Method, device, processor and electronic device for determining credit rating of object

Abstract

The application discloses a kind of determination method, device, processor and electronic equipment of credit rating of object, it is related to artificial intelligence field.The method comprises the following steps: obtaining the attribute information set of object, wherein the attribute information set includes at least one attribute information, and the attribute information is used to indicate the attribute associated with the credit rating of object;From the attribute information of attribute information set, determine target attribute information;Target attribute information is input into target prediction model in target prediction model set to carry out prediction, and the initial credit rating of object is obtained, wherein the target prediction model is obtained by training initial prediction model based on sample attribute information set of sample object, and sample attribute information set includes at least one sample attribute information, and sample attribute information is used to indicate the attribute associated with the credit rating of sample object;Determine the target credit rating associated with initial credit rating.The application solves the technical problem that the determination precision of credit rating is low.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, in particular to a method and device for determining credit rating of an object, a processor and an electronic device. BACKGROUND

[0002] In the process of applying for a credit card, the manual judgment method is used to determine whether to agree to the user's credit card application and the allocated credit card limit, which has the defects of strong subjectivity and lack of theoretical basis.

[0003] In order to avoid the above defects, a constructed model is used to predict the credit rating of the user, and according to the obtained prediction rating, it is directly determined whether to agree to the user's credit card application and the allocated credit card limit. However, the prediction result of this prediction method is not comprehensive, thereby causing the technical problem of low accuracy of determining the credit rating of the user.

[0004] At present, there is no effective solution to the technical problem of low accuracy of determining the credit rating. SUMMARY

[0005] The embodiments of the present application provide a method and device for determining credit rating of an object, a processor and an electronic device, to at least solve the technical problem of low accuracy of determining the credit rating.

[0006] According to an aspect of an embodiment of the present application, a method for determining credit rating of an object is provided, the method comprising: obtaining an attribute information set of the object, wherein the attribute information set comprises at least one attribute information, and the attribute information is used to represent an attribute associated with the credit rating of the object; determining target attribute information from the attribute information of the attribute information set, wherein the degree of association between the target attribute information and the credit rating is greater than the degree of association between the remaining attribute information in the attribute information set except the target attribute information and the credit rating; inputting the target attribute information into a target prediction model in a target prediction model set for prediction to obtain an initial credit rating of the object, wherein the target prediction model is obtained by training an initial prediction model based on a sample attribute information set of a sample object, and the sample attribute information set comprises at least one sample attribute information, and the sample attribute information is used to represent an attribute associated with the credit rating of the sample object; determining a target credit rating associated with the initial credit rating, wherein the initial credit rating is fused in the target credit rating.

[0007] Optionally, the target prediction model set includes a first target prediction model, a second target prediction model, and a third target prediction model, the first target prediction model is a prediction model constructed for a number of overdue periods of a credit card held by the subject, the second target prediction model is a prediction model constructed for an overdue time of the credit card, and the third target prediction model is a prediction model constructed for an overdue amount of the credit card, the target attribute information is input into the target prediction model in the target prediction model set to perform prediction to obtain the initial credit level of the subject, including: inputting the target attribute information into the first target prediction model to perform prediction to obtain a first initial credit level; inputting the target attribute information into the second target prediction model to perform prediction to obtain a second initial credit level; and inputting the target attribute information into the third target prediction model to perform prediction to obtain a third initial credit level.

[0008] Optionally, the target credit level associated with the initial credit level is determined, including: determining a weight of the initial credit level; and determining the target credit level based on the weight and the initial credit level.

[0009] Optionally, the weight includes a first weight of the first initial credit level, a second weight of the second initial credit level, and a third weight of the third initial credit level, and the target credit level is determined based on the weight and the initial credit level, including: determining the target credit level based on the first weight, the second weight, the third weight, and the initial credit level.

[0010] Optionally, the target credit level is determined based on the first weight, the second weight, the third weight, and the initial credit level, including: determining a first product between the first weight and the first initial credit level, a second product between the second weight and the second initial credit level, and a third product between the third weight and the third initial credit level; and determining a sum of the first product, the second product, and the third product as the target credit level.

[0011] Optionally, the target attribute information is determined from the attribute information in the attribute information set, including: determining a degree of association between the attribute information in the attribute information set and the credit level; determining a target degree of association from the degrees of association, wherein the target degree of association is greater than remaining degrees of association other than the target degree of association in the degrees of association; and determining attribute information corresponding to the target degree of association as the target attribute information.

[0012] According to an aspect of an embodiment of the present application, a device for determining a credit level of an object is provided. The device can include: an obtaining unit configured to obtain a set of attribute information of the object, wherein the set of attribute information includes at least one attribute information, and the attribute information is used to represent an attribute associated with the credit level of the object; a first determining unit configured to determine target attribute information from the attribute information of the set of attribute information, wherein a degree of association between the target attribute information and the credit level is greater than a degree of association between remaining attribute information of the set of attribute information other than the target attribute information and the credit level; a predicting unit configured to input the target attribute information into a target prediction model of a set of target prediction models to obtain an initial credit level of the object, wherein the target prediction model is obtained by training an initial prediction model based on a set of sample attribute information of a sample object, and the set of sample attribute information includes at least one sample attribute information, and the sample attribute information is used to represent an attribute associated with a credit level of the sample object; and a second determining unit configured to determine a target credit level associated with the initial credit level, wherein the initial credit level is fused into the target credit level.

[0013] According to another aspect of an embodiment of the present application, a processor is provided. The processor is configured to run a program, wherein the program, when run by the processor, performs the method for determining a credit level of an object in the embodiments of the present application.

[0014] According to another aspect of an embodiment of the present application, an electronic device is provided. The electronic device includes a memory storing an executable program, and a processor configured to run the program, wherein the program, when run, performs the method for determining a credit level of an object in the embodiments of the present application.

[0015] According to another aspect of an embodiment of the present application, a computer readable storage medium is provided. The computer readable storage medium includes a stored program, wherein the program, when run, controls a device in which the computer readable storage medium is located to perform the method for determining a credit level of an object in the embodiments of the present application.

[0016] According to another aspect of an embodiment of the present application, a computer program product is provided. The computer program product includes a computer program, wherein the computer program, when executed by a processor, implements the method for determining a credit level of an object in the embodiments of the present application.

[0017] According to another aspect of an embodiment of the present application, a computer program product is provided. The computer program product includes a non-volatile computer readable storage medium configured to store a computer program, wherein the computer program, when executed by a processor, implements the method for determining a credit level of an object in the embodiments of the present application.

[0018] According to another aspect of the embodiments of the present application, the embodiments of the present application further provide a computer program, which, when executed by a processor, implements the method for determining the credit level of the object in the above embodiments of the present application.

[0019] In the embodiments of the present application, when determining the credit level of the object, the attribute information set of the object can be acquired. From the attribute information in the acquired attribute information set, the target attribute information can be determined, the determined target attribute information is input into the target prediction model in the target prediction model set for prediction, the initial credit level of the object can be obtained, and the target credit level associated with the initial credit level is determined, thereby achieving the purpose of ensuring the comprehensiveness of the prediction result, solving the technical problem of low determination accuracy of the credit level of the user, and further achieving the technical effect of improving the determination accuracy of the credit level of the user. BRIEF DESCRIPTION OF DRAWINGS

[0020] The accompanying drawings, which are included to provide a further understanding of the present application and constitute a part of this application, illustrate embodiments of the present application and together with the description serve to explain the present application. In the drawings:

[0021] Figure 1 FIG. 1 is a hardware structure block diagram of a computer terminal for implementing the method for determining the credit level of the object according to the embodiments of the present application;

[0022] Figure 2 FIG. 2 is a flowchart of the method for determining the credit level of the object according to the embodiments of the present application;

[0023] Figure 3 FIG. 3 is a flowchart of the method for analyzing the credit risk of the user according to the embodiments of the present application;

[0024] Figure 4 FIG. 4 is a schematic diagram of the device for determining the credit level of the object according to the embodiments of the present application;

[0025] Figure 5 FIG. 5 is a structure block diagram of an electronic device according to the embodiments of the present application. DETAILED DESCRIPTION

[0026] In order to enable persons skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by persons skilled in the art without creative labor should fall within the scope of protection of the present application.

[0027] It should be noted that the terms "first", "second", and the like in the description and in the claims of the present application and the above-described accompanying 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 data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" 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 can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0028] According to an embodiment of the present application, a method for determining a credit rating of an object is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0029] The method for determining a credit rating of an object provided by the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 is a hardware structure block diagram of a computer terminal for implementing a method for determining a credit rating of an object according to an embodiment of the present application. As shown in Figure 1 , the computer terminal 10 (or mobile device) can include one or more processors 102 (the processor 102 can include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, which does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 can also include more or fewer components than those shown in Figure 1 , or have a different configuration than that shown in Figure 1 .

[0030] It should be noted that the one or more processors 102 and / or other data processing circuitry described above can be generally referred to herein as "data processing circuitry". The data processing circuitry can be embodied in whole or in part as software, hardware, firmware, or any combination thereof. In addition, the data processing circuitry can be a single standalone processing module or incorporated in whole or in part within any of the other elements of the computer terminal 10 (or mobile device). As referred to in the embodiments of the present application, the data processing circuitry functions as a processor to control, for example, the selection of the variable resistance terminal path connected to the interface.

[0031] The memory 104 can be used to store software programs of application software and modules, such as program instructions / data storage means corresponding to the object credit rating determination method in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, i.e. implements the object credit rating determination method described above. The memory 104 can include a high-speed random access memory and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 can further include a memory remotely arranged with respect to the processor 102, which can be connected to the computer terminal 10 through a network. Examples of the network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0032] The transmission device 106 can be used to receive or send data via a network. Specific examples of the network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network adapter (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module for communicating with the Internet in a wireless manner.

[0033] The display can be, for example, a touch screen type liquid crystal display (LCD) that enables a user to interact with the user interface of the computer terminal 10 (or mobile device).

[0034] In the above operating environment, the present application provides an object credit rating determination method as shown in Figure 2 Figure 2 is a flowchart of an object credit rating determination method according to an embodiment of the present application. The method can include the following steps: ​

[0035] In step S201, a set of attribute information of the object is obtained.

[0036] In the technical solution provided in step S201, the set of attribute information can include at least one attribute information. The attribute information can be used to represent an attribute associated with the credit level of the object. For example, the object can be a current user who needs to apply for a credit card, and the attribute associated with the credit level of the object can include basic information, asset status, historical credit status, and family situation of the current user. The basic information can include age, gender, occupation, working years, and income, which are only examples and are not limited.

[0037] In this embodiment, the set of attribute information of the object is obtained. Optionally, the embodiment filters the background data according to the demand of the object to obtain the set of attribute information of the object, that is, if the demand of the object is to apply for a credit card, the background data is filtered to obtain the set of attribute information of the object. Alternatively, if the demand of the object is to apply for a credit card, the set of attribute information of the object is directly downloaded from the server.

[0038] In step S202, target attribute information is determined from the attribute information of the set of attribute information.

[0039] In the technical solution provided in step S202, the degree of association between the target attribute information and the credit level can be greater than the degree of association between the remaining attribute information in the set of attribute information and the credit level. For example, the degree of association can also be referred to as the degree of influence, which is only an example and is not limited.

[0040] In this embodiment, after obtaining the set of attribute information of the object, the target attribute information is determined from the attribute information of the set of attribute information. Optionally, the embodiment can determine the degree of association between the attribute information of the set of attribute information and the credit level on the basis of obtaining the set of attribute information of the object. The attribute information corresponding to the maximum degree of association between the attribute information of the set of attribute information and the credit level is determined as the target attribute information.

[0041] Optionally, the degree of association between the attribute information of the set of attribute information and the credit level can be determined on the basis of obtaining the set of attribute information of the object. For example, the degree of correlation between the attribute information of the set of attribute information and the credit level can be calculated by using a statistical method, and the calculated degree of correlation is used as the degree of association between the attribute information of the set of attribute information and the credit level. The statistical method can be, but is not limited to, a correlation coefficient, a chi-square test, and the like, which are only examples and are not limited.

[0042] Step S203, inputting the target attribute information into a target prediction model in the target prediction model set to perform prediction, and obtaining an initial credit level of the object.

[0043] In the technical solution provided by step S203 of the present application, the target prediction model can be obtained by training an initial prediction model based on a sample attribute information set of a sample object. The sample attribute information set can include at least one sample attribute information, which can be used to represent an attribute associated with a credit level of the sample object.

[0044] In this embodiment, the initial prediction model can be constructed based on a multi-layer perceptron (MLP), the sample object can be a historical user who has a demand for applying for a credit card, and the attribute associated with the credit level of the sample object can include basic information, asset status, historical credit status, family situation, etc. of the historical user. The basic information can include age, gender, occupation, working years, income, etc. This is only an example and is not limited.

[0045] In this embodiment, the initial credit level can be represented in at least one of the following forms or a combination thereof: English letter form and Arabic numeral form, etc. For example, if the initial credit level is represented in English letter form, the initial credit level of A is higher than that of B, and the initial credit level of B is higher than that of C; if the initial credit level is represented in Arabic numeral form, the initial credit level of 10 is higher than that of 9, and the initial credit level of 9 is higher than that of 8. This is only an example and is not limited.

[0046] In this embodiment, after determining the target attribute information from the attribute information set, the target attribute information is input into a target prediction model in the target prediction model set to perform prediction, and an initial credit level of the object is obtained. Alternatively, based on the determination of the target attribute information, the determined target attribute information is input into a target prediction model in the target prediction model set to perform prediction, and an output layer of the target prediction model is used to output the predicted initial credit level.

[0047] Step S204, determining a target credit level associated with the initial credit level.

[0048] In the technical solution provided by step S204 of the present application, the target credit level is fused with the initial credit level.

[0049] In this embodiment, after the target attribute information is input into the target prediction model in the target prediction model set to perform prediction, an initial credit level of the object is obtained, and a target credit level associated with the initial credit level is determined. Optionally, the embodiment performs data fusion on each initial credit level obtained by using the target prediction model in the target prediction model set on the basis of the initial credit level of the object, and the target credit level can be obtained.

[0050] The above steps S201 to S204 of the present application can obtain the attribute information set of the object when determining the credit level of the object. The target attribute information can be determined from the attribute information in the obtained attribute information set, and the determined target attribute information is input into the target prediction model in the target prediction model set to perform prediction, so that the initial credit level of the object is obtained, and the target credit level associated with the initial credit level is determined. Thus, the comprehensiveness of the prediction result is ensured, thereby solving the technical problem of low determination accuracy of the credit level of the user, and further achieving the technical effect of improving the determination accuracy of the credit level of the user.

[0051] The above method of this embodiment is further introduced as follows.

[0052] As an optional embodiment, step S203 of inputting the target attribute information into the target prediction model in the target prediction model set to perform prediction and obtaining the initial credit level of the object includes: inputting the target attribute information into the first target prediction model to perform prediction and obtaining a first initial credit level; inputting the target attribute information into the second target prediction model to perform prediction and obtaining a second initial credit level; and inputting the target attribute information into the third target prediction model to perform prediction and obtaining a third initial credit level.

[0053] In this embodiment, the target prediction model set can include a first target prediction model, a second target prediction model, and a third target prediction model. The first target prediction model can be a prediction model constructed for the number of overdue credit cards held by the object, the second target prediction model can be a prediction model constructed for the overdue time of the credit card, and the third target prediction model can be a prediction model constructed for the overdue amount of the credit card.

[0054] In this embodiment, after determining the target attribute information from the attribute information set, the target attribute information is input into the first target prediction model for prediction to obtain the first initial credit level. Optionally, based on the determination of the target attribute information, when the target prediction model is the first target prediction model, the determined target attribute information is input into the first target prediction model for prediction, and the output layer of the first target prediction model is used to output the predicted first initial credit level.

[0055] In this embodiment, after determining the target attribute information from the attribute information set, the target attribute information is input into the second target prediction model for prediction to obtain the second initial credit level. Optionally, based on the determination of the target attribute information, when the target prediction model is the second target prediction model, the determined target attribute information is input into the second target prediction model for prediction, and the output layer of the second target prediction model is used to output the predicted second initial credit level.

[0056] In this embodiment, after determining the target attribute information from the attribute information set, the target attribute information is input into the third target prediction model for prediction to obtain the third initial credit level. Optionally, based on the determination of the target attribute information, when the target prediction model is the third target prediction model, the determined target attribute information is input into the third target prediction model for prediction, and the output layer of the third target prediction model is used to output the predicted third initial credit level.

[0057] As an optional embodiment, step S204 of determining the target credit level associated with the initial credit level comprises: determining the weight of the initial credit level; and determining the target credit level based on the weight and the initial credit level.

[0058] In this embodiment, after inputting the target attribute information into the target prediction model in the target prediction model set for prediction to obtain the initial credit level of the object, the weight of the initial credit level is determined. Optionally, based on the obtained initial credit level of the object, the first weight of the first initial credit level, the second weight of the second initial credit level, and the third weight of the third initial credit level can be determined from the weight distribution table. The weight distribution table can include a mapping relationship between different initial information levels and different weights.

[0059] In this embodiment, after the weight of the initial credit level is determined, the target credit level is determined based on the weight and the initial credit level. Alternatively, this embodiment sets the determined weight to the initial credit level based on the determination of the weight of the initial credit level, and the target credit level can be obtained.

[0060] As an optional embodiment, the target credit level is determined based on the weight and the initial credit level, including: determining the target credit level based on the first weight, the second weight, the third weight and the initial credit level.

[0061] In this embodiment, the weight can include a first weight of a first initial credit level, a second weight of a second initial credit level, and a third weight of a third initial credit level.

[0062] In this embodiment, after the weight of the initial credit level is determined, the target credit level is determined based on the first weight, the second weight, the third weight and the initial credit level. Alternatively, this embodiment sets the determined first weight to the first initial credit level, sets the determined second weight to the second initial credit level, and sets the determined third weight to the third initial credit level based on the determination of the weight of the initial credit level, and the target credit level can be obtained.

[0063] As an optional embodiment, the target credit level is determined based on the first weight, the second weight, the third weight and the initial credit level, including: determining a first product between the first weight and the first initial credit level, a second product between the second weight and the second initial credit level, and a third product between the third weight and the third initial credit level; and determining a sum of the first product, the second product and the third product as the target credit level.

[0064] In this embodiment, after the weight of the initial credit level is determined, a first product between the first weight and the first initial credit level, a second product between the second weight and the second initial credit level, and a third product between the third weight and the third initial credit level are determined. Alternatively, this embodiment performs a multiplication operation between the first weight and the first initial credit level based on the determination of the weight of the initial credit level, and the first product between the first weight and the first initial credit level can be obtained. A multiplication operation is performed between the second weight and the second initial credit level, and the second product between the second weight and the second initial credit level can be obtained. A multiplication operation is performed between the third weight and the third initial credit level, and the third product between the third weight and the third initial credit level can be obtained.

[0065] In this embodiment, after determining the first product between the first weight and the first initial credit level, the second product between the second weight and the second initial credit level, and the third product between the third weight and the third initial credit level, the sum between the first product, the second product and the third product is determined as the target credit level. Alternatively, this embodiment can obtain the sum between the first product, the second product and the third product based on the determination of the first product, the second product and the third product, and determine the obtained sum as the target credit level.

[0066] As an optional embodiment, the step S202 of determining the target attribute information from the attribute information in the attribute information set comprises: determining the association degree between the attribute information in the attribute information set and the credit level; determining the target association degree from the association degrees; and determining the attribute information corresponding to the target association degree as the target attribute information.

[0067] In this embodiment, the target association degree can be greater than the remaining association degrees in the association degrees except the target association degree.

[0068] In this embodiment, after obtaining the attribute information set of the object, the association degree between the attribute information in the attribute information set and the credit level is determined. Alternatively, this embodiment can calculate the correlation degree between the attribute information in the attribute information set and the credit level based on the obtained attribute information set by using a statistical method, and determine the calculated correlation degree as the association degree between the attribute information in the attribute information set and the credit level. The statistical method can be, but is not limited to, a correlation coefficient, a chi-square test, etc., which are only used as examples and are not limited specifically.

[0069] In this embodiment, after determining the association degree between the attribute information in the attribute information set and the credit level, the target association degree is determined from the association degrees. Alternatively, this embodiment can compare the association degrees between the attribute information in the attribute information set and the credit level in sequence until the target association degree is found, i.e., the maximum association degree is found, based on the determined association degrees.

[0070] In this embodiment, after determining the target association degree from the association degrees, the attribute information corresponding to the target association degree is determined as the target attribute information. Alternatively, this embodiment can directly determine the attribute information corresponding to the target association degree as the target attribute information based on the determined target association degree, i.e., determine the attribute information corresponding to the maximum association degree as the target attribute information.

[0071] In the embodiment of the present application, when determining the credit level of the object, the attribute information set of the object can be obtained. From the attribute information of the obtained attribute information set, target attribute information can be determined, the determined target attribute information is input into a target prediction model in the target prediction model set for prediction, the initial credit level of the object can be obtained, and the target credit level associated with the initial credit level is determined, thereby achieving the purpose of ensuring the comprehensiveness of the prediction result, thereby solving the technical problem of low determination accuracy of the credit level of the user, and further achieving the technical effect of improving the determination accuracy of the credit level of the user.

[0072] The technical solutions of the embodiments of the present application will be illustrated below in combination with preferred embodiments.

[0073] In the process of applying for a credit card, the manual judgment method is used to determine whether to agree to the credit card application of the user and the credit card limit allocated, which has the defects of strong subjectivity and lack of theoretical basis.

[0074] In order to avoid the above defects, the model constructed is used to predict the credit level of the user, and whether to agree to the credit card application of the user and the credit card limit allocated is directly determined according to the obtained prediction level. However, the prediction result of this prediction method is not comprehensive, thereby causing the technical problem of low determination accuracy of the credit level of the user.

[0075] In order to solve the above technical problem, the embodiment of the present application proposes a method for determining the credit level of an object. From the attribute information of the obtained attribute information set, target attribute information can be determined, the determined target attribute information is input into a target prediction model in the target prediction model set for prediction, the initial credit level of the object can be obtained, and the target credit level associated with the initial credit level is determined, thereby achieving the purpose of ensuring the comprehensiveness of the prediction result, thereby solving the technical problem of low determination accuracy of the credit level of the user, and further achieving the technical effect of improving the determination accuracy of the credit level of the user.

[0076] In this embodiment, the user credit risk analysis method in the present application is performed to obtain the credit level of the user. For example, Figure 3 is a flowchart of a user credit risk analysis method according to an embodiment of the present application, as Figure 3 shown, the method can include the following steps:

[0077] In step S301, an improved particle swarm optimization algorithm is used to optimize the initial weight and bias of each node in the first multi-layer perceptron (MLP).

[0078] In the technical solution provided in the step S301 of the application, the first MLP is applicable to the credit risk analysis model of the overdue card number of a user.

[0079] After the initial weights and biases of each node in the first MLP are optimized, the step S304 is entered, the initial weights and biases obtained through the optimization are substituted into the first MLP to obtain a first target MLP, and the user overdue card number credit risk analysis model is trained by using the first target MLP.

[0080] After the user overdue card number credit risk analysis model is trained, the step S307 is entered, the feature parameters of the user are input into the trained user overdue card number credit risk analysis model for risk prediction, and a first initial credit level can be obtained.

[0081] In the step S302, an improved particle swarm optimization algorithm is used to optimize the initial weights and biases of each node in the second MLP.

[0082] In the technical solution provided in the step S302 of the application, the second MLP is applicable to the credit risk analysis model of the overdue days of a user.

[0083] After the initial weights and biases of each node in the second MLP are optimized, the step S305 is entered, the initial weights and biases obtained through the optimization are substituted into the second MLP to obtain a second target MLP, and the user overdue days credit risk analysis model is trained by using the second target MLP.

[0084] After the user overdue days credit risk analysis model is trained, the step S308 is entered, the feature parameters of the user are input into the trained user overdue days credit risk analysis model for risk prediction, and a second initial credit level can be obtained.

[0085] In the step S303, an improved particle swarm optimization algorithm is used to optimize the initial weights and biases of each node in the third MLP.

[0086] In the technical solution provided in the step S303 of the application, the third MLP is applicable to the credit risk analysis model of the overdue amount of a user.

[0087] In this embodiment, the optimization process mentioned above can include the following steps:

[0088] Step one, initialize parameters, according to the parameter optimization number and the parameter interval, the total number of particles, the particle dimension, the search interval of each parameter of the particle and the search speed interval, the maximum and minimum values of the inertia weight coefficient are determined, and w max = 0.95, w min= 0.45, learning factors c1 = 2.05, c2 = 2.05, annealing process decay parameter lamda = 0.98, and maximum iteration number k max and an error threshold;

[0089] Step two, randomly generating initial speed and initial position of each particle, calculating fitness value of each particle, initializing individual optimal solution of particle and global optimal solution of population, global optimal fitness value of which is Gvalue, setting initial temperature value of simulated annealing as T = Gvalue / ln(5);

[0090] Step three, updating inertia weight coefficient w according to the following formula, and updating speed and position of particle. Wherein, D can represent dimension, flight speed of the i-th particle can be represented as V i = (V i1 , V i2 , … V iD ), position of the i-th particle can be represented as X i = (X i1 , X i2 , … X iD ), individual optimal position can be represented as PBest i = (p i1 , p i2 , … p iD ), and global optimal position of the whole particle population so far can be represented as GBest = (g1, g2, … g D ), can be used to represent the speed of the i-th particle in the d-th dimension in the k-th iteration, and Similarly, and can be used to represent a random number between 0 and 1;

[0091] w = 1.4 × (w max -w min ) + 0.5 × tanh (-2 + 4 × (k max -k / k max )(w max -w min )

[0092]

[0093] Step four, calculating fitness of each particle after moving and individual optimal solution of each particle after moving, integrating the idea of simulated annealing, using Metropolis criterion to determine whether to take the new solution as the individual optimal solution of the particle, if so, performing annealing operation;

[0094] Step five, update the global optimal solution of the particle;

[0095] Step six, the above steps are repeatedly executed until a termination condition is reached;

[0096] After the initial weights and biases of each node in the third MLP are optimized, the initial weights and biases obtained by optimization are substituted into the third multi-layer perceptron in step S306 to obtain a third target multi-layer perceptron, and the user overdue amount credit risk analysis model is trained using the third target multi-layer perceptron.

[0097] In the technical solution provided in the above step S306 of the present application, the user overdue amount credit risk analysis model is trained using the third target multi-layer perceptron, which can be completed by performing the following steps: preprocessing the data to remove outliers; determining the network structure, for example, in the network structure, the input layer can include 10 input nodes and 1 bias node, the output layer can include 4 nodes, the appropriate number of hidden layers and the number of neurons in each hidden layer are determined, and each initial value is set, the initial weights and biases of each neuron can be optimized by the improved particle swarm optimization algorithm to obtain the optimal initial value; for forward propagation, each neuron calculates the weighted input and applies the activation function, where the tanh function is used as the activation function; calculate the loss function, where the loss function can be calculated using the mean square error; for back propagation, calculate the gradient, update the weights and biases; and repeatedly train the network model and evaluate the trained network model until the prediction result of the network model meets the expectation.

[0098] After training the user overdue amount credit risk analysis model, the feature parameters of the user are input into the trained user overdue amount credit risk analysis model for risk prediction in step S309, and a third initial credit rating can be obtained.

[0099] In the technical solution provided in the above step S309 of the present application, on the basis of obtaining the user sample data, the user sample data is divided according to a target division ratio (for example, but not limited to: 7:3), so as to obtain a training set and a test set. The input of the three models can be basic feature data of the user, which can include personal basic information, personal asset status, personal historical credit status and personal family situation, etc., wherein the personal basic information can include age, gender, occupation, working years and income, etc. The above basic feature data is classified, the classified basic feature data is combined, the combined basic feature data is preprocessed, and the preprocessed basic feature data is denoised. By using principal component analysis and other methods, the feature data with the highest correlation degree with the credit level can be determined from the denoised basic feature data as the input of the model, and the feature data with the highest correlation degree with the credit level is normalized and the existing null values are filled.

[0100] In this embodiment, the output of the above three models is the credit level of the user. For the user overdue card number credit risk analysis model, the initial credit level of the user is determined according to the number of overdue cards, for example, if the number of overdue cards is 0, the initial credit level of the user is A level, if the number of overdue cards is 1, the initial credit level of the user is B level, if the number of overdue cards is 2, the initial credit level of the user is C level, and if the number of overdue cards is 3 or more, the initial credit level of the user is D level. Similarly, the user overdue day credit risk analysis model determines the initial credit level of the user according to the number of overdue days, the user overdue amount credit risk analysis model determines the initial credit level of the user according to the size of the overdue amount, and the initial credit level of the user under different analysis conditions is the output of the above three models.

[0101] After obtaining the above first initial credit level, the above second initial credit level and the above third initial credit level, step S310 is entered, the above first initial credit level, the above second initial credit level and the above third initial credit level are weighted and fused, and the target credit level can be obtained.

[0102] In this embodiment, when the automation test script is executed, the set of automation test scripts to be executed can be acquired, and the relationship between the automation test scripts in the set of automation test scripts is determined, respectively, between the normal automation test script and the abnormal automation test script. According to the determined relationship, the set of automation test script tasks to be executed can be constructed, and the automation test scripts in the constructed set of automation test script tasks are executed, and the test result can be obtained, thereby achieving the purpose of avoiding long time consumption of executing the automation test script, thereby solving the technical problem of low efficiency of the automation test, and further realizing the technical effect of improving the efficiency of the automation test.

[0103] According to the embodiments of the present application, a determination device for a credit rating of an object is also provided. It should be noted that the determination device for a credit rating of an object can be used to execute the determination method for a credit rating of an object in the embodiments.

[0104] Figure 4 is a schematic diagram of a determination device for a credit rating of an object according to an embodiment of the present application. As shown in Figure 4 the determination device for a credit rating of an object 400 can include an acquisition unit 401, a first determination unit 402, a prediction unit 403 and a second determination unit 404.

[0105] The acquisition unit 401 is configured to acquire a set of attribute information of an object, wherein the set of attribute information includes at least one attribute information, and the attribute information is used to represent an attribute associated with the credit rating of the object.

[0106] The first determination unit 402 is configured to determine target attribute information from the attribute information in the set of attribute information, wherein the degree of association between the target attribute information and the credit rating is greater than the degree of association between the remaining attribute information in the set of attribute information except the target attribute information and the credit rating.

[0107] The prediction unit 403 is configured to input the target attribute information into a target prediction model in a set of target prediction models for prediction to obtain an initial credit rating of the object, wherein the target prediction model is obtained by training an initial prediction model based on a set of sample attribute information of a sample object, and the set of sample attribute information includes at least one sample attribute information, and the sample attribute information is used to represent an attribute associated with the credit rating of the sample object.

[0108] The second determination unit 404 is configured to determine a target credit rating associated with the initial credit rating, wherein the initial credit rating is integrated in the target credit rating.

[0109] Optionally, the target prediction model set includes a first target prediction model, a second target prediction model, and a third target prediction model, the first target prediction model is a prediction model constructed for the number of overdue periods of a credit card held by the subject, the second target prediction model is a prediction model constructed for the overdue time of the credit card, and the third target prediction model is a prediction model constructed for the overdue amount of the credit card, and the prediction unit 403 can include: a first prediction module configured to input the target attribute information into the first target prediction model to perform prediction and obtain a first initial credit level; a second prediction module configured to input the target attribute information into the second target prediction model to perform prediction and obtain a second initial credit level; and a third prediction module configured to input the target attribute information into the third target prediction model to perform prediction and obtain a third initial credit level.

[0110] Optionally, the second determination unit 404 can include: a first determination module configured to determine a weight of the initial credit level; and a second determination module configured to determine the target credit level based on the weight and the initial credit level.

[0111] Optionally, the weight includes a first weight of the first initial credit level, a second weight of the second initial credit level, and a third weight of the third initial credit level, and the second determination module can include: a determination submodule configured to determine the target credit level based on the first weight, the second weight, the third weight, and the initial credit level.

[0112] Optionally, the determination submodule can determine the target credit level based on the first weight, the second weight, the third weight, and the initial credit level by performing the following steps: determining a first product between the first weight and the first initial credit level, a second product between the second weight and the second initial credit level, and a third product between the third weight and the third initial credit level; and determining a sum of the first product, the second product, and the third product as the target credit level.

[0113] Optionally, the first determination unit 402 can include: a third determination module configured to determine a degree of association between the attribute information in the attribute information set and the credit level; a fourth determination module configured to determine a target degree of association from the degrees of association, wherein the target degree of association is greater than the remaining degrees of association other than the target degree of association among the degrees of association; and a fifth determination module configured to determine the target attribute information as the attribute information corresponding to the target degree of association.

[0114] In this embodiment, the device for determining the credit rating of an object includes the following units: an acquisition unit for acquiring an attribute information set of the object, wherein the attribute information set includes at least one attribute information, which represents an attribute associated with the credit rating of the object; a first determination unit for determining target attribute information from the attribute information in the attribute information set, wherein the correlation between the target attribute information and the credit rating is greater than the correlation between the remaining attribute information in the attribute information set (excluding the target attribute information) and the credit rating; a prediction unit for inputting the target attribute information into a target prediction model in a target prediction model set to predict and obtain the initial credit rating of the object, wherein the target prediction model is trained on the initial prediction model based on a sample attribute information set of a sample object, the sample attribute information set including at least one sample attribute information, which represents an attribute associated with the credit rating of the sample object; and a second determination unit for determining a target credit rating associated with the initial credit rating, wherein the initial credit rating is incorporated into the target credit rating, thereby ensuring the comprehensiveness of the prediction results and solving the technical problem of low accuracy in determining the user's credit rating, thus achieving the technical effect of improving the accuracy of determining the user's credit rating.

[0115] It should be noted that the acquisition unit 401, the first determination unit 402, the prediction unit 403, and the second determination unit 404 mentioned above correspond to steps S201 to S204 in the embodiments. The four units and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules or units can be hardware components or software components stored in memory (e.g., memory 104) and processed by one or more processors (e.g., processors 102a, 102b, ..., 102n). The above modules can also be part of the device and run in the computer terminal 10 provided in Embodiment 1.

[0116] Embodiments of this application may provide an electronic device. Figure 5 This is a structural block diagram of an electronic device according to an embodiment of this application. Figure 5 As shown, the electronic device 500 may include: one or more ( Figure 5 (Only one is shown) processor 502, memory 504, memory controller 506, and peripheral interface 508, wherein peripheral interface 508 is connected to radio frequency module 5101, audio module 5102 and display 5103.

[0117] The memory can be used to store software programs and modules, such as program instructions / modules corresponding to the method and device in the embodiments of the present application. The processor executes various functions and data processing by running the software programs and modules stored in the memory, that is, implements the above method. The memory can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory can further include a memory remotely arranged with respect to the processor, which can be connected to the terminal through a network. Examples of the above network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0118] The processor can call information and applications stored in the memory through the transmission device to perform the following steps: obtaining an attribute information set of the object, wherein the attribute information set includes at least one attribute information, and the attribute information is used to represent an attribute associated with the credit level of the object; determining target attribute information from the attribute information of the attribute information set, wherein the degree of association between the target attribute information and the credit level is greater than the degree of association between the remaining attribute information in the attribute information set except the target attribute information and the credit level; inputting the target attribute information into a target prediction model in a target prediction model set for prediction to obtain an initial credit level of the object, wherein the target prediction model is obtained by training an initial prediction model based on a sample attribute information set of a sample object, and the sample attribute information set includes at least one sample attribute information, and the sample attribute information is used to represent an attribute associated with the credit level of the sample object; determining a target credit level associated with the initial credit level, wherein the initial credit level is fused in the target credit level.

[0119] The processor can also call information and applications stored in the memory through the transmission device to perform the following steps: inputting the target attribute information into a first target prediction model for prediction to obtain a first initial credit level; inputting the target attribute information into a second target prediction model for prediction to obtain a second initial credit level; inputting the target attribute information into a third target prediction model for prediction to obtain a third initial credit level.

[0120] The processor can also call information and applications stored in the memory through the transmission device to perform the following steps: determining a weight of the initial credit level; determining the target credit level based on the weight and the initial credit level.

[0121] The processor can further call information and application programs stored in the memory through the transmission device to perform the following steps: determining a first product between the first weight and the first initial credit level, a second product between the second weight and the second initial credit level, and a third product between the third weight and the third initial credit level; determining a sum between the first product, the second product and the third product as the target credit level.

[0122] The processor can further call information and application programs stored in the memory through the transmission device to perform the following steps: determining the target credit level based on the first weight, the second weight, the third weight and the initial credit level.

[0123] The processor can further call information and application programs stored in the memory through the transmission device to perform the following steps: determining a degree of association between the attribute information of the attribute information set and the credit level; determining a target degree of association from the degrees of association, wherein the target degree of association is greater than the remaining degrees of association except the target degree of association; and determining the attribute information corresponding to the target degree of association as the target attribute information.

[0124] In the embodiments of the present application, when determining the credit level of the object, the attribute information set of the object can be obtained. From the attribute information of the obtained attribute information set, the target attribute information can be determined, and the determined target attribute information is input into the target prediction model in the target prediction model set for prediction, so that the initial credit level of the object can be obtained, and the target credit level associated with the initial credit level can be determined. Thus, the purpose of ensuring the comprehensiveness of the prediction result is achieved, thereby solving the technical problem of low determination accuracy of the credit level of the user, and further achieving the technical effect of improving the determination accuracy of the credit level of the user.

[0125] Figure 5 is a structural block diagram of an electronic device according to an embodiment of the present application. Those skilled in the art can understand that, Figure 5 The structure shown is only schematic, and the electronic device can also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD terminal device, etc. Figure 5 It does not limit the structure of the above-mentioned electronic device. For example, the electronic device can further include more or fewer components (such as a network interface, a display device, etc.) than those shown in Figure 5 or have a different configuration from that shown in Figure 5 .

[0126] Those skilled in the art can understand that all or part of the steps of the various methods in the above embodiments can be instructed by programs to terminal device related hardware, and the programs can be stored in a computer readable storage medium, which can include a flash disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, etc.

[0127] According to the embodiments of the present application, a processor is also provided, which is used to run a program, wherein the program is executed to implement the method for reproducing the exception information according to the embodiments when the program is run by the processor.

[0128] According to the embodiments of the present application, an electronic device is also provided, which includes a memory storing an executable program, and a processor used to run the program, wherein the program is executed to implement the method for determining the credit level of the object according to the embodiments when the program is run.

[0129] According to another aspect of the embodiments of the present application, a computer readable storage medium is also provided. The computer readable storage medium includes a stored program, wherein the program is executed to control the device where the computer readable storage medium is located to implement the method for determining the credit level of the object according to the embodiments when the program is run.

[0130] According to the embodiments of the present application, a computer program product is also provided, which includes a computer program, wherein the computer program is executed to implement the method for determining the credit level of the object according to the embodiments when the computer program is run by a processor.

[0131] According to the embodiments of the present application, a computer program product is also provided, which includes a non-volatile computer readable storage medium used to store a computer program, wherein the computer program is executed to implement the method for determining the credit level of the object according to the embodiments when the computer program is run by a processor.

[0132] According to the embodiments of the present application, a computer program is also provided, which is executed to implement the method for determining the credit level of the object according to the embodiments when the computer program is run by a processor.

[0133] The above-mentioned serial numbers of the embodiments of the present application are only for description, and do not represent the advantages or disadvantages of the embodiments.

[0134] In the above-mentioned embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.

[0135] In several embodiments provided in the present application, it should be understood that the disclosed technology can be implemented by other ways. Among them, the above-described device embodiments are only schematic, for example, the division of units can be a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0136] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they can be located in one place, or they can be distributed to multiple units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0137] In addition, each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically, or two or more units can be integrated into one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0138] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the whole or part of the related technology that essentially contribute can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the embodiments of the present application. The foregoing storage medium includes: U disk, ROM, RAM, mobile hard disk, magnetic disk or optical disk and various program code storage media.

[0139] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can be made, and these improvements and refinements should be considered as the protection scope of the present application.

Claims

1. A method for determining the credit rating of an object, characterized in that, include: Obtain an attribute information set of an object, wherein the attribute information set includes at least one attribute information, the attribute information being used to represent an attribute associated with the credit rating of the object; From the attribute information in the attribute information set, target attribute information is determined, wherein the degree of correlation between the target attribute information and the credit rating is greater than the degree of correlation between the remaining attribute information in the attribute information set excluding the target attribute information and the credit rating. The target attribute information is input into the target prediction model in the target prediction model set for prediction to obtain the initial credit rating of the object. The target prediction model is trained on the initial prediction model based on the sample attribute information set of the sample object. The sample attribute information set includes at least one sample attribute information. The sample attribute information is used to represent the attribute associated with the credit rating of the sample object. The initial prediction model is constructed based on a multilayer perceptron. Determine a target credit rating associated with the initial credit rating, wherein the initial credit rating is incorporated into the target credit rating; The target prediction model set includes a first target prediction model, a second target prediction model, and a third target prediction model. The first target prediction model is a prediction model constructed for the number of overdue credit cards held by the object. The second target prediction model is a prediction model constructed for the overdue time of the credit cards. The third target prediction model is a prediction model constructed for the overdue amount of the credit cards. The target attribute information is input into the target prediction models in the target prediction model set for prediction to obtain the object's initial credit rating. This includes: inputting the target attribute information into the first target prediction model for prediction to obtain a first initial credit rating; inputting the target attribute information into the second target prediction model for prediction to obtain a second initial credit rating; and inputting the target attribute information into the third target prediction model for prediction to obtain a third initial credit rating. Determining a target credit rating associated with the initial credit rating includes: determining a weight for the initial credit rating, wherein the weight includes a first weight for the first initial credit rating, a second weight for the second initial credit rating, and a third weight for the third initial credit rating; and determining the target credit rating based on the first weight, the second weight, the third weight, and the initial credit rating.

2. The method according to claim 1, characterized in that, Determining the target credit rating based on the first weight, the second weight, the third weight, and the initial credit rating includes: Determine a first product between the first weight and the first initial credit rating, a second product between the second weight and the second initial credit rating, and a third product between the third weight and the third initial credit rating; The sum of the first product, the second product, and the third product is determined as the target credit rating.

3. The method according to claim 1, characterized in that, Determining target attribute information from the attribute information set includes: Determine the degree of correlation between the attribute information in the attribute information set and the credit rating; From the association degrees, a target association degree is determined, wherein the target association degree is greater than the remaining association degrees other than the target association degree. The attribute information corresponding to the degree of correlation of the target is determined as the target attribute information.

4. A device for determining the credit rating of an object, characterized in that, include: An acquisition unit is used to acquire an attribute information set of an object, wherein the attribute information set includes at least one attribute information, the attribute information being used to represent an attribute associated with the credit rating of the object; The first determining unit is configured to determine target attribute information from the attribute information in the attribute information set, wherein the degree of correlation between the target attribute information and the credit rating is greater than the degree of correlation between the remaining attribute information in the attribute information set excluding the target attribute information and the credit rating. A prediction unit is used to input the target attribute information into a target prediction model in a target prediction model set to make a prediction and obtain the initial credit rating of the object. The target prediction model is trained on the initial prediction model based on a sample attribute information set of the sample object. The sample attribute information set includes at least one sample attribute information. The sample attribute information is used to represent the attribute associated with the credit rating of the sample object. The initial prediction model is constructed based on a multilayer perceptron. The second determining unit is used to determine a target credit rating associated with the initial credit rating, wherein the initial credit rating is incorporated into the target credit rating; The target prediction model set includes a first target prediction model, a second target prediction model, and a third target prediction model. The first target prediction model is a prediction model constructed for the number of overdue credit cards held by the object. The second target prediction model is a prediction model constructed for the overdue time of the credit cards. The third target prediction model is a prediction model constructed for the overdue amount of the credit cards. The prediction unit is used to perform the following steps to input the target attribute information into the target prediction models in the target prediction model set for prediction to obtain the object's initial credit rating: inputting the target attribute information into the first target prediction model for prediction to obtain a first initial credit rating; inputting the target attribute information into the second target prediction model for prediction to obtain a second initial credit rating; inputting the target attribute information into the third target prediction model for prediction to obtain a third initial credit rating. The second determining unit is configured to determine a target credit rating associated with the initial credit rating by performing the following steps: determining a weight of the initial credit rating, wherein the weight includes a first weight of the first initial credit rating, a second weight of the second initial credit rating, and a third weight of the third initial credit rating; and determining the target credit rating based on the first weight, the second weight, the third weight, and the initial credit rating.

5. A processor, characterized in that, The processor is used to run a program, wherein the program, when run by the processor, executes the method for determining the credit rating of the object according to any one of claims 1 to 3.

6. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, executes a method for determining the credit rating of the object according to any one of claims 1 to 3.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored executable program, wherein, when the executable program is executed, it controls the device on which the storage medium is located to perform a method for determining the credit rating of the object according to any one of claims 1 to 3.

Images