first type of tripartite score data and the second type of tripartite score data, and further form the mapping from the tripartite score S2 sequence to the tripartite score S1 sequence. The technical solution provided by the present invention will use the cumulative default rate sequence as the connection key to connect the two tripartite score sequences, and further form the mapping from the tripartite score S2 sequence to the tripartite score S1 sequence.

[0064] Step S140: Use the second type of tripartite score data as the input and the first type of tripartite score data as the output to obtain the corresponding functional relationship.

[0065] The technical solution provided by the present invention, in step S140, includes:

[0066] Based on the mapping from the tripartite score S2 sequence to the tripartite score S1 sequence, use the tripartite score S2 sequence as the input and the tripartite score S1 sequence as the output, and perform fitting using the polynomial regression algorithm. Finally, obtain the functional relationship f(x) from the tripartite score S2 to the tripartite score S1, that is, S1 = f(S2).

[0067] The technical solution provided by the present invention also provides a credit score mapping device based on the cumulative bad debt rate, as Figure 3 shown, including:

[0068] An acquisition module, configured to respectively acquire the first type of tripartite score data and the second type of tripartite score data, where the first type of tripartite score data and the second type of tripartite score data respectively include the score data of multiple users;

[0069] A calculation module, configured to respectively calculate the cumulative default rate of the users for the first type of tripartite score data and the second type of tripartite score data;

[0070] A connection module, configured to connect first-party scoring data of a first type and second-party scoring data of a second type according to the cumulative default rate, so as to obtain a mapping from the second-party scoring data of the second type to the first-party scoring data of the first type;

[0071] A function obtaining module, configured to obtain a corresponding functional relationship by using the second-party scoring data of the second type as an input and the first-party scoring data of the first type as an output.

[0072] The present invention further provides a storage medium, in which a computer program is stored, and when the computer program is executed by a processor, it is used to implement the methods provided by the above various embodiments.

[0073] Wherein, the storage medium may be a computer storage medium or a communication medium. The communication medium includes any medium facilitating the transmission of a computer program from one place to another. The computer storage medium may be any available medium accessible by a general-purpose or special-purpose computer. For example, the storage medium is coupled to the processor, so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium may also be a component of the processor. The processor and the storage medium may be located in an application specific integrated circuit (ASIC). In addition, the ASIC may be located in a user device. Of course, the processor and the storage medium may also exist as discrete components in a communication device. The storage medium may be a read-only memory (ROM), a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0074] The present invention further provides a program product, which includes execution instructions stored in a storage medium. At least one processor of a device can read the execution instructions from the storage medium, and the execution of the execution instructions by at least one processor causes the device to implement the methods provided by the above various embodiments.

[0075] In the above embodiments of the terminal or the server, it should be understood that the processor may be a central processing unit (CPU for short), and may also be other general-purpose processors, a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the present invention may be directly embodied as being executed and completed by a hardware processor, or executed and completed by a combination of hardware and software modules in the processor.

[0076] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A credit score mapping method based on the cumulative bad debt rate, characterized in that, Including: Obtain the first type of third-party scoring data and the second type of third-party scoring data respectively. The first type of third-party scoring data and the second type of third-party scoring data each include the scoring data of multiple users. Among them, the first type of third-party scoring data is , and the second type of third-party scoring data is . is the scoring data of the th user in the first type of third-party scoring data, is the scoring data of the th user in the first type of third-party scoring data, is the scoring data of the th user in the second type of third-party scoring data, is the scoring data of the th user in the second type of third-party scoring data; Calculating the cumulative default rate of the user from the first type of third-party scoring data and the second type of third-party scoring data respectively; Connecting the first type of third-party scoring data and the second type of third-party scoring data according to the cumulative default rate to obtain a mapping from the second type of third-party scoring data to the first type of third-party scoring data, wherein, using the sequence of the cumulative default rate as the connection key, connecting the first type of third-party scoring data and the second type of third-party scoring data, and then forming a mapping from the third-party scoring S2 sequence to the third-party scoring S1 sequence; Taking the second type of third-party scoring data as the input and the first type of third-party scoring data as the output to obtain the corresponding functional relationship; Wherein, in the step of calculating the cumulative default rate of the user from the first type of third-party scoring data and the second type of third-party scoring data respectively, it further includes: Sorting the scoring data of the users in the first type of third-party scoring data and the second type of third-party scoring data respectively to obtain the first type of third-party scoring data and the second type of third-party scoring data with a preset order; For the first type of three-party scoring data with a preset order and the second type of three-party scoring data calculate the corresponding cumulative default rates respectively. Among them, in the step of calculating the corresponding cumulative default rates for the first type of three-party scoring data and the second type of three-party scoring data with a preset order, it includes: for the first type of three-party scoring data and the second type of three-party scoring data calculate respectively to obtain the cumulative default rate sequences of the first type of three-party scoring data and the second type of three-party scoring data and , where is the proportion of bad customers with a three-party score S1 below , where is the proportion of bad customers with a three-party score S1 below , is the proportion of bad customers with a three-party score S2 below , is the proportion of bad customers with a three-party score S2 below .

2. The credit scoring mapping method based on the cumulative bad debt rate according to claim 1, characterized in that In the step of sorting the rating data of users in the first type of third-party rating data and the second type of third-party rating data respectively to obtain the first type of third-party rating data and the second type of third-party rating data with a preset order, it includes: respectively obtaining the user label sequences of users in the first type of third-party rating data and the second type of third-party rating data, and the user label sequence is , where 0 is a customer of the first preset type, 1 is a customer of the second preset type, is the user label of the th user, is the user label of the th user, ; sorting the first type of third-party rating data and the second type of third-party rating data based on the user label sequence to obtain the sorted first type of third-party rating data , the second type of third-party rating data , where is the rating data of the th user in the updated first type of third-party rating data, is the rating data of the th user in the updated first type of third-party rating data, is the rating data of the th user in the updated second type of third-party rating data, is the rating data of the th user in the updated second type of third-party rating data.

3. The credit scoring mapping method based on the cumulative bad debt rate according to claim 2, characterized in that In the step of taking the second type of third-party scoring data as the input and the first type of third-party scoring data as the output to obtain the corresponding functional relationship, it includes: Based on the mapping from the three-party score S2 sequence to the three-party score S1 sequence, taking the three-party score S2 sequence as the input and the three-party score S1 sequence as the output, a polynomial regression algorithm is used for fitting, and finally the functional relationship from the three-party score S2 to the three-party score S1 is obtained , namely .

4. The credit scoring mapping method based on the cumulative bad debt rate according to claim 1, characterized in that is the number of the same users in the first type of three-party scoring data and the second type of three-party scoring data.

5. Storage medium, characterized in that, A computer program is stored in the storage medium, and when the computer program is executed by a processor, it is used to implement the method according to any one of claims 1 to 3.

Images