User financial risk identification method, device and electronic device based on incremental learning

Abstract

The application discloses a user financial risk identification method and device based on incremental learning and electronic equipment, and the method comprises the following steps: historical user data is divided into multiple time periods according to the starting time sequence of user financial behavior performance periods, historical user data of a time period with an end point t i‑1 is used to train a classification model M i , historical user data of a time period with an end point t i is input into the trained classification model M i for training, so as to obtain a classification model M i+1 , and the financial risk of a new user is identified according to each classification model M1, M2,..., M n+1 . The classification model M i of the application learns the sample features of the time period t i‑1 and the sample features of the time period t i , and finally, the sufficiency of the early sample can be retained, and the representativeness of the recent sample can be learned, so that the accuracy of the judgment on the recent behavior of the user is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer information processing technology, and in particular to a method, device, electronic device and computer-readable medium for identifying user financial risks based on incremental learning. Background Art

[0002] Currently, companies with comprehensive risk identification capabilities often hope to accurately identify users' risk levels in risky scenarios. Existing solutions typically use existing user data to train a risk prediction model, which is then used to predict users' potential risks.

[0003] When identifying user risk, many users only have short-term user data, such as short-term borrowers. The inventors discovered that using long-term user data, due to its long data cycle, cannot accurately predict recent user behavior. Using short-term user data, however, is insufficiently expressive due to its short duration, making it difficult to comprehensively predict user behavior. Therefore, a method is needed that reflects authentic data from existing after-sales users while avoiding the problem of data being too long, which makes it difficult to predict recent user risks. Summary of the Invention

[0004] The present invention aims to solve the technical problem that simply using long-term user data or simply using short-term user data in risk prediction will lead to inaccurate prediction of users' recent risks.

[0005] In order to solve the above technical problems, the first aspect of the present invention proposes a user financial risk identification method based on incremental learning, the method comprising:

[0006] The historical user data is divided into multiple time periods according to the starting time sequence of the user's financial behavior performance period, and the separation points of each time period are recorded as t0, t1, ..., t i ,…,t n , where i and n are positive integers;

[0007] Use the end point t i-1 The historical user data of the time period is used to train the classification model M i , set the end point to t i The historical user data of the time period is input into the trained classification model M i Train the classification model M i+1 ;

[0008] According to each classification model M1, M2, ..., M n+1 Identify financial risks of new users.

[0009] According to a preferred embodiment of the present invention, the endpoint is t iThe length of time contained in the time period decreases as i increases.

[0010] According to a preferred embodiment of the present invention, the classification model is a decision tree model.

[0011] According to a preferred embodiment of the present invention, the decision tree model M i+1 Based on the decision tree model M i Residual value learning.

[0012] According to a preferred embodiment of the present invention, the classification models M1, M2, ..., M n+1 Identifying financial risks for new users includes:

[0013] The user data of the new user is input into each decision tree model M1, M2, ..., M n+1 ;

[0014] Each decision tree model M1, M2, ..., M n+1 The output results are accumulated to obtain the final output result;

[0015] The financial risk of the new user is identified based on the final output result.

[0016] According to a preferred embodiment of the present invention, the historical user data includes at least one of: a user resource return rate, a user resource request number, and user attribute information.

[0017] In order to solve the above technical problems, the second aspect of the present invention provides a user financial risk identification device based on incremental learning, the device comprising:

[0018] The segmentation module is used to divide historical user data into multiple time periods according to the start time sequence of the user's financial behavior performance period. The separation points of each time period are recorded as t0, t1, ..., t i ,…,t n , where i and n are positive integers;

[0019] Training module for using the endpoint t i-1 The historical user data of the time period is used to train the classification model M i , set the end point to t i The historical user data of the time period is input into the trained classification model M i Train the classification model M i+1 ;

[0020] Identification module, used to classify the models M1, M2, ..., M n+1 Identify financial risks of new users.

[0021] According to a preferred embodiment of the present invention, the endpoint is t i The length of time contained in the time period decreases as i increases.

[0022] According to a preferred embodiment of the present invention, the classification model is a decision tree model.

[0023] According to a preferred embodiment of the present invention, the decision tree model M i+1 Based on the decision tree model M i Residual value learning.

[0024] According to a preferred embodiment of the present invention, the identification module includes:

[0025] Input module, used to input the user data of new users into each decision tree model M1, M2, ..., M n+1 ;

[0026] The accumulation module is used to accumulate the decision tree models M1, M2, ..., M n+1 The output results are accumulated to obtain the final output result;

[0027] The sub-identification module is used to identify the financial risk of the new user based on the final output result.

[0028] According to a preferred embodiment of the present invention, the historical user data includes at least one of: a user resource return rate, a user resource request number, and user attribute information.

[0029] To solve the above technical problems, the present invention provides an electronic device according to a third aspect, including:

[0030] processor; and

[0031] A memory storing computer executable instructions, which, when executed, cause the processor to perform the above method.

[0032] In order to solve the above technical problems, the fourth aspect of the present invention proposes a computer-readable storage medium, wherein the computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, the above method is implemented.

[0033] The present invention divides historical user data into multiple time periods according to the start time sequence of the user's financial behavior performance period, using the end point t i-1 The historical user data of the time period is used to train the classification model M i , set the end point to t i The historical user data of the time period is input into the trained classification model Mi for training, and the classification model M is obtained. i+1; Then according to each classification model M1, M2, ..., M n+1 Identify the financial risks of new users. In this way, the classification model M i Having learned t i-1 The sample characteristics of the time period are also learned i The sample characteristics of the time period, thereby realizing the classification model M i Through iterative learning of samples from different time periods, the final output can not only retain the adequacy of previous samples, but also learn the representativeness of recent samples, thereby improving the accuracy of judging users' recent behavior. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to make the technical problems solved by the present invention, the technical means adopted, and the technical effects achieved more clearly, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. However, it should be noted that the drawings described below are only drawings of exemplary embodiments of the present invention. Those skilled in the art can derive drawings of other embodiments based on these drawings without inventive effort.

[0035] Figure 1 This is a flow chart of a method for identifying user financial risks based on incremental learning according to the present invention;

[0036] Figure 2 This is a schematic diagram of the present invention dividing historical user data into multiple time periods according to the starting time sequence of the user's financial behavior performance period;

[0037] Figure 3 This is a schematic diagram of the structural framework of a user financial risk identification device based on incremental learning according to the present invention;

[0038] Figure 4 is a structural block diagram of an exemplary embodiment of an electronic device according to the present invention;

[0039] Figure 5 is a schematic diagram of an embodiment of a computer-readable medium of the present invention. DETAILED DESCRIPTION

[0040] The exemplary embodiments of the present invention will now be described more fully with reference to the accompanying drawings. Although each exemplary embodiment can be implemented in a variety of specific ways, it should not be understood that the present invention is limited to the embodiments set forth herein. Rather, these exemplary embodiments are provided to make the content of the present invention more complete and to more fully convey the inventive concept to those skilled in the art.

[0041] Under the premise of being consistent with the technical concept of the present invention, the structure, performance, effect or other characteristics described in a specific embodiment may be combined with one or more other embodiments in any appropriate manner.

[0042] In the introduction of specific embodiments, the detailed description of the structure, performance, effect or other features is intended to enable those skilled in the art to fully understand the embodiments. However, this does not preclude those skilled in the art from implementing the present invention with a technical solution that does not include the aforementioned structure, performance, effect or other features under specific circumstances.

[0043] The flowcharts in the accompanying drawings are merely illustrative of the process flow and do not necessarily include all of the content, operations, and steps in the flowcharts, nor do they necessarily imply that all of the steps in the flowcharts must be executed in the order shown. For example, some of the steps in the flowcharts may be separated, some may be combined or partially combined, and so on. The execution order shown in the flowcharts may be changed according to actual circumstances without departing from the spirit of the present invention.

[0044] Frames in the accompanying drawings Figure 1 The term "functional entity" generally refers to a functional entity and does not necessarily correspond to a physically independent entity. That is, these functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0045] The same reference numerals in the accompanying drawings represent the same or similar elements, components or parts, and thus repeated descriptions of the same or similar elements, components or parts may be omitted below. It should also be understood that although the first, second, third and other numbered adjectives may be used herein to describe various devices, elements, components or parts, these devices, elements, components or parts should not be limited by these adjectives. In other words, these adjectives are only used to distinguish one from another. For example, the first device may also be called the second device, but this does not deviate from the essential technical solution of the present invention. In addition, the terms "and / or" and "and / or" refer to all combinations including any one or more of the listed items.

[0046] In the present invention, incremental learning refers to a learning system that continuously learns new knowledge from new samples while retaining most of the previously learned knowledge. Incremental learning reduces storage space usage by eliminating the need to store historical data. Furthermore, it leverages historical training results in current sample training, significantly reducing subsequent training time. Currently, incremental learning is primarily used in two areas: first, for very large databases, such as web logs; and second, for streaming data, which constantly changes over time, such as stock trading data. While researching how risk control systems can predict near-term user risk, the inventors discovered that using long-term user data, due to its long data cycle, cannot accurately predict recent user behavior. Using short-term user data, however, is too short and lacks expressiveness, making it incapable of comprehensively predicting user behavior. Therefore, the inventors combined long-term and short-term user data to train a model based on incremental learning. This ensures that the model's final output retains the sufficiency of previous samples while also capturing the representativeness of recent samples, thereby improving the accuracy of predictions about recent user behavior.

[0047] See also Figure 1 , Figure 1 This is a flowchart of a method for identifying user financial risks based on incremental learning provided by the present invention. Figure 1 As shown, the method includes:

[0048] S1. Divide historical user data into multiple time periods according to the starting time sequence of the user's financial behavior performance period.

[0049] In the present invention, the separation points of each time period are respectively recorded as t0, t1, ..., t i ,…,t n , where i and n are positive integers; the user's financial behavior performance period refers to financial installment, for example, 3-month installment, 6-month installment, etc.

[0050] In principle, time periods can be divided based on the start and end time sequence of the user's financial behavior. The length of each time period can be set arbitrarily. For example, the time period ending at t0 can be 3 months long, the time period ending at t1 can be 3 months long, the time period ending at t2 can be 5 months long, and so on.

[0051] In order to improve the accuracy of the final prediction, the end point of the present invention is t i The length of time contained in the time period decreases with the increase of i. i The time period from the current time point decreases in length. For example Figure 2In the example, the current time point is T, the starting time point of the user's historical data is t0, and the time t0-T is divided into the following categories according to the starting time sequence of the user's financial behavior performance period: the time period ending at t0, the time period t0-t1, and the time period t1-t2. Among them, the time length from the time period with end point t0 to the current time point T is greater than the time length from the time period with end point t1 to the current time point T, and the time length from the time period with end point t1 to the current time point T is greater than the time length from the time period with end point t2 to the current time point T, that is, t0, t1, and t2 are getting closer and closer to the current time point, and the time lengths of the time period with end point t0, the time period t0-t1, and the time period t1-t2 decrease successively. For example, the time period with end point t0 is a time period 6-12 months before the current time T, the time period t0-t1 is a time period 3-6 months before the current time T, and the time period t1-t2 is a time period 2-3 months before the current time T. Therefore, the time period with end point t0 and the time period t0-t1 have sufficient performance samples, and the time period t1-t2 is closest to the current time point T.

[0052] In the present invention, the historical user data is determined based on actual business operations. For example, in the case of financial services, the user data includes at least one of the following: user resource return rate, number of user resource requests, and user attribute information. User attribute information may include user identification information (such as user ID) and contact information (such as mobile phone number and address). It may also include information reflecting user income, such as occupation and educational background.

[0053] S2, use end point t i-1 The historical user data of the time period is used to train the classification model M i , set the end point to t i The historical user data of the time period is input into the trained classification model M i Train the classification model M i+1 ;

[0054] The classification model can be a decision tree model or a neural network model, with the present invention preferably using a decision tree model. A decision tree is a prediction model that represents a mapping relationship between object attributes and object values. Each node in the tree represents an object, each forked path represents a possible attribute value, and each leaf node corresponds to the value of the object represented by the path from the root node to the leaf node.

[0055] Specifically, the decision tree model M1 is trained by the historical user data of the time period ending at t0; the historical user data of the time period t0-t1 is input into the trained decision tree M1 for training to obtain the decision tree model M2, and so on. i-1The historical user data of the time period is used to train the decision tree model M i , set the end point to t i The historical user data of the time period is input into the trained decision tree model M i Train the decision tree model M i+1 . In this way, the decision tree model M i Since we have learned that the end point is t i-1 The sample features of the time period are also learned with the end point t i The sample characteristics of the time period.

[0056] In the present invention, the decision tree model M i Training can be performed using any of the ID3, C4.5, C5.0, or CART algorithms.

[0057] In the training process of the present invention, for the same sample, the decision tree model M i+1 Based on the decision tree model M i The residual value. That is, for the same historical user, first use the data from the time period ending at t0 to train decision tree model M1. At the leaf nodes, the predicted value for the historical user and the residual after the prediction can be obtained. Then, the historical user data from the time period t0-t1 is input into the trained decision tree M1 for training, resulting in decision tree model M2. Decision tree model M2 is trained based on the residual of decision tree model M1, and so on, until the residual between the predicted value and the true value is zero. For example, in a sample training, if the total score predicted by decision tree model M1 is 0.9, decision tree model M2 is learned based on the residual of M1, 0.1.

[0058] S3, according to each classification model M1, M2, ..., M n+1 Identify financial risks of new users.

[0059] The classification models M1, M2, ..., M with iterative features trained through the above step S2 are n+1 This step iterates the results of each classification model to identify the financial risk of new users. For example, taking the decision tree model as an example, this step includes:

[0060] S31, input the user data of the new user into each decision tree model M1, M2, ..., M n+1 ;

[0061] S32, each decision tree model M1, M2, ..., M n+1 The output results are accumulated to obtain the final output result;

[0062] Since during the training process, the decision tree model M i+1 The total score of the decision tree model Mi The residual value of . Then the predicted value for the new user is the residual value of each decision tree model M1, M2, ..., M n+1 The whole process is to learn a little bit each time (a part of the true value) and then accumulate it. i By iteratively learning samples from different time periods, the final output result can not only retain the adequacy of previous samples, but also learn the representativeness of recent samples, thereby improving the accuracy of judging users' recent behavior.

[0063] S33. Identify the financial risk of the new user according to the final output result.

[0064] In this step, in order to facilitate the identification of user risks, the final output of the model can be converted into a value R between 0 and 1 through the Sigmoid function, and the financial risk of new users can be identified based on the size of the value R.

[0065] Figure 3 This is a schematic diagram of the architecture of a user financial risk identification device based on incremental learning in the present invention. Figure 3 As shown, the device includes:

[0066] The segmentation module 31 is used to divide the historical user data into multiple time periods according to the start time sequence of the user's financial behavior performance period, and the separation points of each time period are recorded as t0, t1, ..., t i ,…,t n , where i and n are positive integers; the historical user data includes: at least one of: user resource return rate, number of user resource requests, and user attribute information.

[0067] Training module 32, for using the end point t i-1 The historical user data of the time period is used to train the classification model M i , set the end point to t i The historical user data of the time period is input into the trained classification model M i Train the classification model M i+1 ;

[0068] Identification module 33, for identifying n+1 Identify financial risks of new users.

[0069] In a specific embodiment, the endpoint is t i The length of time contained in the time period decreases as i increases.

[0070] Preferably, the classification model is a decision tree model. i+1 The total score of the decision tree model Mi The residual value of .

[0071] In a specific embodiment, the identification module 33 includes:

[0072] Input module 331 is used to input the user data of the new user into each decision tree model M1, M2, ..., M n+1 ;

[0073] The accumulation module 332 is used to accumulate the decision tree models M1, M2, ..., M n+1 The output results are accumulated to obtain the final output result;

[0074] The sub-identification module 333 is configured to identify the financial risk of the new user based on the final output result.

[0075] Those skilled in the art will appreciate that the modules in the above device embodiments may be distributed in the device as described, or may be modified accordingly and distributed in one or more devices different from the above embodiments. The modules in the above embodiments may be combined into one module or further split into multiple submodules.

[0076] The following describes an electronic device embodiment of the present invention. This electronic device can be considered a physical implementation of the method and apparatus embodiments of the present invention described above. Details described in the electronic device embodiment of the present invention should be considered supplementary to the above-described method or apparatus embodiments; details not disclosed in the electronic device embodiment of the present invention can be implemented with reference to the above-described method or apparatus embodiments.

[0077] Figure 4 is a structural block diagram of an exemplary embodiment of an electronic device according to the present invention. Figure 4 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0078] like Figure 4 As shown, the electronic device 400 of this exemplary embodiment is implemented as a general-purpose data processing device. Components of the electronic device 400 may include, but are not limited to, at least one processing unit 410, at least one storage unit 420, a bus 430 connecting various electronic device components (including the storage unit 420 and the processing unit 410), a display unit 440, and the like.

[0079] The storage unit 420 stores a computer-readable program, which may be a source program or a code of a read-only program. The program may be executed by the processing unit 410, so that the processing unit 410 performs the steps of various embodiments of the present invention. For example, the processing unit 410 may perform the following steps: Figure 1 Steps shown.

[0080] The storage unit 420 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 4201 and / or a cache memory unit 4202, and may further include a read-only memory unit (ROM) 4203. The storage unit 420 may also include a program / utility 4204 having a set (at least one) of program modules 4205. Such program modules 4205 include, but are not limited to, operating the electronic device, one or more application programs, other program modules, and program data. Each of these examples or some combination thereof may include the implementation of a network environment.

[0081] Bus 430 may represent one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0082] The electronic device 400 may also communicate with one or more external devices 300 (e.g., a keyboard, a display, a network device, a Bluetooth device, etc.), so that a user can interact with the electronic device 400 via these external devices 400, and / or so that the electronic device 400 can communicate with one or more other data processing devices (e.g., a router, a modem, etc.). Such communication may be performed through an input / output (I / O) interface 450, or may be performed with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) through a network adapter 460. The network adapter 460 may communicate with other modules of the electronic device 400 via the bus 430. It should be understood that although Figure 4 Not shown, other hardware and / or software modules may be used in the electronic device 400, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID electronic devices, tape drives, and data backup storage electronic devices.

[0083] Figure 5 Schematic diagram of a computer readable medium embodiment of the present invention. Figure 5As shown, the computer program can be stored on one or more computer-readable media. The computer-readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can be, for example, but not limited to, an electronic device, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. When the computer program is executed by one or more data processing devices, the computer-readable medium is enabled to implement the above-mentioned method of the present invention, namely: dividing the historical user data into a plurality of time periods according to the starting time sequence of the user's financial behavior performance period, and the separation points of each time period are respectively recorded as t0, t1, ..., t i ,…,t n , where i and n are positive integers; the end point is t i-1 The historical user data of the time period is used to train the classification model M i , set the end point to t i The historical user data of the time period is input into the trained classification model M i Train the classification model M i+1 ; According to each classification model M1, M2, ..., M n+1 Identify financial risks of new users.

[0084] From the above description of the embodiments, those skilled in the art will readily appreciate that the exemplary embodiments described herein can be implemented via software or via a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present invention can be embodied in the form of a software product, which can be stored on a computer-readable storage medium (such as a CD-ROM, USB flash drive, or portable hard drive) or on a network and includes instructions for causing a data processing device (such as a personal computer, server, or network device) to execute the above-described methods according to the present invention.

[0085] The computer-readable storage medium may include a data signal propagated in baseband or as part of a carrier wave, wherein the readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction-executing electronic device, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination thereof.

[0086] The program code for performing the operations of the present invention may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, and the like, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computing device, partially on the user's device, as a stand-alone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0087] In summary, the present invention can be implemented by a method, apparatus, electronic device, or computer-readable medium that executes a computer program. In practice, a general-purpose data processing device such as a microprocessor or digital signal processor (DSP) can be used to implement some or all of the functions of the present invention.

[0088] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the present invention is not inherently related to any specific computer, virtual device, or electronic device, and various general-purpose devices can also implement the present invention. The above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.

Claims

1. A user financial risk identification method based on incremental learning, characterized in that: The method comprises: Determine multiple time periods according to the start time sequence of the user financial installment of the historical user data, and divide the historical user data into multiple time periods corresponding to the time periods; the separation points of each time period are respectively recorded as t0, t1, ..., t i ,…,t n , where i and n are positive integers; the end point is t i The time interval between the time period and the current time point decreases as i increases; Combine the long-term historical user data and short-term historical user data in the historical user data to train the decision tree model M i ; Wherein: the installment duration of the financial installment of the long-term historical user data is longer than the installment duration of the financial installment of the short-term historical user data; the end point of use is t i-1 The historical user data of the time period is used to train the classification model M i , set the end point to t i The historical user data of the time period is input into the trained classification model M i Train the classification model M i+1 ; Decision tree model M i Since we have learned that the end point is t i-1 The sample features of the time period are also learned with the end point t i Sample features of the time period; implement the classification model M i Iterative learning of samples from different time periods; the final output can retain the sufficiency of previous samples while learning the representativeness of recent samples; According to each classification model M1, M2, ..., M n+1 Identify financial risks of new users.

2. The method according to claim 1, characterized in that The end point is t i The length of time contained in the time period decreases as i increases.

3. The method according to any one of claims 1 to 2, characterized in that The classification model is a decision tree model.

4. The method according to claim 3, characterized in that Decision tree model M i+1 Based on the decision tree model M i Residual value learning.

5. The method according to claim 4, characterized in that According to the classification models M1, M2, ..., M n+1 Identifying financial risks for new users includes: The user data of the new user is input into each decision tree model M1, M2, ..., M n+1 ; Each decision tree model M1, M2, ..., M n+1 The output results are accumulated to obtain the final output result; The financial risk of the new user is identified based on the final output result.

6. The method according to claim 1, characterized in that The historical user data includes at least one of: a user resource return rate, a user resource request number, and user attribute information.

7. A user financial risk identification device based on incremental learning, characterized in that: The device comprises: The segmentation module is used to determine multiple time periods according to the start time sequence of the user financial installment of the historical user data, and divide the historical user data into multiple time periods corresponding to the time period. The separation points of each time period are respectively recorded as t0, t1, ..., t i ,…,t n , where i and n are positive integers; the end point is t i The time interval between the time period and the current time point decreases as i increases; The training module is used to combine the long-term historical user data and the short-term historical user data in the historical user data to train the decision tree model M i ; Wherein: the installment duration of the financial installment of the long-term historical user data is longer than the installment duration of the financial installment of the short-term historical user data; the end point of use is t i-1 The historical user data of the time period is used to train the classification model M i , set the end point to t i The historical user data of the time period is input into the trained classification model M i Train the classification model M i+1 , decision tree model M i Since we have learned that the end point is t i-1 The sample features of the time period are also learned with the end point t i Sample features of the time period; implement the classification model M i Iterative learning of samples from different time periods; the final output can retain the sufficiency of previous samples while learning the representativeness of recent samples; Identification module, used to classify the models M1, M2, ..., M n+1 Identify financial risks of new users.

8. An electronic device comprising: processor; as well as A memory storing computer executable instructions which, when executed, cause the processor to perform the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, wherein: The computer-readable storage medium stores one or more programs, and when the one or more programs are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images