CN113450211A

CN113450211A - User credit granting method, device, electronic equipment and storage medium

Info

Publication number: CN113450211A
Application number: CN202111013686.4A
Authority: CN
Inventors: 吴云崇; 邹志鹏; 陈志鹏; 陈鹏
Original assignee: Alipay Hangzhou Information Technology Co Ltd
Current assignee: Alipay Hangzhou Information Technology Co Ltd
Priority date: 2021-08-31
Filing date: 2021-08-31
Publication date: 2021-09-28

Abstract

The application provides a user credit granting method, a user credit granting device, electronic equipment and a storage medium. Wherein the method may comprise: and acquiring user data corresponding to the user to be trusted. And inputting the user data into a loan credit granting model for calculation to obtain a loan credit granting result corresponding to the user. The loan credit granting model comprises an artificial intelligence model which takes the standard of financial risk indexes after loan facing a credit granting user group as a constraint condition, takes the maximization of the loan facing the credit granting user group as an optimization target, and periodically performs dynamic update based on the user data of historical users. And loan is carried out to the user based on the obtained loan credit granting result.

Description

User credit granting method, device, electronic equipment and storage medium

Technical Field

One or more embodiments of the present application relate to the field of credit, and in particular, to a user credit granting method, apparatus, electronic device, and storage medium.

Background

The credit granting means that the credit granting organization provides funds for the user to be credited.

After receiving the credit granting application of the user to be granted, the credit granting organization can determine whether the user is a credit granting user or not by using a credit granting model based on the user data corresponding to the user to be granted, and determine the credit granting amount and the credit granting rate of the user under the condition that the user is determined to be the credit granting user, so as to credit the user based on the determined credit granting amount and the credit granting rate.

The credit granting organization obtains income by depending on credit granting for the user, and the high income can promote the credit granting behavior of the credit granting organization, thereby promoting the social development. Thus, there is a need for a user trust method that maximizes the long-term revenue for an organization.

Disclosure of Invention

In view of this, the present application provides a user credit granting method. The method can comprise the following steps: acquiring user data corresponding to a user to be trusted; inputting the user data into a loan credit granting model for calculation to obtain a loan credit granting result corresponding to the user; the loan trust model comprises an artificial intelligence model which takes the financial risk index standard after loan facing the trust user group as a constraint condition, takes the profit maximization of loan facing the trust user group as an optimization target, and periodically performs dynamic update based on the user data of the historical users; and loan is carried out to the user based on the obtained loan credit granting result.

In some embodiments, the loan credit model comprises an artificial intelligence model which takes credit score, loan credit line and loan credit rate of a user as model variables, takes the financial risk index standard after loan facing a credit user group as a constraint condition, and maximizes the profit facing the credit user group for loan as an optimization target; the credit score characterizes a probability of a user loan default; the loan credit granting result comprises a loan credit granting amount corresponding to the credit score of the user and a loan credit granting interest rate corresponding to the credit score of the user.

In some embodiments, the loan credit model includes a credit assessment sub-model, a credit line sub-model, and a credit rate sub-model; the credit evaluation submodel is used for calculating the credit score of the user; the credit line sub-model is used for calculating a loan credit line corresponding to the credit score of the user; the credit granting rate sub-model is used for calculating the credit granting rate of the loan corresponding to the credit score of the user; inputting the user data into a loan credit granting model for calculation to obtain a loan credit granting result corresponding to the user, wherein the loan credit granting result comprises: inputting the user data into the credit evaluation submodel to obtain a credit score corresponding to the user; and respectively inputting the credit scores into the credit line sub-model and the credit interest rate sub-model to obtain the credit line and the credit interest rate of loan corresponding to the credit scores of the users.

In some embodiments, the credit evaluation submodel comprises a scorecard model trained based on user data of historical users; the credit line sub-model comprises a mathematical relationship between credit scores of the historical users and loan credit lines of the historical users, which is learned based on the user data of the historical users; the credit granting rate sub-model comprises a mathematical relationship between credit scores of the historical users and loan credit granting rates of the historical users, which is learned based on the user data of the historical users.

In some embodiments, the financial risk indicators include financial disability rates; the financial risk indicators are up to standard, including: the financial reject rate is lower than a preset threshold.

In some embodiments, the optimization objectives include: the credit evaluation sub-model, the credit line sub-model and the credit interest rate sub-model are used for loan of N users obtained based on historical user data, and the per-capita loan yield is maximized; the constraint conditions include: and after N users obtained based on historical user data are loaned by using the credit evaluation submodel, the credit granting amount submodel and the credit granting interest rate submodel, the ratio of the overdue repayment amount to the daily average loan balance is lower than the preset threshold.

In some embodiments, the method further comprises:

adding the user data of the user as the user data of the historical user to an offline user database; the off-line user database is used for storing the user data of the historical users.

The application also provides a user credit granting device, comprising: the acquisition module is used for acquiring user data corresponding to a user to be trusted; the calculation module is used for inputting the user data into a loan and credit granting model for calculation to obtain a loan and credit granting result corresponding to the user; the loan trust model comprises an artificial intelligence model which takes the financial risk index standard after loan facing the trust user group as a constraint condition, takes the profit maximization of loan facing the trust user group as an optimization target, and periodically performs dynamic update based on the user data of the historical users; and the loan module is used for loan to the user based on the obtained loan credit granting result.

In some embodiments, the loan credit model includes a credit assessment sub-model, a credit line sub-model, and a credit rate sub-model; the credit evaluation submodel is used for calculating the credit score of the user; the credit line sub-model is used for calculating a loan credit line corresponding to the credit score of the user; the credit granting rate sub-model is used for calculating the credit granting rate of the loan corresponding to the credit score of the user; the calculation module is specifically configured to: inputting the user data into the credit evaluation submodel to obtain a credit score corresponding to the user; and respectively inputting the credit scores into the credit line sub-model and the credit interest rate sub-model to obtain the credit line and the credit interest rate of loan corresponding to the credit scores of the users.

In some embodiments, the apparatus further comprises: the adding module is used for adding the user data of the user as the user data of the historical user to an offline user database; the off-line user database is used for storing the user data of the historical users.

The present application further proposes an electronic device, comprising: a processor; a memory for storing processor-executable instructions; the processor executes the executable instructions to implement the user credit granting method as shown in any one of the foregoing embodiments.

The present application also proposes a computer-readable storage medium, which stores a computer program for causing a processor to execute the user trust method as shown in any one of the foregoing embodiments.

The technical solutions shown in the foregoing embodiments at least include the following technical effects:

firstly, the loan credit granting model can be updated by using historical user data, so that the income obtained after loan of a credit granting user can be maximized by the loan credit granting model, and the borne financial risk index reaches the standard, and therefore when the loan credit granting model is used for determining the loan credit granting result of the user to be granted and the user to be granted is loaned based on the loan credit granting result, the credit granting mechanism can maximize the income under the condition of controllable financial risk, the long-term income of the mechanism is promoted, and the sustainable development of the mechanism is promoted.

Secondly, the loan credit model can be dynamically updated periodically based on historical user data, and the loan credit model can be updated optimally periodically according to maintained historical data, so that the income of a credit granting organization can be kept optimal all the time.

And thirdly, model optimization updating can be carried out by using the real credit granting data of the historical user, so that model optimization more fitting the historical situation can be carried out by combining the real influence of the historical credit data on the credit granting organization, and the optimized credit granting model fits the real situation better.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Drawings

In order to more clearly illustrate one or more embodiments of the present application or technical solutions in the related art, the drawings needed to be used in the description of the embodiments or the related art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in one or more embodiments of the present application, and other drawings can be obtained by those skilled in the art without inventive exercise.

Fig. 1 is a flowchart of a method of a user credit granting method shown in the present application;

FIG. 2 is a method flow diagram illustrating one method of updating the loan credit model according to the present application;

FIG. 3 is a schematic illustration of a loan granting process according to the present application;

fig. 4 is a schematic structural diagram of a user credit granting device shown in the present application;

fig. 5 is a schematic diagram of a hardware structure of an electronic device shown in the present application.

Detailed Description

Reference will now be made in detail to the exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, like numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of the present application. Rather, they are merely examples of apparatus and methods consistent with certain aspects of one or more embodiments of the application, as detailed in the claims which follow.

It should be noted that: in other embodiments, the steps of the respective methods are not necessarily performed in the order shown and described herein. In some other embodiments, the method may include more or fewer steps than those described herein. Moreover, individual steps described in this application may be broken down into multiple steps for description in other embodiments; multiple steps described in this application may be combined into a single step in other embodiments.

In view of the above, the present application provides a method for granting credit. The method can regularly update the loan credit model by using maintained historical user data, so that the income obtained after loan is given to a credit granting user can be maximized by the loan credit model, and the borne financial risk index reaches the standard, and on one hand, the loan credit model can be regularly updated in an optimized manner, so that the income of a credit granting organization can be constantly kept optimal; on the other hand, when the loan credit granting result of the user is determined by using the loan credit granting model and the user to be credited is loaned based on the loan credit granting result, the credit granting organization maximizes the income under the condition that the financial risk is controllable, thereby improving the long-term income of the organization and promoting the sustainable development of the organization.

The method may be applied to an electronic device corresponding to a credit agency. The electronic device may execute the method by mounting a software device corresponding to the credit granting method. The electronic equipment can be a notebook computer, a server, a mobile phone, a PAD terminal and the like. The specific type of the electronic device is not particularly limited in this application. It will be appreciated that the electronic device may be a client-side or server-side device. The server may be a server or a cloud provided by a server, a server cluster, or a distributed server cluster. The following description will be given taking an execution body as an electronic device (hereinafter simply referred to as a device) as an example.

Referring to fig. 1, fig. 1 is a flowchart illustrating a method for a user trust method according to the present application.

As shown in fig. 1, the method may include S102-S106.

And S102, acquiring user data corresponding to the user to be trusted.

The user to be trusted refers to a user with a trust requirement. In some embodiments, the service promotion platform may direct the user to be trusted to make a trust application.

The user data may include user base data, behavioral data, credit data, and the like. Wherein the underlying data may include age, work, asset, etc. data. The behavioral data may include consumption data. The credit data may include loan, repayment, etc. data. Credit scoring may be performed by the user data.

In some embodiments, a user to be trusted may initiate a trust application to an organization through a user client provided by the organization. And sending user data corresponding to the user to be trusted to the mechanism in response to the user initiating the trust application. The electronic device corresponding to the mechanism can acquire the user data sent by the user client.

S104, inputting the user data into a loan and credit model for calculation to obtain a loan and credit result corresponding to the user;

the loan credit granting model comprises an artificial intelligence model which takes the standard of financial risk indexes subjected to loan release facing a credit granting user group as a constraint condition, takes the maximization of the loan release facing the credit granting user group as an optimization target, and periodically performs dynamic update based on the user data of historical users.

The loan credit granting model can be dynamically updated periodically based on historical user data. The loan credit granting model comprises an artificial intelligence model which takes the standard of financial risk indexes subjected to loan release facing a credit granting user group as a constraint condition, takes the maximization of the loan release facing the credit granting user group as an optimization target, and periodically performs dynamic update based on the user data of historical users.

The financial risk indicators may represent losses that an organization may incur after lending to a user to be credited. The financial risk index meeting the standard can indicate that the loss possibly borne by the organization is lower than a preset value after the user to be credited is loan-placed.

In some embodiments, a dynamic optimization model may be constructed using model parameters included in the loan credit model. The dynamic optimization model may determine, on one hand, the profit obtained after lending the historical user according to the historical user data by using the model parameter, and may determine, on the other hand, the financial risk assumed after lending the historical user according to the historical user data by using the model parameter.

In some embodiments, when the loan credit granting model is updated, historical user data corresponding to the historical user may be obtained, and then the historical user data is input into the dynamic optimization model, so as to obtain revenue obtained after loan for the historical user and financial risk assumed after loan for the historical user. Model parameters of the dynamic optimization model may then be continually adjusted until two predetermined constraints are met, namely, revenue maximization and financial risk being below a predetermined threshold. And then, determining the current model parameters of the dynamic optimization model as the model parameters of the loan credit granting model, and completing the parameter updating aiming at the loan credit granting model. The loan granting model is used for outputting a loan granting result aiming at the user to be granted based on the user data of the user to be granted. In some embodiments, the loan credit model comprises an artificial intelligence model which takes credit score, loan credit line and loan credit rate of a user as model variables, takes the financial risk index standard after loan facing a credit user group as a constraint condition, and maximizes the profit facing the credit user group for loan as an optimization target; the credit score characterizes a probability of a user loan default; the loan credit granting result comprises a loan credit granting amount corresponding to the credit score of the user and a loan credit granting interest rate evaluation sub-model credit granting interest rate sub-model corresponding to the credit score of the user. Therefore, credit score, loan credit line and loan credit rate of the user are adopted as model variables of the loan credit model, and the loan credit accuracy of the user can be improved.

In some embodiments, the loan credit model includes a credit assessment sub-model, a credit line sub-model, and a credit rate sub-model; the credit evaluation submodel is used for calculating the credit score of the user; the credit line sub-model is used for calculating a loan credit line corresponding to the credit score of the user; and the credit granting rate sub-model is used for calculating the credit granting rate of the loan corresponding to the credit score of the user.

In step S104, the user data may be input into the credit evaluation submodel, so as to obtain a credit score corresponding to the user. And then the credit score can be respectively input into the credit line sub-model and the credit interest rate sub-model to obtain a credit line and a credit interest rate of loan corresponding to the credit score of the user.

The credit evaluation submodel may output a user score for a user to be credited based on user data of the user to be credited. The user score may reflect the probability of the user's loan default, i.e., the trustworthiness of the user to be credited. In some embodiments, the credit evaluation submodel may include a mapping of user data to user scores, which may be a linear or non-linear mapping. And according to the user data of the user to be credited, mapping to obtain the user score of the user to be credited. In some embodiments, after the user score is obtained, it may be determined whether the user score reaches a score threshold, and if so, it may be determined that the user to be trusted is a trusted user.

And the credit line sub-model is used for outputting the credit line aiming at the user to be credited based on the user score of the user to be credited. And the credit line granting mechanism grants the loan line of the user. The higher the amount of loan granted, the greater the amount the user may loan.

In some embodiments, the credit limit submodel may include a credit limit table. The table can maintain the corresponding relation between the credit rating and the credit limit of the user. In some embodiments, a user credit rating may be determined according to a user score of a user to be credited, and then the credit rating table may be queried according to the determined user credit rating to determine a credit limit of the user to be credited.

And the credit granting interest rate sub-model is used for outputting the credit granting interest rate aiming at the user to be granted based on the user score of the user to be granted. The credit granting rate indicates the loan rate granted to the user by the institution. The lower the loan interest rate, the lower the amount the user needs to pay.

In some embodiments, the credit interest rate submodel may include a credit interest rate table. The table can maintain the corresponding relation between the user credit rating and the credit interest rate. In some embodiments, a user credit granting level may be determined according to a user score of a user to be granted, and then the credit granting interest rate table may be queried according to the determined user credit granting level to determine a credit granting interest rate of the user to be granted.

And S106, loan is made to the user based on the obtained loan credit granting result. In some embodiments, in executing S106, a loan may be made to the user based on the credit interest rate and credit line of the user to be credited obtained in S104.

In the scheme, on one hand, the loan credit granting model can be updated by using historical user data, so that the income obtained after the loan is given to a credit granting user can be maximized, and the borne financial risk index reaches the standard, and therefore, when the loan credit granting model is used for determining the loan credit granting result of the user to be granted and the user to be granted is loaned based on the loan credit granting result, the income of a credit granting organization can be maximized under the condition that the financial risk is controllable, the long-term income of the organization can be improved, and the sustainable development of the organization can be promoted.

On the other hand, the loan credit model is dynamically updated periodically based on historical user data, so that the loan credit model can be updated optimally and periodically according to the maintained historical data, and the income of a credit institution can be kept optimal all the time.

In some embodiments, the device to which the organization corresponds may further include an offline user database. The off-line user database is used for storing user data of historical users. After the device completes credit granting for the user to be granted, the user data of the user can be used as the user data of the historical user and added to an offline user database for data statistics, and therefore optimization updating of a loan credit granting model by utilizing the historical user data is facilitated.

In some embodiments, the financial risk comprises a financial disability rate. The financial risk meeting the standard comprises that the financial reject ratio is lower than a preset threshold value. The financial failure rate may be used to measure annual losses for the credit agency. Financial risks possibly borne by an organization after loan placement can be accurately represented through the financial reject ratio, and the optimization effect of the loan credit granting model is improved.

In some embodiments, the optimization objectives for the loan credit model may include:

and utilizing the credit evaluation sub-model, the credit granting amount sub-model and the credit granting interest rate sub-model to loan the N users obtained based on the historical user data, so as to maximize the per-capita loan yield.

Constraints on the loan credit model may include:

and after N users obtained based on historical user data are loaned by using the credit evaluation submodel, the credit granting amount submodel and the credit granting interest rate submodel, the ratio of the overdue repayment amount to the daily average loan balance is lower than the preset threshold.

In some embodiments, the user data of the second user newly added within the first preset time period may be generated according to the maintained user data of the first user. A set of users may then be generated to dynamically update the loan credit model based on the first user and the second user.

The first user may be a first user selected from historical users. The user data of the second user may be user data obtained by reasonably estimating (for example, estimating means such as averaging and maximum value) the user data of the first user.

The credit data that has actually occurred may be included in the user data of the first user. The loan and credit model is updated through the data of the users in the user set, model optimization which is more suitable for the historical situation can be made by combining the real influence of the historical credit data on the credit granting organization, and therefore the optimized loan and credit model is more suitable for the real situation.

Referring to fig. 2, fig. 2 is a flowchart illustrating a method for updating a loan credit model according to the present application.

As shown in fig. 2, the method may include:

s202, solving the target credit evaluation sub-model, the target credit line sub-model and the target credit interest rate sub-model so as to meet the condition A and the condition B.

And S204, determining the target credit evaluation submodel, the target credit line submodel and the target credit interest rate submodel as the credit evaluation submodel, the credit line submodel and the credit interest rate included in the loan credit model.

Wherein condition a comprises: and utilizing the credit evaluation sub-model, the credit granting amount sub-model and the credit granting interest rate sub-model to loan the N users obtained based on the historical user data, so as to maximize the per-capita loan yield.

The condition B includes: and after N users obtained based on historical user data are loaned by using the credit evaluation submodel, the credit granting amount submodel and the credit granting interest rate submodel, the ratio of the overdue repayment amount to the daily average loan balance is lower than the preset threshold.

In some embodiments, in order to solve the target credit evaluation submodel, the target credit line submodel and the target credit interest rate submodel, an expression may be set for the condition a and the condition B.

Wherein, the expression of the condition a may include:

wherein the content of the first and second substances,

representing the number of users in the set of users,

representing each user in the set of users.

Representing the target credit rating submodel that needs to be solved,

representing a target credit line submodel to be solved,

and representing the target credit granting interest rate submodel to be solved.

Representing the determination of a user using the target credit evaluation submodel

Whether it is a trusted user. In some embodiments, the target credit evaluation submodel may include a mapping function and a scoring threshold. Users can be identified by mapping function

Is mapped to a user score. If the obtained user score reaches a score threshold, the user can be determined

Is a trusted user, and then the user can use the method

Is set to 1. If the obtained user score does not reach the score threshold value, the user can be determined

For non-trusted users, the user can be connected with the network at the moment

Is set to 0. According to the user

The user rating of (2) may also be determined.

Indicating the user determined by the target credit evaluation submodel and the target credit line submodel

The credit line. In some embodiments, the target credit limit submodel may include a target credit limit table. The table maintains the corresponding relationship between the user level and the credit line. The user can be determined through the target credit evaluation submodel

According to the user grade, the user can be determined

The credit line. The organization can give the user the credit limit

And (7) loan.

Representing the user-specific determined using the target credit evaluation submodel and the target credit interest rate submodel

The credit and interest rate of. In some embodiments, the target credit interest rate submodel may include a target credit interest rate table. The table maintains the corresponding relationship between the user level and the credit and interest rate. The user can be determined through the target credit evaluation submodel

According to the user grade, the user can be determined

The credit and interest rate of. The organization can give credit to the user according to the credit interest rate

And (7) loan.

Representing the user determined by using the data representing the loan limit usage rate in the user data of the first user

Credit usage of. The first user is a user who really performs credit behavior, and the corresponding user data can comprise data representing the credit amount usage rate of the first user. In some embodiments, the highest, lowest, or average loan amount usage among the first users may be selected as the one

. In some embodiments, a neural network may also be employed to predict the loan amount usage based on the user data of the first user.

By passing

The income that the institution can obtain after lending each user in the user set can be obtained. The income is divided by the number n of people to obtain the per-capita income which can be obtained by the institution after each user in the user set is put in credit.

The expression of condition B may include:

first, the same variables in the expression of the condition B and the expression of the condition a represent the same meanings, and are not described in detail here. Secondly, in order to improve the optimization effect of the loan credit granting model, the condition B can be satisfied in any month with k being 1-12.

Wherein the content of the first and second substances,

and the preset threshold value is used. This value can be set according to the business requirements. If the financial risk that the institution is required to undertake is low, a smaller value is selected as

If the higher financial risk that the organization can undertake, a larger value is selected as

。

Representing the number of second users newly added in a month.

Representing each of the newly added second users.

Is shown as

And (4) month.

Indicating a second determination using data characterizing an amount of overdue among the user data of the first user

Historical overdue amounts of the month; the first user is a user who really takes credit action, and the corresponding user data can comprise data which represents overdue amount. In some embodiments, the highest, lowest or average amount of overdue among the first users may be selected as the

. In some embodiments, a neural network may also be employed to predict the amount of overdue based on the user data of the first user.

Whether it is a overdue user. In some embodiments, the credit evaluation submodel further includes an overdue score, and if the user score of the user reaches the overdue score, the user may be determined to be an overdue user, in which case the user may be determined to be an overdue user

Setting as 1; if the user score of the user reaches the timeoutA term score may determine that the user is not a overdue user, at which point the user may be scored

Is set to 0.

Show to proceed

The new overdue amount added after the loan of a month. It should be noted that two reasonable assumptions are made here. First, the overdue user will make the loan overdue within the maximum credit line given by the institution. For example, if the organization grants credit based on user j

If the user j is judged as the overdue user, the corresponding overdue amount

. Secondly, in the case where the number of users is sufficiently large, the number of users newly added per month and the composition are approximately the same, that is, the amount of overdue money newly generated per month is the same. Thereby the device is provided with

Then can proceed

The new overdue amount added after the loan of a month.

Data representing the average daily loan balance in the user data of the first user, and the second

The historical daily average of the month loan balance. The first user is realThe user who takes credit action may include data characterizing his/her average daily loan balance in the corresponding user data. In some embodiments, the highest, lowest, or average daily loan balance among the first users may be selected as the

. In some embodiments, a neural network may also be employed to predict the average daily loan balance based on the user data of the first user.

Show to proceed

The daily average loan balance after the month loan.

The average daily loan balance is the amount of loan released by the institution. For example, if user j is a trusted user, the amount of loan issued by the institution to user j is

. In addition, the assumption that the number of users and the composition of components newly added each month are approximately the same in the case where the number of users is sufficiently large is also utilized here. That is to say that the first and second electrodes,

can express to proceed

The daily average loan balance after the month loan.

By passing

The overdue amount which is possibly borne by the institution after the k months is credited can be obtained by adding the historical overdue amount actually generated in the k month with the newly added overdue amount.

By passing

The daily average loan balance k months after the institution loan is put can be obtained by adding the historical daily average loan balance actually occurring in the kth month and the newly added daily average loan balance.

And dividing the overdue amount possibly borne by the institution after the institution credits for k months by the daily average loan balance, so as to obtain the financial reject ratio possibly borne by the institution after the institution credits for k months.

In some embodiments, in step S202, the target credit evaluation sub-model, the target credit line sub-model and the target credit interest rate sub-model may be solved through a preset intelligent optimization algorithm. The intelligent optimization algorithm may include a genetic algorithm, a particle swarm algorithm, an ant colony algorithm, and the like. The intelligent optimization algorithm is not particularly limited in this application.

After the solution is completed, S204 may be executed, and the solved target credit evaluation submodel, target credit line submodel and target credit interest rate submodel are determined as the credit evaluation submodel, credit line submodel and credit interest rate included in the loan credit model.

In the solution disclosed in the foregoing embodiment, on one hand, by optimizing and updating the credit evaluation sub-model, the credit granting amount sub-model and the credit granting rate sub-model included in the loan granting model with the condition a and the condition B as optimization targets, the profit obtained after each user in the user set is loaned by the optimized loan granting model can be maximized, and the borne financial failure rate is lower than the threshold value, so that when the loan granting result of the user to be granted is determined by using the loan granting model, and the user to be granted is loaned based on the loan granting result, the profit of the granting organization can be maximized under the condition that the financial risk is controllable, thereby improving the long-term profit of the organization and promoting the sustainable development of the organization.

On the other hand, model optimization updating can be performed by using the real credit granting data of the first user (historical user), so that model optimization more fitting the historical situation can be performed by combining the real influence of the historical credit data on the credit granting organization, and the optimized credit granting model fits the real situation better.

The following embodiments are described in connection with a trust scenario.

Referring to fig. 3, fig. 3 is a schematic diagram illustrating a loan granting process according to the present application.

The trust system (hereinafter, referred to as a system) shown in fig. 3 may include a system developed by a trust authority for granting a user trust. As shown in fig. 3, the credit granting system may include an online system and an offline system. The credit granting system can grant the user credit through the computing power provided by the electronic equipment.

The online system can credit the user through a loan credit model. The loan and trust model can comprise a trust condition, a credit evaluation sub-model, a credit limit sub-model and a credit interest rate sub-model. The credit limit submodel can comprise a credit limit table; the credit interest rate submodel may include a credit interest rate table. And the credit granting condition is used for preliminarily filtering the users which do not meet the credit granting condition.

The offline system can maintain historical user data and regularly optimize and update the credit evaluation sub-model, the credit granting amount table and the credit granting interest rate table, so that the loan credit granting model is regularly optimized and updated.

As shown in fig. 3, the system may execute S31, and the online system obtains the user data of the user to be trusted through the drainage platform. The drainage platform may be a platform including the development of a trust authority or a third party authority. The user to be trusted can comprise a user who initiates a trust application through a link provided by the drainage platform. And responding to the user to be trusted to initiate a credit granting application, and synchronously authorizing to send user data to an online system. The user data may include basic data such as height and occupation of the user to be trusted, behavior data such as consumption, historical credit data, and the like.

And then the system can execute S32 to determine whether the user to be credited meets the preset crediting condition, if so, the system can further determine the user score of the user to be credited according to the user data of the user to be credited by utilizing the credit evaluation sub-model, and when the user score reaches the score threshold value, the user to be credited is determined to be a creditable user.

And then S33 can be executed, the credit rating of the user to be credited is respectively input into the credit line submodel and the credit interest rate submodel, the credit line and the credit interest rate of the user to be credited are determined by utilizing the credit rate table and the credit interest rate table, and the user to be credited is subjected to loan and repayment tracking according to the determined credit line and the determined credit interest rate.

The system may further execute S34 to track and count the behavior of the user to be trusted, such as loan, repayment, etc., and maintain the user data through the online database.

The system may also execute S35 to periodically import the user data maintained in the online database into the offline database.

The system comprises an offline system which can generate user data of a second user according to the user data of a first user in an offline database, and generate a user set according to the first user and the second user.

The system may perform S36 for credit usage prediction, historical expected credit prediction, and historical loan daily average balance prediction based on the user data of the first user.

And then S37 can be executed, according to the expressions respectively corresponding to the condition A and the condition B maintained in the system, the credit evaluation sub-model, the credit granting amount table and the credit granting rate are solved through an optimization solving algorithm, and the credit evaluation sub-model, the credit granting amount table and the credit granting rate included in the loan credit granting model are optimized and updated regularly.

Therefore, firstly, by taking the condition A and the condition B as optimization targets, the credit assessment submodel, the credit granting amount submodel and the credit granting rate submodel which are included in the loan credit granting model are optimized and updated, the income obtained after each user in the user set is loan-released by the optimized loan credit granting model can be maximized, and the borne financial failure rate is lower than a threshold value, so that when the loan credit granting model is used for determining the loan credit granting result of the user to be granted, and the user to be granted is loaned based on the loan credit granting result, the income of the credit granting organization can be maximized under the condition that the financial risk is controllable, the long-term income of the organization is improved, and the sustainable development of the organization is promoted.

Secondly, model optimization updating can be carried out by utilizing the real credit granting data of the first user (historical user), so that model optimization more fitting the historical situation can be carried out by combining the real influence of the historical credit data on the credit granting organization, and the optimized credit granting model fits the real situation better.

Thirdly, the loan credit model is dynamically updated periodically based on historical user data, so that the loan credit model can be updated optimally periodically according to maintained historical data, and the income of a credit granting organization can be kept optimal all the time.

Corresponding to the foregoing embodiments, the present application further provides a user credit granting device 40.

Referring to fig. 4, fig. 4 is a schematic structural diagram of a user credit granting device shown in the present application.

As shown in fig. 4, the apparatus 40 may include:

the obtaining module 41 obtains user data corresponding to a user to be trusted;

the calculation module 42 is used for inputting the user data into a loan and credit granting model for calculation to obtain a loan and credit granting result corresponding to the user; the loan trust model comprises an artificial intelligence model which takes the financial risk index standard after loan facing the trust user group as a constraint condition, takes the profit maximization of loan facing the trust user group as an optimization target, and periodically performs dynamic update based on the user data of the historical users;

and a loan module 43 for lending to the user based on the obtained loan granting result.

In some embodiments, the loan credit model includes a credit assessment sub-model, a credit line sub-model, and a credit rate sub-model; the credit evaluation submodel is used for calculating the credit score of the user; the credit line sub-model is used for calculating a loan credit line corresponding to the credit score of the user; the credit granting rate sub-model is used for calculating the credit granting rate of the loan corresponding to the credit score of the user;

the calculating module 42 is specifically configured to:

inputting the user data into the credit evaluation submodel to obtain a credit score corresponding to the user;

and respectively inputting the credit scores into the credit line sub-model and the credit interest rate sub-model to obtain the credit line and the credit interest rate of loan corresponding to the credit scores of the users.

In some embodiments, the optimization objectives include:

the credit evaluation sub-model, the credit line sub-model and the credit interest rate sub-model are used for loan of N users obtained based on historical user data, and the per-capita loan yield is maximized;

the constraint conditions include:

In some embodiments, the apparatus 40 further comprises:

an adding module 44, configured to add the user data of the user to an offline user database as the user data of the historical user; the off-line user database is used for storing the user data of the historical users.

The embodiment of the user credit granting device shown in the application can be applied to electronic equipment. Accordingly, the present application discloses an electronic device, which may comprise: a processor.

A memory for storing processor-executable instructions.

The processor is configured to call the executable instructions stored in the memory to implement the user credit granting method shown in any one of the foregoing embodiments.

Referring to fig. 5, fig. 5 is a schematic diagram of a hardware structure of an electronic device shown in the present application.

As shown in fig. 5, the electronic device may include a processor for executing instructions, a network interface for making network connections, a memory for storing operating data for the processor, and a non-volatile memory for storing instructions corresponding to the user trusted device.

The embodiments of the apparatus may be implemented by software, or by hardware, or by a combination of hardware and software. Taking a software implementation as an example, as a logical device, the device is formed by reading, by a processor of the electronic device where the device is located, a corresponding computer program instruction in the nonvolatile memory into the memory for operation. In terms of hardware, in addition to the processor, the memory, the network interface, and the nonvolatile memory shown in fig. 5, the electronic device in which the apparatus is located in the embodiment may also include other hardware according to an actual function of the electronic device, which is not described again.

It is understood that, in order to increase the processing speed, the user instruction corresponding to the apparatus may also be directly stored in the memory, which is not limited herein.

The present application proposes a computer-readable storage medium, which stores a computer program, which can be used to cause a processor to execute a user trust method as shown in any of the foregoing embodiments.

One skilled in the art will recognize that one or more embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, one or more embodiments of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, one or more embodiments of the present application may take the form of a computer program product embodied on one or more computer-usable storage media (which may include, but are not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

"and/or" as recited herein means having at least one of two, for example, "a and/or B" includes three scenarios: A. b, and "A and B".

The embodiments in the present application are described in a progressive manner, and the same and similar parts among the embodiments can be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the data processing apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to part of the description of the method embodiment.

Specific embodiments of the present application have been described. Other embodiments are within the scope of the following claims. In some cases, the acts or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

Embodiments of the subject matter and functional operations described in this application may be implemented in the following: digital electronic circuitry, tangibly embodied computer software or firmware, computer hardware including the structures disclosed in this application and their structural equivalents, or a combination of one or more of them. Embodiments of the subject matter described in this application can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions, encoded on a tangible, non-transitory program carrier for execution by, or to control the operation of, data processing apparatus. Alternatively or additionally, the program instructions may be encoded on an artificially generated propagated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal, that is generated to encode and transmit information to suitable receiver apparatus for execution by the data processing apparatus. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or a combination of one or more of them.

The processes and logic flows described in this application can be performed by one or more programmable computers executing one or more computer programs to perform corresponding functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application-specific integrated circuit).

Computers suitable for executing computer programs include, for example, general and/or special purpose microprocessors, or any other type of central processing system. Generally, a central processing system will receive instructions and data from a read-only memory and/or a random access memory. The essential components of a computer include a central processing system for implementing or executing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto-optical disks, or optical disks. However, a computer does not necessarily have such a device. Moreover, a computer may be embedded in another device, e.g., a mobile telephone, a Personal Digital Assistant (PDA), a mobile audio or video player, a game console, a Global Positioning System (GPS) receiver, or a portable storage device such as a Universal Serial Bus (USB) flash drive, to name a few.

Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including by way of example semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., an internal hard disk or a removable disk), magneto-optical disks, and 0xCD _00 ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

Although this application contains many specific implementation details, these should not be construed as limiting the scope of any disclosure or of what may be claimed, but rather as merely describing features of particular disclosed embodiments. Certain features that are described in this application in the context of separate embodiments can also be implemented in combination in a single embodiment. In other instances, features described in connection with one embodiment may be implemented as discrete components or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination.

Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In some cases, multitasking and parallel processing may be advantageous. Moreover, the separation of various system modules and components in the described embodiments is not to be understood as requiring such separation in all embodiments, and it is to be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

Thus, particular embodiments of the subject matter have been described. Other embodiments are within the scope of the following claims. In some cases, the actions recited in the claims can be performed in a different order and still achieve desirable results. Further, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some implementations, multitasking and parallel processing may be advantageous.

The above description is only for the purpose of illustrating the preferred embodiments of the present application and is not intended to limit the present application to the particular embodiments of the present application, and any modifications, equivalents, improvements, etc. made within the spirit and principles of the present application should be included within the scope of the present application.

Claims

1. A user credit granting method comprises the following steps:

acquiring user data corresponding to a user to be trusted;

inputting the user data into a loan credit granting model for calculation to obtain a loan credit granting result corresponding to the user; the loan trust model comprises an artificial intelligence model which takes the financial risk index standard after loan facing the trust user group as a constraint condition, takes the profit maximization of loan facing the trust user group as an optimization target, and periodically performs dynamic update based on the user data of the historical users;

and loan is carried out to the user based on the obtained loan credit granting result.

2. The method of claim 1, wherein the loan credit model comprises an artificial intelligence model which takes credit score, loan credit line and loan credit interest rate of a user as model variables, takes the financial risk index reaching the standard after loan release facing a credit user group as a constraint condition, and takes the income maximization facing the credit user group as an optimization target; the credit score characterizes a probability of a user loan default; the loan credit granting result comprises a loan credit granting amount corresponding to the credit score of the user and a loan credit granting interest rate corresponding to the credit score of the user.

3. The method of claim 2, wherein the loan credit model comprises a credit assessment submodel, a credit line submodel, and a credit interest rate submodel; the credit evaluation submodel is used for calculating the credit score of the user; the credit line sub-model is used for calculating a loan credit line corresponding to the credit score of the user; the credit granting rate sub-model is used for calculating the credit granting rate of the loan corresponding to the credit score of the user;

inputting the user data into a loan credit granting model for calculation to obtain a loan credit granting result corresponding to the user, wherein the loan credit granting result comprises:

4. The method of claim 3, the credit evaluation submodel comprising a scorecard model trained based on user data of historical users; the credit line sub-model comprises a mathematical relationship between credit scores of the historical users and loan credit lines of the historical users, which is learned based on the user data of the historical users; the credit granting rate sub-model comprises a mathematical relationship between credit scores of the historical users and loan credit granting rates of the historical users, which is learned based on the user data of the historical users.

5. The method of claim 3, the financial risk indicators comprising financial disability rates; the financial risk indicators are up to standard, including: the financial reject rate is lower than a preset threshold.

6. The method of claim 5, the optimization objective comprising:

the constraint conditions include:

7. The method of claim 1, further comprising:

8. A user trust apparatus, comprising:

the acquisition module is used for acquiring user data corresponding to a user to be trusted;

the calculation module is used for inputting the user data into a loan and credit granting model for calculation to obtain a loan and credit granting result corresponding to the user; the loan trust model comprises an artificial intelligence model which takes the financial risk index standard after loan facing the trust user group as a constraint condition, takes the profit maximization of loan facing the trust user group as an optimization target, and periodically performs dynamic update based on the user data of the historical users;

and the loan module is used for loan to the user based on the obtained loan credit granting result.

9. The device of claim 8, wherein the loan credit model comprises an artificial intelligence model which takes credit score, loan credit line and loan credit interest rate of a user as model variables, takes the financial risk index reaching the standard after loan release facing a credit user group as a constraint condition, and takes the income maximization facing the credit user group as an optimization target; the credit score characterizes a probability of a user loan default; the loan credit granting result comprises a loan credit granting amount corresponding to the credit score of the user and a loan credit granting interest rate corresponding to the credit score of the user.

10. The apparatus of claim 9, wherein the loan credit model comprises a credit assessment sub-model, a credit line sub-model and a credit interest rate sub-model; the credit evaluation submodel is used for calculating the credit score of the user; the credit line sub-model is used for calculating a loan credit line corresponding to the credit score of the user; the credit granting rate sub-model is used for calculating the credit granting rate of the loan corresponding to the credit score of the user;

the calculation module is specifically configured to:

11. The apparatus of claim 10, the credit evaluation submodel comprising a scorecard model trained based on user data of historical users; the credit line sub-model comprises a mathematical relationship between credit scores of the historical users and loan credit lines of the historical users, which is learned based on the user data of the historical users; the credit granting rate sub-model comprises a mathematical relationship between credit scores of the historical users and loan credit granting rates of the historical users, which is learned based on the user data of the historical users.

12. The apparatus of claim 10, the financial risk indicator comprising a financial disability rate; the financial risk indicators are up to standard, including: the financial reject rate is lower than a preset threshold.

13. The apparatus of claim 12, the optimization objective comprising:

the constraint conditions include:

14. The apparatus of claim 8, further comprising:

the adding module is used for adding the user data of the user as the user data of the historical user to an offline user database; the off-line user database is used for storing the user data of the historical users.

15. An electronic device, comprising:

a processor;

a memory for storing processor-executable instructions;

wherein the processor executes the executable instructions to implement the user credit granting method according to any one of claims 1 to 7.

16. A computer-readable storage medium, which stores a computer program for causing a processor to execute the user granting method according to any one of claims 1 to 7.