A risk prediction method, device and equipment for bank corporate customers

Abstract

This article belongs to the field of bank risk control technology, and particularly relates to a risk prediction method, device, and apparatus for a bank's corporate customers, wherein the method comprises: obtaining first account information and first transaction information of the corporate customer; inputting the first account information and first transaction information of the corporate customer into a pre-trained profit risk prediction model to obtain a positive profit probability score for the corporate customer; obtaining second account information and second transaction information of the corporate customer; inputting the second account information, second transaction information, and the positive profit probability score of the corporate customer into a pre-trained churn risk prediction model to obtain a churn probability score for the corporate customer; and when the churn probability score is higher than a first preset threshold, determining that the corporate customer is at risk of churn. This article addresses the problem of how to further improve the accuracy of predicting churn risk for corporate customers.

Description

Technical Field

[0001] The present invention relates to the technical field of bank risk control, and in particular to a method, device and equipment for predicting risks for corporate customers of a bank. Background Art

[0002] For corporate customer risk management, existing technologies mainly monitor corporate customer transaction data, activity levels, capital flows, etc. to predict the risk of corporate customer loss, so that relevant measures can be taken in a timely manner when corporate customers show a tendency to lose, to ensure the stability of corporate customers.

[0003] However, there are also transactions between corporate customers and their respective users. The accounts, number of users, and user status of each user under the corporate customer all affect the corporate customer's revenue, and further affect the risk of corporate customer churn. Therefore, the existing technology lacks consideration of the impact of each user under the corporate customer, and thus the accuracy of predicting the risk of corporate customer churn needs to be further improved. Summary of the Invention

[0004] The present invention provides a method, device and equipment for predicting the risk of corporate customers of a bank, which are used to solve the problem of how to further improve the accuracy of predicting the risk of corporate customer churn.

[0005] In order to solve the above technical problems, a first aspect of this document provides a risk prediction method for corporate customers of a bank, the method comprising:

[0006] Obtaining first account information and first transaction information of a corporate customer, wherein the first account information includes the ledger data of each user of the corporate customer, the number of users, the number of products held, and the number of active users; and the first transaction information includes the transaction amount, the number of transactions, and the transaction frequency of the corporate customer;

[0007] Inputting the first account information and the first transaction information of the corporate customer into a pre-trained profit-risk prediction model to obtain a positive profit probability score for the corporate customer;

[0008] Obtaining second account information and second transaction information of the corporate customer, where the second account information includes the number of products held by the corporate customer, a satisfaction survey score, and complaint frequency, and the second transaction information includes the transaction amount, number of transactions, transaction frequency, and knowledge graph feature value of the corporate customer;

[0009] Inputting the second account information, second transaction information, and the positive profit probability score of the corporate customer into a pre-trained churn risk prediction model to obtain the churn probability score of the corporate customer;

[0010] When the churn probability score is higher than a first preset threshold, it is determined that the corporate customer has a churn risk.

[0011] Furthermore, the first transaction information also includes the interest rate allocated by the corporate customer to each user and the user's early interest settlement rate;

[0012] The interest ratio represents the ratio of the deposit interest allocated by the corporate client to each user;

[0013] The user early interest settlement rate refers to the proportion of users who initiated refunds and interest settlements in advance before the interest settlement date to the total number of users under the corporate customer.

[0014] Furthermore, the generation process of the return-risk prediction model includes:

[0015] Obtaining first account information, first transaction information, and income information of each historical corporate customer, wherein the income information includes income data of each historical corporate customer within a predetermined period;

[0016] determining a first data set based on the first account information of each historical corporate customer and the first transaction information of each historical corporate customer;

[0017] Determining label data of the first data set based on the historical revenue information of each corporate customer;

[0018] A first machine learning model is trained based on the first data set and the label data of the first data set to obtain a return-risk prediction model.

[0019] Furthermore, the second transaction information also includes the amount transferred by the corporate customer to the corporate customer's external institution account and the frequency of the corporate customer transferring money to the corporate customer's external institution account.

[0020] Furthermore, the generation process of the churn risk prediction model includes:

[0021] Acquire the second account information, second transaction information, and churn information of each historical corporate customer, wherein the churn information includes identification data indicating whether each historical corporate customer has churned;

[0022] determining a second data set based on the second account information of each historical corporate customer and the second transaction information of each historical corporate customer;

[0023] determining label data of the second data set based on the historical churn information of each corporate customer;

[0024] A second machine learning model is trained based on the second data set and the label data of the second data set to obtain a churn risk prediction model.

[0025] Furthermore, when the churn probability score is higher than a first preset threshold, it is determined that the corporate customer has a churn risk, further comprising:

[0026] Obtaining the satisfaction survey score, complaint frequency, knowledge graph feature value, and positive profit probability score of the corporate customers;

[0027] When the churn probability score is higher than the first preset threshold, and satisfies at least one of the following conditions: the satisfaction survey score is lower than the second preset threshold, the complaint frequency is higher than the third preset threshold, the knowledge graph feature value is lower than the fourth preset threshold, and the positive benefit probability score is lower than the fifth preset threshold, it is determined that there is a risk of churn for the corporate customer.

[0028] Furthermore, the risk prediction method for corporate bank customers also includes:

[0029] Determining whether the positive return probability score is lower than a sixth preset threshold;

[0030] If the judgment is yes, it is determined that the corporate customer has a profit risk.

[0031] A second aspect of the present invention provides a risk prediction device for a bank's corporate customers, the device comprising:

[0032] A first acquisition module is configured to acquire first account information and first transaction information of a corporate customer, wherein the first account information includes the ledger data of each user of the corporate customer, the number of users, the number of products held, and the number of active users; and the first transaction information includes the transaction amount, the number of transactions, and the transaction frequency of the corporate customer;

[0033] a first input module, configured to input the first account information and first transaction information of the corporate customer into a pre-trained profit-risk prediction model to obtain a positive profit probability score for the corporate customer;

[0034] a second acquisition module configured to acquire second account information and second transaction information of the corporate customer, wherein the second account information includes the number of products held by the corporate customer, a satisfaction survey score, and complaint frequency, and the second transaction information includes the transaction amount, number of transactions, transaction frequency, and knowledge graph feature value of the corporate customer;

[0035] a second input module, configured to input the second account information, second transaction information, and the positive profit probability score of the corporate customer into a pre-trained churn risk prediction model to obtain a churn probability score of the corporate customer;

[0036] The risk determination module is configured to determine that the corporate customer has a churn risk when the churn probability score is higher than a first preset threshold.

[0037] The third aspect of this document provides a computer device comprising a memory, a processor, and a computer program stored on the memory, wherein when the computer program is run by the processor, the computer program executes the instructions of the risk prediction method for corporate customers of a bank as described in any of the aforementioned embodiments.

[0038] A fourth aspect of this document provides a computer storage medium having a computer program stored thereon, which, when executed by a processor of a computer device, executes instructions of the risk prediction method for corporate customers of a bank as described in any of the aforementioned embodiments.

[0039] A fifth aspect of this document provides a computer program product, which includes a computer program. When the computer program is run by a processor of a computer device, it executes instructions of the risk prediction method for corporate customers of a bank as described in any of the aforementioned embodiments.

[0040] The risk prediction method, device and equipment for corporate customers of a bank provided in this article obtain a profit risk prediction model through the account information, transaction information and user information of the corporate customers. The profit risk prediction model can predict the positive profit probability score of the corporate customers and provide a data reference for predicting the risk of corporate customer churn. The churn risk prediction model is then obtained by combining the account information, transaction information and positive profit probability score of the corporate customers, thereby predicting the risk of corporate customer churn in a more comprehensive and accurate manner.

[0041] In order to make the above and other purposes, features and advantages of this article more obvious and easy to understand, the following specifically cites preferred embodiments and provides detailed descriptions in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of this article or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of this article. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0043] Figure 1 A flow chart of a risk prediction method for corporate customers of a bank according to an embodiment of the present invention is shown;

[0044] Figure 2 The following is a flow chart showing the generation process of the return-risk prediction model according to the embodiment of this article;

[0045] Figure 3 The following is a flow chart showing the generation process of the churn risk prediction model according to the embodiment of this article;

[0046] Figure 4A flow chart of determining the risk of corporate customer churn according to an embodiment of this invention is shown;

[0047] Figure 5 A flow chart for determining the risk of corporate customer returns according to an embodiment of this invention is shown;

[0048] Figure 6 It shows the structure of the risk prediction device for corporate customers of a bank according to the embodiment of this invention;

[0049] Figure 7 The diagram shows the structure of the computer device according to the embodiment of this article.

[0050] Description of the accompanying symbols:

[0051] 610. First acquisition module;

[0052] 620, first input module;

[0053] 630, second acquisition module;

[0054] 640, second input module;

[0055] 650, Risk Determination Module;

[0056] 702. Computer equipment;

[0057] 704, processor;

[0058] 706. Memory;

[0059] 708, driving mechanism;

[0060] 710, input / output module;

[0061] 712. Input devices;

[0062] 714. Output device;

[0063] 716. Presentation equipment;

[0064] 718. Graphical User Interface;

[0065] 720, network interface;

[0066] 722, communication link;

[0067] 724. Communication bus. DETAILED DESCRIPTION

[0068] The following will be combined with the accompanying drawings to clearly and completely describe the technical solutions in the embodiments of this document. Obviously, the embodiments described are only part of the embodiments of this document, not all of the embodiments. Based on the embodiments of this document, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this document.

[0069] It should be noted that the terms "first," "second," and the like in the specification and claims herein and in the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, product, or device comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or devices.

[0070] This specification provides method operation steps as described in the embodiments or flowcharts, but more or fewer operation steps may be included based on routine or non-creative work. The order of steps listed in the embodiments is only one way of executing the steps among many orderings and does not represent the only execution order. When a system or device product is actually executed, the method can be executed in the order shown in the embodiments or the drawings or in parallel.

[0071] It should be noted that the acquisition, storage, use, and processing of data in the technical solutions of the embodiments of this specification comply with the relevant provisions of national laws and regulations.

[0072] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned. They should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0073] In one embodiment of the present invention, a risk prediction method for corporate customers of a bank is provided to solve the problem of how to further improve the accuracy of predicting the risk of corporate customer churn.

[0074] Specifically, such as Figure 1 As shown in Figure 2, the risk prediction methods for corporate customers of banks include:

[0075] Step 110: Obtain first account information and first transaction information of the corporate customer;

[0076] In step 110, the first account information includes the user ledger data, number of users, number of products held, and number of active users of the corporate customer; the first transaction information includes the corporate customer's transaction amount, number of transactions, and transaction frequency;

[0077] Step 120: Input the first account information and first transaction information of the corporate customer into a pre-trained profit-risk prediction model to obtain a positive profit probability score for the corporate customer;

[0078] Step 130: Obtain the second account information and second transaction information of the corporate customer;

[0079] In step 130, the second account information includes the number of products held by the corporate customer, satisfaction survey score, and complaint frequency, and the second transaction information includes the corporate customer's transaction amount, number of transactions, transaction frequency, and knowledge graph feature value;

[0080] Step 140: Input the second account information, second transaction information, and positive profit probability score of the corporate customer into a pre-trained churn risk prediction model to obtain a churn probability score for the corporate customer;

[0081] Step 150: When the churn probability score is higher than a first preset threshold, it is determined that the corporate customer has a churn risk.

[0082] This embodiment fully considers the account information, transaction information and user information of corporate customers. Through the account information, transaction information and user information of corporate customers, a profit risk prediction model is obtained. The profit risk prediction model can predict the positive profit probability score of corporate customers and provide data reference for the prediction of corporate customer churn risk. Combined with the account information, transaction information and positive profit probability score of corporate customers, a churn risk prediction model is obtained, thereby predicting the churn risk of corporate customers more comprehensively and accurately.

[0083] In some embodiments herein, a ledger is maintained within the corporate client's unified management account to enable real-time monitoring of each user's account information and transaction information. This ledger data includes funds in user accounts, records of fund transfers in and out, and records of interest earned on the corporate client's account.

[0084] In some embodiments of this document, at every scheduled time interval, the first account information and first transaction information of the corporate customer within the scheduled time period are obtained, and the income risk prediction model is input for prediction to obtain a positive income probability score for the corporate customer. The scheduled time can be daily, weekly, monthly, etc., and the scheduled period can be one consecutive week, two consecutive weeks, three consecutive weeks, one consecutive month, two consecutive months, etc., which is not limited in this document.

[0085] In some embodiments of this document, at every predetermined time interval, the second account information and second transaction information of the corporate customer within the predetermined time period are obtained, and the positive profit probability score is combined to input into the churn risk prediction model for prediction to obtain the corporate customer churn probability score. When the churn probability score is higher than the first preset threshold, it means that there is a churn risk for the corporate customer. The predetermined time can be daily, weekly, monthly, etc., and the predetermined period can be one consecutive week, two consecutive weeks, three consecutive weeks, one consecutive month, two consecutive months, etc., which is not limited in this document, but should be consistent with the predetermined time and predetermined period corresponding to the input positive profit probability score. The first preset threshold is set according to actual conditions and is not limited in this document.

[0086] In some embodiments of this document, satisfaction survey scores of corporate customers are collected at specific intervals, and the frequency of complaints from corporate customers within a predetermined period is counted. The satisfaction survey scores at the time of model input are based on the latest collected scores, and the complaint frequency is based on the complaint frequency within the predetermined period. The specific interval time of the specific intervals is not limited in this document.

[0087] In some embodiments of this invention, a relationship graph between customers is established within the bank to obtain the knowledge graph feature values ​​of each customer. The knowledge graph feature values ​​include in-degree, out-degree, betweenness centrality, closeness centrality, etc. When inputting the churn risk prediction model, the knowledge graph feature values ​​of corporate customers are also input.

[0088] In some embodiments herein, the first transaction information also includes the interest ratio allocated by the corporate client to each user and the user's early interest settlement rate; the interest ratio indicates the ratio of the deposited interest allocated by the corporate client to each user; the user's early interest settlement rate indicates the proportion of users who initiated a refund and settled the interest in advance before the interest settlement date to the total number of users under the corporate client. The corporate client will consider signing a contract with the user and allocating part of the profits from the funds stored by the user in its account to the user in order to maintain user stickiness; the user will also request early interest settlement, request a refund, and require the corporate client to fulfill the contract and pay the relevant profits. Therefore, the interest ratio allocated by the corporate client to each user and the user's early interest settlement rate can be related to the corporate client's profits to a certain extent.

[0089] In some embodiments herein, the second transaction information also includes the amount transferred by the corporate customer to the corporate customer's account at an external institution, and the frequency of transfers from the corporate customer to the corporate customer's account at the external institution. An external institution refers to a financial institution other than the bank's own. When a corporate customer transfers funds to an account at an external institution, there are generally two scenarios: one in which the corporate customer transfers funds to its own account at the external institution, and the other in which the corporate customer transacts with other customers of the external institution. The former indicates a potential churn risk for the corporate customer. Therefore, the transaction amount and frequency of the former are also considered factors and input into the churn risk prediction model for prediction.

[0090] In one embodiment of this invention, Figure 2 As shown, the generation process of the return-risk prediction model includes:

[0091] Step 210: Obtain the first account information, first transaction information, and income information of each historical corporate customer;

[0092] In step 210, the revenue information includes the revenue data of each historical corporate customer within a predetermined period;

[0093] Step 220: determining a first data set based on the first account information of each historical corporate customer and the first transaction information of each historical corporate customer;

[0094] Step 230: determining label data of the first data set based on the historical revenue information of each corporate customer;

[0095] Step 240: Train a first machine learning model based on the first data set and the label data of the first data set to obtain a return-risk prediction model.

[0096] This embodiment obtains the account information, transaction information, and income information of each historical corporate customer, uses the account information and transaction information as training data, and the income information as label data for the training data, and trains a machine learning model to obtain an income risk prediction model, providing a method for predicting subsequent customer income risks.

[0097] In some embodiments of this document, the account information, transaction information, and income information of each historical corporate customer within a predetermined period are obtained at predetermined time intervals. The predetermined period may be one consecutive week, two consecutive weeks, three consecutive weeks, one consecutive month, two consecutive months, etc., which is not limited in this document.

[0098] In some embodiments of this document, the revenue information may be marked as 0, indicating negative revenue growth, or marked as 1, indicating positive revenue growth. The probability of being marked as 1 is used as the positive revenue probability score for corporate customers. Other marking methods may also be used, which are not limited in this document.

[0099] In some embodiments of this document, the machine learning model may be a traditional machine learning model such as SVM or random forest, or a deep learning model, which is not limited in this document.

[0100] In one embodiment of this invention, Figure 3 As shown, the generation process of the churn risk prediction model includes:

[0101] Step 310: Obtain the second account information, second transaction information, and churn information of each historical corporate customer;

[0102] In step 310, the churn information includes identification data of whether each historical corporate customer has churned;

[0103] Step 320: determining a second data set based on the second account information of each historical corporate customer and the second transaction information of each historical corporate customer;

[0104] Step 330: determining label data of the second data set based on the historical churn information of each corporate customer;

[0105] Step 340: Train a second machine learning model based on the second data set and the label data of the second data set to obtain a churn risk prediction model.

[0106] This embodiment obtains the account information, transaction information, and churn information of each historical corporate customer, uses the account information and transaction information as training data, and the churn information as label data for the training data, and trains a machine learning model to obtain a churn risk prediction model, providing a method for predicting subsequent customer churn risks.

[0107] In some embodiments herein, at predetermined intervals, the account information, transaction information, and churn information of each historical corporate customer within a predetermined time period, as well as the positive return probability score obtained by the return risk prediction model, are obtained. The account information, transaction information, and positive return probability score serve as training data, and the churn information serves as label data for the training data, to train a machine learning model, thereby generating a churn risk prediction model. The predetermined time period can be a period when corporate customers have not churned within the bank, or it can be a period before corporate customers churn, and this is not limited herein.

[0108] In some embodiments of this document, the churn information may be marked as 0, indicating that the corporate customer has churned, or marked as 1, indicating that the corporate customer has not churned. The probability of being marked as 0 is used as the churn probability score of the corporate customer. Other identification methods may also be used, which are not limited in this document.

[0109] In some embodiments of this document, the machine learning model may be a traditional machine learning model such as SVM, random forest, or a deep learning model, which is not limited in this document.

[0110] In one embodiment of this invention, Figure 4 As shown, when the churn probability score is higher than a first preset threshold, it is determined that the corporate customer has a churn risk, further comprising:

[0111] Step 410: Obtain the satisfaction survey score, complaint frequency, knowledge graph feature value, and positive profit probability score of the corporate customer;

[0112] Step 420: When the churn probability score is higher than the first preset threshold, and satisfies at least one of the following conditions: the satisfaction survey score is lower than the second preset threshold, the complaint frequency is higher than the third preset threshold, the knowledge graph feature value is lower than the fourth preset threshold, and the positive benefit probability score is lower than the fifth preset threshold, it is determined that there is a risk of churn for the corporate customer.

[0113] After obtaining the churn probability score of corporate customers through the churn risk prediction model, this embodiment also considers extracting some important features as a judgment basis alone. Combining them with the churn probability score to judge the churn risk of corporate customers can more accurately determine corporate customers with churn risk.

[0114] In some embodiments of this document, when the churn probability score is higher than a first preset threshold, but any one of the satisfaction survey score, complaint frequency, knowledge graph feature value, and positive benefit probability score does not meet the relevant conditions, the churn probability score is considered unreliable and can be left unprocessed or predicted again in the next period of time to avoid increasing the human cost of verification.

[0115] In some embodiments of this document, it is possible to select, based on actual conditions, a case where the churn probability score is higher than a first preset threshold and matches one of the satisfaction survey score, complaint frequency, knowledge graph feature value, and positive benefit probability score, to consider that a corporate customer is at risk of churn. It is also possible to select a case where a case where the churn probability score is higher than a first preset threshold and matches at least two, or at least three, or all of the satisfaction survey score, complaint frequency, knowledge graph feature value, and positive benefit probability score, to consider that a corporate customer is at risk of churn. The specific selection is not limited in this document.

[0116] In one embodiment of this invention, Figure 5 As shown, the risk prediction method for corporate customers of banks also includes:

[0117] Step 510, determining whether the positive return probability score is lower than a sixth preset threshold;

[0118] Step 520: If the judgment is yes, it is determined that the corporate customer has a profit risk.

[0119] This embodiment takes into account the positive return probability score obtained through the return risk prediction model, which can reflect the return forecast of corporate customers, thereby helping customers to do risk prevention and control in real time based on the positive return probability score and improving the stickiness of corporate customers.

[0120] In some embodiments of this document, when it is determined that the positive return probability score is lower than the sixth preset threshold, the customer is considered to have a return risk, and relevant personnel can be arranged to verify the situation and then notify the customer. Multiple predictions can also be made in different time periods to ensure the accuracy of the predictions.

[0121] Based on the same inventive concept, this document also provides a risk prediction device for corporate bank customers, as described in the following embodiments. Because the principles underlying the risk prediction device for corporate bank customers are similar to those of the risk prediction method for corporate bank customers, the implementation of the risk prediction device for corporate bank customers can be referenced to the risk prediction method for corporate bank customers, and any repetitions will not be repeated.

[0122] Specifically, such as Figure 6 As shown, the risk prediction device for corporate customers of a bank includes:

[0123] A first acquisition module 610 is configured to acquire first account information and first transaction information of a corporate customer. The first account information includes the ledger data of each user of the corporate customer, the number of users, the number of products held, and the number of active users. The first transaction information includes the transaction amount, number of transactions, and transaction frequency of the corporate customer.

[0124] A first input module 620 is configured to input the first account information and first transaction information of the corporate customer into a pre-trained profit-risk prediction model to obtain a positive profit probability score for the corporate customer;

[0125] A second acquisition module 630 is configured to acquire second account information and second transaction information of the corporate customer, wherein the second account information includes the number of products held by the corporate customer, a satisfaction survey score, and complaint frequency, and the second transaction information includes the transaction amount, number of transactions, transaction frequency, and knowledge graph feature value of the corporate customer;

[0126] A second input module 640 is configured to input the second account information, second transaction information, and the positive profit probability score of the corporate customer into a pre-trained churn risk prediction model to obtain a churn probability score for the corporate customer;

[0127] The risk determination module 650 is configured to determine that the corporate customer has a churn risk when the churn probability score is higher than a first preset threshold.

[0128] This embodiment fully considers the account information, transaction information and user information of corporate customers. Through the account information, transaction information and user information of corporate customers, a profit risk prediction model is obtained. The profit risk prediction model can predict the positive profit probability score of corporate customers and provide data reference for the prediction of corporate customer churn risk. Combined with the account information, transaction information and positive profit probability score of corporate customers, a churn risk prediction model is obtained, thereby predicting the churn risk of corporate customers more comprehensively and accurately.

[0129] The risk prediction method, device and equipment for corporate customers of a bank provided in this article obtain a profit risk prediction model through the account information, transaction information and user information of the corporate customers. The profit risk prediction model can predict the positive profit probability score of the corporate customers and provide a data reference for predicting the risk of corporate customer churn. The churn risk prediction model is then obtained by combining the account information, transaction information and positive profit probability score of the corporate customers, thereby predicting the risk of corporate customer churn in a more comprehensive and accurate manner.

[0130] In one embodiment of the present invention, a computer device is further provided for implementing the method described in any of the above embodiments, such as Figure 7 The diagram shows a schematic diagram of the structure of a computer device according to an embodiment of the present invention. The computer device 702 may include one or more processors 704, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. The computer device 702 may also include any memory 706 for storing any type of information, such as code, settings, data, etc. For example, and without limitation, the memory 706 may include any one or more combinations of the following: any type of RAM, any type of ROM, a flash memory device, a hard disk, an optical disk, etc. More generally, any memory may use any technology to store information. Furthermore, any memory may provide volatile or non-volatile retention of information. Furthermore, any memory may represent a fixed or removable component of the computer device 702. In one embodiment, when the processor 704 executes associated instructions stored in any memory or combination of memories, the computer device 702 may perform any operation of the associated instructions. The computer device 702 also includes one or more drive mechanisms 708, such as a hard disk drive mechanism, an optical disk drive mechanism, etc., for interacting with any memory.

[0131] The computer device 702 may also include an input / output module 710 (I / O) for receiving various inputs (via input devices 712) and for providing various outputs (via output devices 714). A specific output mechanism may include a presentation device 716 and an associated graphical user interface (GUI) 718. In other embodiments, the input / output module 710 (I / O), input devices 710, and output devices 714 may not be included, and the computer device 702 may simply be a computer device in a network. The computer device 702 may also include one or more network interfaces 720 for exchanging data with other devices via one or more communication links 722. One or more communication buses 724 couple the components described above together.

[0132] The communication link 722 may be implemented in any manner, for example, via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 722 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc., governed by any protocol or combination of protocols.

[0133] Corresponding to Figure 1-Figure 5 The embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which executes the steps of the above method when executed by a processor.

[0134] The embodiment of the present invention also provides a computer readable instruction, wherein when the processor executes the instruction, the program causes the processor to execute the following Figures 1 to 5 The method shown.

[0135] It should be understood that in the various embodiments of this document, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this document.

[0136] It should also be understood that in the embodiments herein, the term "and / or" merely describes an association between associated objects, indicating that three possible relationships exist. For example, "A and / or B" could represent: A alone, A and B simultaneously, or B alone. Furthermore, the character " / " in this document generally indicates an "or" relationship between the associated objects.

[0137] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this document.

[0138] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0139] In the several embodiments provided herein, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices, or units, or can be an electrical, mechanical, or other form of connection.

[0140] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments herein.

[0141] In addition, the functional units in the various embodiments herein may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0142] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this article is essentially or the part that contributes to the existing technology, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of this article. The aforementioned storage medium includes: various media that can store program code, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0143] This article uses specific embodiments to illustrate the principles and implementation methods of this article. The description of the above embodiments is only used to help understand the methods and core ideas of this article. At the same time, for those skilled in the art, based on the ideas of this article, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation to this article.

Claims

1. A risk prediction method for corporate bank customers, characterized in that: The method comprises: Obtaining first account information and first transaction information of a corporate customer, wherein the first account information includes the ledger data of each user of the corporate customer, the number of users, the number of products held, and the number of active users; and the first transaction information includes the transaction amount, number of transactions, and transaction frequency of the corporate customer; Inputting the first account information and the first transaction information of the corporate customer into a pre-trained profit-risk prediction model to obtain a positive profit probability score for the corporate customer; Obtaining second account information and second transaction information of the corporate customer, where the second account information includes the number of products held by the corporate customer, a satisfaction survey score, and complaint frequency, and the second transaction information includes the transaction amount, number of transactions, transaction frequency, and knowledge graph feature value of the corporate customer; Inputting the second account information, second transaction information, and the positive profit probability score of the corporate customer into a pre-trained churn risk prediction model to obtain the churn probability score of the corporate customer; When the churn probability score is higher than a first preset threshold, determining that the corporate customer has a churn risk; The generation process of the return-risk prediction model includes: Obtaining first account information, first transaction information, and income information of each historical corporate customer, wherein the income information includes income data of each historical corporate customer within a predetermined period; determining a first data set based on the first account information of each historical corporate customer and the first transaction information of each historical corporate customer; Determining label data of the first data set based on the historical revenue information of each corporate customer; Training a first machine learning model based on the first data set and the label data of the first data set to obtain a return-risk prediction model; The generation process of the churn risk prediction model includes: Acquire the second account information, second transaction information, and churn information of each historical corporate customer, wherein the churn information includes identification data indicating whether each historical corporate customer has churned; determining a second data set based on the second account information of each historical corporate customer, the second transaction information of each historical corporate customer, and the positive return probability score; determining label data of the second data set based on the historical churn information of each corporate customer; A second machine learning model is trained based on the second data set and the label data of the second data set to obtain a churn risk prediction model.

2. The method according to claim 1, wherein The first transaction information also includes the interest rate allocated by the corporate customer to each user and the user's early interest settlement rate; The interest ratio represents the ratio of the deposit interest allocated by the corporate client to each user; The user early interest settlement rate refers to the proportion of users who initiated refunds and interest settlements in advance before the interest settlement date to the total number of users under the corporate customer.

3. The method according to claim 1, wherein The second transaction information also includes the amount transferred by the corporate customer to the corporate customer's external institution account and the frequency of the corporate customer transferring money to the corporate customer's external institution account.

4. The method according to claim 1, wherein When the churn probability score is higher than a first preset threshold, determining that the corporate customer has a churn risk further includes: Obtaining the satisfaction survey score, complaint frequency, knowledge graph feature value, and positive profit probability score of the corporate customers; When the churn probability score is higher than the first preset threshold, and satisfies at least one of the following conditions: the satisfaction survey score is lower than the second preset threshold, the complaint frequency is higher than the third preset threshold, the knowledge graph feature value is lower than the fourth preset threshold, and the positive benefit probability score is lower than the fifth preset threshold, it is determined that there is a risk of churn for the corporate customer.

5. The method according to claim 1, wherein The method further comprises: Determining whether the positive return probability score is lower than a sixth preset threshold; If the judgment is yes, it is determined that the corporate customer has a profit risk.

6. A risk prediction device for corporate customers of a bank, characterized in that: The device comprises: A first acquisition module is configured to acquire first account information and first transaction information of a corporate customer, wherein the first account information includes the ledger data of each user of the corporate customer, the number of users, the number of products held, and the number of active users; and the first transaction information includes the transaction amount, the number of transactions, and the transaction frequency of the corporate customer; a first input module, configured to input the first account information and first transaction information of the corporate customer into a pre-trained profit-risk prediction model to obtain a positive profit probability score for the corporate customer; a second acquisition module configured to acquire second account information and second transaction information of the corporate customer, wherein the second account information includes the number of products held by the corporate customer, a satisfaction survey score, and complaint frequency, and the second transaction information includes the transaction amount, number of transactions, transaction frequency, and knowledge graph feature value of the corporate customer; a second input module, configured to input the second account information, second transaction information, and the positive profit probability score of the corporate customer into a pre-trained churn risk prediction model to obtain a churn probability score of the corporate customer; a risk determination module, configured to determine that the corporate customer has a churn risk when the churn probability score is higher than a first preset threshold; The generation process of the return-risk prediction model includes: Obtaining first account information, first transaction information, and income information of each historical corporate customer, wherein the income information includes income data of each historical corporate customer within a predetermined period; determining a first data set based on the first account information of each historical corporate customer and the first transaction information of each historical corporate customer; Determining label data of the first data set based on the historical revenue information of each corporate customer; Training a first machine learning model based on the first data set and the label data of the first data set to obtain a return-risk prediction model; The generation process of the churn risk prediction model includes: Acquire the second account information, second transaction information, and churn information of each historical corporate customer, wherein the churn information includes identification data indicating whether each historical corporate customer has churned; determining a second data set based on the second account information of each historical corporate customer, the second transaction information of each historical corporate customer, and the positive return probability score; determining label data of the second data set based on the historical churn information of each corporate customer; A second machine learning model is trained based on the second data set and the label data of the second data set to obtain a churn risk prediction model.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method according to any one of claims 1 to 5 is implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor of a computer device, the method according to any one of claims 1 to 5 is implemented.

9. A computer program product, comprising a computer program, characterized in that: When the computer program is executed by a processor of a computer device, the method according to any one of claims 1 to 5 is implemented.

Images