A personal credit assessment method and device based on artificial intelligence

Abstract

The present application relates to the field of data processing technology, and in particular to an artificial intelligence-based personal credit assessment method and device for improving the accuracy of personal credit assessment. The main scheme is: based on historical consumption data, predict the expected consumption data of the user to be assessed and the expected consumption data of the immediate family members in the next period corresponding to the user to be assessed and his immediate family members respectively; calculate the support coefficient of the user to be assessed through the user basic data, expected consumption data, and historical repayment record data of the immediate family members; determine the data feature vector of the user to be assessed through the user basic data, expected consumption data, historical repayment record data, support coefficient, and repayment ability coefficient of the user to be assessed; input the data feature vector of the user to be assessed into the personal credit assessment model to obtain the personal credit assessment level corresponding to the user to be assessed.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a personal credit assessment method and device based on artificial intelligence. Background Art

[0002] Credit investigation is the collection, organization, preservation, and processing of credit information of natural persons, legal persons, and other organizations in accordance with the law, and the provision of credit reports, credit assessments, credit consulting, and other services to help customers determine and control credit risks and conduct credit management activities. Credit investigation plays an important basic role in promoting the development of the credit economy and the construction of a social credit system: first, preventing credit risks; second, expanding credit transactions; third, improving economic efficiency; and fourth, promoting the construction of a social credit system.

[0003] With the development of the market economy, people's consumption habits have gradually shifted to a model of consuming first and paying later. The negative impact of the consumption method of prepaying future assets is that there are more and more defaults, which has led to credit institutions paying more and more attention to customers' credit rating evaluation and credit risk analysis. The traditional method of using established default facts to evaluate personal credit ratings has a certain delay and a low evaluation accuracy rate. Summary of the invention

[0004] In view of this, the present application provides a personal credit assessment method, device, electronic device and storage medium based on artificial intelligence, which are used to improve the accuracy of personal credit assessment.

[0005] In a first aspect, an embodiment of the present application provides a personal credit assessment method based on artificial intelligence, the method comprising:

[0006] Obtain the user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members;

[0007] Based on the historical consumption data, predict the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members;

[0008] Calculate the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives;

[0009] Calculate the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated;

[0010] Determine the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated;

[0011] The data feature vector of the user to be evaluated is input into the personal credit evaluation model to obtain the personal credit evaluation grade corresponding to the user to be evaluated.

[0012] In an optional embodiment provided by the present invention, the predicting, based on the historical consumption data, the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members includes:

[0013] Acquire historical consumption data of the current cycle and N cycles before it, and historical consumption data of the same period as the next cycle from the historical consumption data of the user to be evaluated, and use the acquired historical consumption data as first historical consumption data;

[0014] Acquire historical consumption data of the current cycle and N cycles before it, and historical consumption data of the same period as the next cycle from the historical consumption data of the direct relatives, and use the acquired historical consumption data as second historical consumption data;

[0015] The expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members corresponding to the user to be evaluated and the immediate family members in the next period are predicted respectively according to the first historical consumption data and the second historical consumption data.

[0016] In an optional embodiment provided by the present invention, predicting the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members of the user to be evaluated and the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members according to the first historical consumption data and the second historical consumption data, includes:

[0017] Converting the first historical consumption data and the second historical consumption data into a first eigenvector matrix and a second eigenvector matrix respectively;

[0018] The first eigenvector matrix and the second eigenvector matrix are respectively input into the user consumption prediction model to obtain the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period.

[0019] In an optional embodiment provided by the present invention, the calculation of the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives includes:

[0020] Calculating a first consumption difference by using the expected consumption data and historical repayment record data of the direct relatives;

[0021] The assistance coefficient of the user to be evaluated is calculated according to the user basic data of the immediate family member and the first consumption difference.

[0022] In an optional embodiment provided by the present invention, the calculating the assistance coefficient of the user to be evaluated according to the user basic data of the direct relatives and the first consumption difference includes:

[0023] Convert the basic user data of each direct relative and the first consumption difference into feature vectors respectively;

[0024] Merge the eigenvectors of all immediate family members to obtain the target eigenvector matrix;

[0025] Calculating the similarity between the target eigenvector matrix and each eigenvector matrix in the eigenvector library;

[0026] The assistance coefficient corresponding to the feature vector matrix having the highest similarity with the target feature vector matrix in the feature vector library is determined as the assistance coefficient of the user to be evaluated.

[0027] In an optional embodiment provided by the present invention, the calculation of the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated includes:

[0028] Calculating a second consumption difference by using the expected consumption data and historical repayment record data of the user to be evaluated;

[0029] The repayment ability coefficient of the user to be evaluated is calculated according to the user basic data of the user to be evaluated, the second consumption difference and the assistance coefficient.

[0030] In an optional embodiment provided by the present invention, the calculating the repayment ability coefficient of the user to be evaluated according to the user basic data of the user to be evaluated, the second consumption difference and the assistance coefficient includes:

[0031] The user basic data, the second consumption difference and the assistance coefficient of the user to be evaluated are input into the repayment ability model to obtain the repayment ability coefficient of the user to be evaluated.

[0032] In an optional embodiment provided by the present invention, determining the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated includes:

[0033] Determining a first repayment weight and a second repayment weight respectively according to the assistance coefficient and the repayment capacity coefficient;

[0034] A data feature vector of the user to be evaluated is determined according to the user basic data of the user to be evaluated, the first repayment weight, the second repayment weight, and the expected consumption data.

[0035] In an optional embodiment provided by the present invention, determining the data feature vector of the user to be evaluated according to the user basic data of the user to be evaluated, the first repayment weight, the second repayment weight and the expected consumption data includes:

[0036] Calculating the product of the first repayment weight and the second repayment weight with the expected repayment data respectively, and adding the obtained product results to obtain the expected repayment data;

[0037] The basic user data and expected repayment data of the user to be evaluated are converted into a data feature vector of the user to be evaluated.

[0038] In a second aspect, the embodiment of the present application further provides a personal credit assessment device based on artificial intelligence, the device comprising:

[0039] An acquisition module is used to obtain user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members;

[0040] A prediction module, used for predicting the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members based on the historical consumption data;

[0041] A calculation module, used to calculate the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives;

[0042] The calculation module is further used to calculate the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated;

[0043] A determination module, used to determine the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated;

[0044] The prediction module is further used to input the data feature vector of the user to be evaluated into the personal credit evaluation model to obtain the personal credit evaluation level corresponding to the user to be evaluated.

[0045] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and the machine-readable instructions are executed by the processor to execute the steps of the artificial intelligence-based personal credit assessment method of the first aspect.

[0046] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the artificial intelligence-based personal credit assessment method in the first aspect are executed.

[0047] An artificial intelligence-based personal credit assessment method, device, electronic device and storage medium provided in an embodiment of the present application first obtain user basic data, historical consumption data and historical repayment record data corresponding to the user to be assessed and his / her immediate family members respectively, then predict the expected consumption data of the user to be assessed and the expected consumption data of the immediate family members corresponding to the user to be assessed and his / her immediate family members in the next period respectively based on the historical consumption data, then calculate the support coefficient of the user to be assessed through the user basic data, expected consumption data and historical repayment record data of the immediate family members, and calculate the repayment capacity coefficient of the user to be assessed through the expected consumption data, user basic data, historical repayment record data and support coefficient of the user to be assessed; determine the data feature vector of the user to be assessed through the user basic data, expected consumption data, historical repayment record data of the user to be assessed and the support coefficient and repayment capacity coefficient of the user to be assessed; finally, input the data feature vector of the user to be assessed into the personal credit assessment model to obtain the personal credit assessment level corresponding to the user to be assessed. In the prior art, personal credit is evaluated based on the established facts of default. This application determines the personal credit evaluation level of the user to be evaluated by predicting the expected consumption data, the assistance coefficient, and the repayment ability coefficient. This application can improve the timeliness and accuracy of the personal credit evaluation level.

[0048] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A flowchart of an artificial intelligence-based personal credit assessment method provided in an embodiment of the present application is shown;

[0051] Figure 2 A structural block diagram of an artificial intelligence-based personal credit assessment device provided in an embodiment of the present application is shown;

[0052] Figure 3A schematic diagram of a computer device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0053] The terms "first", "second", "third", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects rather than to limit a specific order.

[0054] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way for easy understanding.

[0055] In the description of this application, unless otherwise specified, " / " indicates that the objects associated before and after are in an "or" relationship, for example, A / B can represent A or B; "and / or" in this application is only a description of the association relationship of associated objects, indicating that there can be three relationships, for example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. In addition, in the description of this application, unless otherwise specified, "multiple" refers to two or more than two. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0056] In the embodiments of the present application, at least one can also be described as one or more, and multiple can be two, three, four or more, which is not limited in the present application.

[0057] like Figure 1 As shown, the embodiment of the present application provides a personal credit assessment method based on artificial intelligence, which is applied to a terminal device. The personal credit assessment method based on artificial intelligence provided by the present application may include:

[0058] S10. Obtain user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members.

[0059] Among them, user basic data may include: name, age, occupation, education, income, contact information, home address and other data; historical consumption data may include consumption data on various consumption platforms and consumption data of entities; historical repayment record data is the historical repayment record data of the user to be evaluated and his / her immediate family members, which is not specifically limited in this embodiment. It should be noted that the immediate family members in this embodiment may be the parents, children, spouse and other relatives with legal relationship of the user to be evaluated.

[0060] S20. Predicting expected consumption data of the user to be evaluated and expected consumption data of the immediate family members of the user to be evaluated and the immediate family members in the next period respectively based on the historical consumption data.

[0061] In this embodiment, the repayment cycle can be set monthly, weekly or at predetermined intervals, and then the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members are predicted based on the historical consumption data of the historical repayment cycle. The expected consumption data is the expected consumption amount of the user in the next cycle, which can be obtained by statistical calculation of the historical consumption data, or by prediction through a time prediction model, or by fitting according to a linear fitting model to obtain the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next cycle, which is not specifically limited in this embodiment.

[0062] In an optional embodiment provided in the present application, the predicting of the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members of the user to be evaluated and the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members based on the historical consumption data includes:

[0063] S201. Obtain historical consumption data of the current cycle and N cycles before it, as well as historical consumption data of the same period as the next cycle from historical consumption data of the user to be evaluated, and use the acquired historical consumption data as first historical consumption data.

[0064] For example, if the current cycle is October, you can obtain historical consumption data from June to October of this year (N is 4), as well as historical consumption data for November of historical years (such as historical consumption data for November of 2020-2023). Then, all the acquired historical data will be used as the first historical consumption data, so that in subsequent steps, the expected consumption data of the user to be evaluated in the next cycle can be predicted based on the first historical consumption data.

[0065] S202: Obtain historical consumption data of the current cycle and the N cycles before it, as well as historical consumption data of the same period as the next cycle from historical consumption data of the direct relatives, and use the acquired historical consumption data as second historical consumption data.

[0066] It should be noted that the second historical consumption data is obtained in the same manner as the first historical consumption data, which will not be described in detail in this embodiment.

[0067] S203 , predicting the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members according to the first historical consumption data and the second historical consumption data.

[0068] In this embodiment, the expected consumption data of the user to be evaluated in the next cycle is predicted based on the first historical consumption data; and the expected consumption data of the immediate family members is predicted based on the second historical consumption data.

[0069] Specifically, predicting the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members corresponding to the user to be evaluated and the immediate family members in the next period respectively based on the first historical consumption data and the second historical consumption data includes: converting the first historical consumption data and the second historical consumption data into a first eigenvector matrix and a second eigenvector matrix respectively; inputting the first eigenvector matrix and the second eigenvector matrix into a user consumption prediction model respectively, to obtain the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively.

[0070] Among them, the user consumption prediction model is a time prediction model, which is trained based on sample data and sample labels, that is, the sample data is the consumption data of N historical periods and M historical consumption data corresponding to the same period, which is then converted into a feature vector matrix and input into the time prediction model for model training. When the accuracy of the model reaches a certain value, the training of the user consumption prediction model is completed.

[0071] In this embodiment, the user consumption prediction model includes a first training module and a second training module. The first training module is used to learn and train the consumption data of N historical periods in the feature vector matrix; the second training module is used to learn and train M historical consumption data of the same period, and then calculate a loss value respectively. When the weighted result of the two loss values ​​is less than a certain value, the training of the user consumption prediction model is completed.

[0072] In the present embodiment, the training of the user consumption prediction model retains both the data analysis of the user's recent consumption and the historical consumption data of the same period, so that the expected consumption data of the next cycle can be accurately predicted by the user consumption prediction model in the present embodiment. For example, the consumption of the annual Double 11 will be burst-type incremental consumption, but because the user is affected by the economic situation this year, his income has been reduced, and the reduction in income naturally leads to a reduction in his consumption capacity. And only by predicting its expected consumption data through its recent consumption data, there will be predicted consumption data that will be lower than its actual consumption (because it will be stockpiled for consumption on Double 11), and only based on its historical consumption data in November, there will be predicted consumption data that is higher than its actual consumption (due to this year's income reduction and consumption capacity is also reduced), so the present embodiment takes into account the above two data, so the accuracy of the expected consumption data prediction can be improved by the consumption prediction model used in the present embodiment.

[0073] S30. Calculate the assistance coefficient for the user to be evaluated through the basic user data, expected consumption data, and historical repayment record data of the immediate family members.

[0074] Among them, the assistance coefficient is used to indicate the ability coefficient of direct relatives to provide assistance in repayment for the user to be evaluated. The higher the coefficient, the higher the repayment amount provided by the direct relatives for the user to be evaluated; conversely, the lower the coefficient, the lower the repayment amount provided by the direct relatives for the user to be evaluated.

[0075] It should be noted that in real life, if the user to be evaluated is a student who is still in school or has just graduated, his income is not enough to fully cover his consumption capacity. At this time, his credit repayment mainly depends on his parents or other immediate family members. Therefore, this embodiment needs to calculate the assistance coefficient of the immediate family members who can help the user to be evaluated to repay the loan based on the user's basic data, expected consumption data, and historical repayment record data. The higher the coefficient, the higher the amount that can be repaid for the user to be evaluated.

[0076] In an optional embodiment provided by the present application, the calculation of the assistance coefficient for the user to be evaluated through the basic user data, expected consumption data, and historical repayment record data of the direct relatives includes:

[0077] S301. Calculate a first consumption difference using expected consumption data and historical repayment record data of direct relatives.

[0078] Among them, the first consumption difference can be used to represent the consumption balance of the next cycle. For example, if the expected consumption data of the direct relatives in the next cycle is 1000, the corresponding historical repayment record data is averaged to 2000. After calculation, the first consumption difference is 1000, which means that the direct relatives have at least 1000 to help the user to be evaluated, thereby realizing the credit repayment of the user to be evaluated.

[0079] S302: Calculate the assistance coefficient of the user to be evaluated according to the basic user data of the immediate family members and the first consumption difference.

[0080] Specifically, the method of calculating the assistance coefficient of the user to be evaluated based on the user basic data of the immediate family members and the first consumption difference includes: converting the user basic data of each immediate family member and the first consumption difference into feature vectors respectively; merging the feature vectors of all immediate family members to obtain a target feature vector matrix; calculating the similarity between the target feature vector matrix and each feature vector matrix in a feature vector library; and determining the assistance coefficient corresponding to the feature vector matrix in the feature vector library that has the highest similarity to the target feature vector matrix as the assistance coefficient of the user to be evaluated.

[0081] It should be noted that since the user to be evaluated may correspond to multiple direct relatives, the user basic data and the first consumption difference of each direct love letter are required, and they are converted into feature vectors respectively. Then, the feature vectors of all relatives are merged to obtain the target feature vector matrix, and each row in the target feature vector matrix represents the feature vector of a direct relative.

[0082] In this embodiment, the feature vector library stores the support coefficients corresponding to multiple feature vectors. This embodiment first calculates the similarity between each row of feature vectors in the target feature vector matrix and the feature vector matrix in the feature vector library, then obtains the feature vector matrix with the highest similarity, and then determines the support coefficient corresponding to the feature vector matrix with the highest similarity as the support coefficient of the user to be evaluated. Finally, the support coefficients of all direct relatives for the user to be evaluated are calculated, and then all support coefficients are accumulated and summed to obtain the final support coefficient.

[0083] S40, calculating the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated.

[0084] In an optional embodiment provided in the present application, the repayment ability coefficient of the user to be evaluated is calculated by using the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated, including:

[0085] S401. Calculate a second consumption difference using expected consumption data and historical repayment record data of the user to be evaluated.

[0086] The calculation method of the second consumption difference is the same as the calculation method of the first consumption difference, which will not be described in detail in this embodiment.

[0087] S402: Calculate the repayment ability coefficient of the user to be evaluated according to the user basic data of the user to be evaluated, the second consumption difference and the assistance coefficient.

[0088] Specifically, calculating the repayment ability coefficient of the user to be evaluated based on the user basic data of the user to be evaluated, the second consumption difference and the assistance coefficient includes: inputting the user basic data of the user to be evaluated, the second consumption difference and the assistance coefficient into a repayment ability model to obtain the repayment ability coefficient of the user to be evaluated.

[0089] Among them, the repayment ability model is trained based on sample data and corresponding sample labels. The sample data includes the basic user data, the second consumption difference and the assistance coefficient of the sample user. The sample label is the marked repayment ability coefficient. The repayment ability coefficient can be specifically represented by a number, such as 0-1. 0 means that the sample user cannot be provided with assistance repayment, and 1 means that the sample user can be fully repaid. For example, if the sample user owes 3,000 yuan, his corresponding immediate family members can repay 3,000 yuan in full for him.

[0090] Specifically, the repayment ability model includes an input layer, a feature extraction layer, a convolution layer and an output layer. In this implementation, the user basic data, the second consumption difference and the assistance coefficient of the sample user are input into the input layer of the repayment ability model, and then input into the feature extraction layer through the input layer. The data features corresponding to the user basic data, the second consumption difference and the assistance coefficient are extracted through the feature extraction layer, and then the predicted repayment ability coefficient corresponding to the sample user is calculated through the convolution layer. Then, the loss value is calculated based on the predicted repayment ability coefficient and the marked repayment ability coefficient. When the loss value is less than the preset value, the training of the repayment ability model is completed.

[0091] Among them, the loss value can be calculated by the following loss function:

[0092]

[0093] Among them, n is the number of sample data, ab i is the predicted repayment capacity coefficient obtained based on the user basic data of the sample user and the second consumption difference; ac i is the predicted repayment capacity coefficient obtained based on the user basic data and assistance coefficient of the sample user; bc iis the predicted repayment capacity coefficient obtained based on the second consumption difference and assistance coefficient of the sample user, y i is the marked repayment ability coefficient corresponding to the sample user.

[0094] S50, determining the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, support coefficient, and repayment ability coefficient of the user to be evaluated.

[0095] In an optional embodiment provided by the present application, determining the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated includes:

[0096] S501. Determine a first repayment weight and a second repayment weight according to the assistance coefficient and the repayment capacity coefficient respectively.

[0097] In this embodiment, the corresponding first repayment weight and second repayment weight can be determined based on the historical repayment record data of the user to be evaluated and the direct relatives. If the historical repayment record data of the user to be evaluated shows that his repayment credit is poor, and the historical repayment record data of the direct relatives shows that his repayment credit is good, then the corresponding weight value is set according to their historical repayment situation, and then the assistance coefficient and the corresponding weight value are multiplied to obtain the first repayment weight, and the repayment ability coefficient and the corresponding weight value are multiplied to obtain the second repayment weight.

[0098] S502: Determine a data feature vector of the user to be evaluated according to the user basic data, the first repayment weight, the second repayment weight, and the expected consumption data of the user to be evaluated.

[0099] Specifically, determining the data feature vector of the user to be evaluated based on the user basic data, the first repayment weight, the second repayment weight and the expected consumption data of the user to be evaluated includes: multiplying the first repayment weight and the second repayment weight by the expected repayment data respectively, and adding the obtained product results to obtain the expected repayment data; converting the user basic data and the expected repayment data of the user to be evaluated into the data feature vector of the user to be evaluated.

[0100] S60: Input the data feature vector of the user to be evaluated into a personal credit evaluation model to obtain a personal credit evaluation grade corresponding to the user to be evaluated.

[0101] The personal credit assessment model is a pre-trained neural network model, that is, it is trained based on the sample data feature vector and its corresponding credit assessment label. The method of obtaining the sample data feature vector is the same as the method of obtaining the data feature vector of the user to be assessed.

[0102] An artificial intelligence-based personal credit assessment method provided in an embodiment of the present application first obtains user basic data, historical consumption data, and historical repayment record data corresponding to the user to be assessed and his / her immediate family members, respectively; then, based on the historical consumption data, predicts the expected consumption data of the user to be assessed and the expected consumption data of the immediate family members corresponding to the user to be assessed and his / her immediate family members in the next period, respectively; then, calculates the assistance coefficient of the user to be assessed through the user basic data, expected consumption data, and historical repayment record data of the immediate family members; and calculates the repayment capacity coefficient of the user to be assessed through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be assessed; determines the data feature vector of the user to be assessed through the user basic data, expected consumption data, historical repayment record data of the user to be assessed, as well as the assistance coefficient and the repayment capacity coefficient of the user to be assessed; and finally, inputs the data feature vector of the user to be assessed into a personal credit assessment model to obtain the personal credit assessment grade corresponding to the user to be assessed. In the prior art, personal credit is evaluated based on the established facts of default. This application determines the personal credit evaluation level of the user to be evaluated by predicting the expected consumption data, the assistance coefficient, and the repayment ability coefficient. This application can improve the timeliness and accuracy of the personal credit evaluation level.

[0103] In the case of dividing each functional module into corresponding functional modules, Figure 2 A possible schematic diagram of the composition of the personal credit assessment device based on artificial intelligence involved in the above and embodiments is shown, such as Figure 2 As shown, the personal credit assessment device based on artificial intelligence may include:

[0104] The acquisition module 21 is used to obtain the user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members;

[0105] A prediction module 22, configured to predict expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members of the user to be evaluated and the immediate family members in the next period respectively based on the historical consumption data;

[0106] A calculation module 23, used to calculate the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives;

[0107] The calculation module 23 is further used to calculate the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated;

[0108] A determination module 24, configured to determine a data feature vector of the user to be evaluated through basic user data, expected consumption data, historical repayment record data, assistance coefficient, and repayment capacity coefficient of the user to be evaluated;

[0109] The prediction module 22 is further used to input the data feature vector of the user to be evaluated into the personal credit evaluation model to obtain the personal credit evaluation level corresponding to the user to be evaluated.

[0110] In an optional embodiment provided by the present invention, the prediction module 22 is specifically used for:

[0111] Acquire historical consumption data of the current cycle and N cycles before it, and historical consumption data of the same period as the next cycle from the historical consumption data of the user to be evaluated, and use the acquired historical consumption data as first historical consumption data;

[0112] Acquire historical consumption data of the current cycle and N cycles before it, and historical consumption data of the same period as the next cycle from the historical consumption data of the direct relatives, and use the acquired historical consumption data as second historical consumption data;

[0113] The expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members corresponding to the user to be evaluated and the immediate family members in the next period are predicted respectively according to the first historical consumption data and the second historical consumption data.

[0114] In an optional embodiment provided by the present invention, the prediction module 22 is specifically used for:

[0115] Converting the first historical consumption data and the second historical consumption data into a first eigenvector matrix and a second eigenvector matrix respectively;

[0116] The first eigenvector matrix and the second eigenvector matrix are respectively input into the user consumption prediction model to obtain the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period.

[0117] In an optional embodiment provided by the present invention, the calculation module 23 is specifically used for:

[0118] Calculating a first consumption difference by using the expected consumption data and historical repayment record data of the direct relatives;

[0119] The assistance coefficient of the user to be evaluated is calculated according to the user basic data of the immediate family member and the first consumption difference.

[0120] In an optional embodiment provided by the present invention, the calculation module 23 is specifically used for:

[0121] Convert the basic user data of each direct relative and the first consumption difference into feature vectors respectively;

[0122] Merge the eigenvectors of all immediate family members to obtain the target eigenvector matrix;

[0123] Calculating the similarity between the target eigenvector matrix and each eigenvector matrix in the eigenvector library;

[0124] The assistance coefficient corresponding to the feature vector matrix having the highest similarity with the target feature vector matrix in the feature vector library is determined as the assistance coefficient of the user to be evaluated.

[0125] In an optional embodiment provided by the present invention, the calculation module 23 is specifically used for:

[0126] Calculating a second consumption difference by using the expected consumption data and historical repayment record data of the user to be evaluated;

[0127] The repayment ability coefficient of the user to be evaluated is calculated according to the user basic data of the user to be evaluated, the second consumption difference and the assistance coefficient.

[0128] In an optional embodiment provided by the present invention, the calculation module 23 is specifically used for:

[0129] The user basic data, the second consumption difference and the assistance coefficient of the user to be evaluated are input into the repayment ability model to obtain the repayment ability coefficient of the user to be evaluated.

[0130] In an optional embodiment provided by the present invention, the determination module 24 is specifically configured to:

[0131] Determining a first repayment weight and a second repayment weight respectively according to the assistance coefficient and the repayment capacity coefficient;

[0132] A data feature vector of the user to be evaluated is determined according to the user basic data of the user to be evaluated, the first repayment weight, the second repayment weight, and the expected consumption data.

[0133] In an optional embodiment provided by the present invention, the determination module 24 is specifically configured to:

[0134] Calculating the product of the first repayment weight and the second repayment weight with the expected repayment data respectively, and adding the obtained product results to obtain the expected repayment data;

[0135] The basic user data and expected repayment data of the user to be evaluated are converted into a data feature vector of the user to be evaluated.

[0136] For the specific definition of the device, please refer to the definition of the personal credit assessment method based on artificial intelligence above, which will not be repeated here. Each module in the above device can be implemented in whole or in part by software, hardware and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0137] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a personal credit assessment method based on artificial intelligence is implemented.

[0138] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0139] Obtain the user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members;

[0140] Based on the historical consumption data, predict the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members;

[0141] Calculate the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives;

[0142] Calculate the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated;

[0143] Determine the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated;

[0144] The data feature vector of the user to be evaluated is input into the personal credit evaluation model to obtain the personal credit evaluation grade corresponding to the user to be evaluated.

[0145] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0146] Obtain the user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members;

[0147] Based on the historical consumption data, predict the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members;

[0148] Calculate the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives;

[0149] Calculate the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated;

[0150] Determine the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated;

[0151] The data feature vector of the user to be evaluated is input into the personal credit evaluation model to obtain the personal credit evaluation grade corresponding to the user to be evaluated.

[0152] In one embodiment, a computer program product is provided, the computer program product comprising a computer program, the computer program being executed by a processor to implement the following steps:

[0153] Obtain the user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members;

[0154] Based on the historical consumption data, predict the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members;

[0155] Calculate the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives;

[0156] Calculate the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated;

[0157] Determine the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated;

[0158] The data feature vector of the user to be evaluated is input into the personal credit evaluation model to obtain the personal credit evaluation grade corresponding to the user to be evaluated.

[0159] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0160] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0161] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A personal credit assessment method based on artificial intelligence, characterized in that: The method comprises: Obtain the user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members; Based on the historical consumption data, predict the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members; Calculate the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives; Calculate the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated; Determine the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated; The data feature vector of the user to be evaluated is input into the personal credit evaluation model to obtain the personal credit evaluation grade corresponding to the user to be evaluated.

2. The method according to claim 1, characterized in that The predicting, based on the historical consumption data, the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members includes: Acquire historical consumption data of the current cycle and N cycles before it, and historical consumption data of the same period as the next cycle from the historical consumption data of the user to be evaluated, and use the acquired historical consumption data as first historical consumption data; Acquire historical consumption data of the current cycle and N cycles before it, and historical consumption data of the same period as the next cycle from the historical consumption data of the direct relatives, and use the acquired historical consumption data as second historical consumption data; The expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members corresponding to the user to be evaluated and the immediate family members in the next period are predicted respectively according to the first historical consumption data and the second historical consumption data.

3. The method according to claim 2, characterized in that The predicting, based on the first historical consumption data and the second historical consumption data, respectively the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period corresponding to the user to be evaluated and the immediate family members, respectively, comprises: Converting the first historical consumption data and the second historical consumption data into a first eigenvector matrix and a second eigenvector matrix respectively; The first eigenvector matrix and the second eigenvector matrix are respectively input into the user consumption prediction model to obtain the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period.

4. The method according to claim 1, characterized in that The calculation of the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives includes: Calculating a first consumption difference by using the expected consumption data and historical repayment record data of the direct relatives; The assistance coefficient of the user to be evaluated is calculated according to the user basic data of the immediate family member and the first consumption difference.

5. The method according to claim 4, characterized in that The calculating the assistance coefficient of the user to be evaluated according to the user basic data of the direct relatives and the first consumption difference includes: Convert the user basic data of each direct relative and the first consumption difference into feature vectors respectively; Merge the eigenvectors of all immediate family members to obtain the target eigenvector matrix; Calculating the similarity between the target eigenvector matrix and each eigenvector matrix in the eigenvector library; The assistance coefficient corresponding to the feature vector matrix having the highest similarity with the target feature vector matrix in the feature vector library is determined as the assistance coefficient of the user to be evaluated.

6. The method according to claim 5, characterized in that The calculation of the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated includes: Calculating a second consumption difference by using the expected consumption data and historical repayment record data of the user to be evaluated; The repayment ability coefficient of the user to be evaluated is calculated according to the user basic data of the user to be evaluated, the second consumption difference and the assistance coefficient.

7. The method according to claim 6, characterized in that The calculating the repayment ability coefficient of the user to be evaluated according to the user basic data of the user to be evaluated, the second consumption difference and the assistance coefficient includes: The user basic data, the second consumption difference and the assistance coefficient of the user to be evaluated are input into the repayment ability model to obtain the repayment ability coefficient of the user to be evaluated.

8. The method according to claim 4 or 6, characterized in that: The step of determining the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated includes: Determining a first repayment weight and a second repayment weight respectively according to the assistance coefficient and the repayment capacity coefficient; A data feature vector of the user to be evaluated is determined according to the user basic data of the user to be evaluated, the first repayment weight, the second repayment weight, and the expected consumption data.

9. The method according to claim 8, characterized in that The step of determining the data feature vector of the user to be evaluated according to the user basic data of the user to be evaluated, the first repayment weight, the second repayment weight, and the expected consumption data includes: Calculating the product of the first repayment weight and the second repayment weight with the expected repayment data respectively, and adding the obtained product results to obtain the expected repayment data; The basic user data and expected repayment data of the user to be evaluated are converted into a data feature vector of the user to be evaluated.

10. A personal credit assessment device based on artificial intelligence, characterized in that: The device comprises: An acquisition module is used to obtain user basic data, historical consumption data, and historical repayment record data corresponding to the user to be evaluated and his / her immediate family members; A prediction module, used for predicting the expected consumption data of the user to be evaluated and the expected consumption data of the immediate family members in the next period respectively corresponding to the user to be evaluated and the immediate family members based on the historical consumption data; A calculation module, used to calculate the assistance coefficient for the user to be evaluated based on the basic user data, expected consumption data, and historical repayment record data of the direct relatives; The calculation module is further used to calculate the repayment ability coefficient of the user to be evaluated through the expected consumption data, user basic data, historical repayment record data, and assistance coefficient of the user to be evaluated; A determination module, used to determine the data feature vector of the user to be evaluated through the user basic data, expected consumption data, historical repayment record data, assistance coefficient, and repayment ability coefficient of the user to be evaluated; The prediction module is further used to input the data feature vector of the user to be evaluated into the personal credit evaluation model to obtain the personal credit evaluation level corresponding to the user to be evaluated.

Images