Bank customer risk prediction method, system equipment and storage medium

Abstract

The invention discloses a bank customer risk prediction method, system equipment and a storage medium, risk customers and related customers are screened out from massive customers by using geocoding and customer characteristics, and then risk prediction is further performed on suspicious customers by using social risk relationships through a knowledge graph system. The calculation amount of data is greatly reduced, the performance of the whole model is improved, geographic grids of customer dimensions are constructed, different types of customer risk characteristics and geographic grid codes are divided from the customer dimensions, the accuracy of the model is improved, risk characteristic data of grid dimensions are constructed, various risk conditions in the grids are summarized, and the risk characteristics of the grids are analyzed. And the interpretability of risk grid prediction is improved. In combination with the social graph, the grid features are used as a new side relationship, and the risk degree of the new application document is predicted by using the social relationship between the new application document and the existing application document and between the new application documents, so that the accuracy of application document risk prediction is improved from multiple dimensions.

Description

Technical Field

[0001] The present invention belongs to the field of anti-fraud technology, and in particular, relates to the technical direction of integrating geographic coding and knowledge graphs, and more particularly to a bank customer risk prediction method, system equipment and storage medium. Background Art

[0002] With the vigorous development of Internet finance, the thresholds for financial management, investment, loans and financing have been significantly lowered, bringing unprecedented convenience to the public. However, there are many fraud methods in the current credit card industry, and ordinary big data risk control cannot timely intercept the risk characteristics of customer input address clustering.

[0003] To this end, the invention patent with patent application number CN202311008053 provides an anti-fraud method based on geocoding. For the target geocoding area and the area adjacent to the target GPS coordinates, the number of applicants at different times is counted respectively, and it is determined whether the number of applicants at different statistical times is abnormal. When there is an abnormality, a fraud warning is issued. However, this method limits the geocoding to an 8-bit geographic grid, and the actual geographic distance is about 50 meters * 50 meters. For the whole country, the number of applications in most geographic grids is small, and there is no way to be adaptive. In addition, the changes in applications within a time period are not comprehensively considered. Only the observation window of a certain day is observed to change slightly, resulting in inaccurate risk prediction. In addition, this method directly uses the number of applications as the criterion for judging risk, without carefully considering the risk characteristics of the customer itself, and there is a problem of inaccurate risk prediction.

[0004] Therefore, how to propose an effective technical solution has become a hot issue that technical personnel in the financial industry need to solve urgently. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a bank customer risk prediction method, system equipment and storage medium, which solves the inaccurate prediction caused by the different number of applicants in different regions in the prior art, resulting in a small number of applications in most geographical grids, and comprehensively considers the changes in applications within a period of time, and does not simply use the number of applications as a criterion for judging risk, but considers at least one technical problem of the risk characteristics of the customer itself.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a method for predicting bank customer risk, comprising:

[0007] Obtain address information of historical application documents for analysis;

[0008] Performing grid division according to the analytical information;

[0009] Cluster the historical applications by grid codes, calculate the number of applications in the same grid, and perform label analysis on the customers in the grid;

[0010] Analyze the distribution of risky customers in the grid and the distribution of the overall risk grid, divide the grid by threshold, obtain the risk attributes of the newly added application grid, and screen the risk grid;

[0011] Obtaining customer basic information and social relationships from the customer basic information, and constructing a customer-customer same-medium graph;

[0012] Conduct risk transmission to customers within the grid and output high social risk customers within the risk grid.

[0013] Furthermore, the acquisition of historical application address information for parsing includes: converting unstructured address data into longitude and latitude data through address standardization technology and address parsing technology.

[0014] Furthermore, the grid division according to the parsed information includes: using Geohash coding technology to calculate the longitude and latitude data into a 5-bit grid code as a grid reference; when the application files are unevenly distributed within the geographic grid, the grid is adaptively lowered; for areas where the number of applications is lower than the set value, the calculation is directly performed according to the 5-bit grid code; for areas with a high level of application density, the grid code is adaptively lowered.

[0015] Furthermore, the tag analysis of customers in the grid includes:

[0016] Calculate customer risk points based on risk tags. Customer risks are divided into:

[0017] score cust =∑ i a i w i ; [1]

[0018] where a i Indicates the risk label determined by the business end, w i Indicates the weight of the corresponding risk label;

[0019] The overall risk score of the grid is calculated by combining the risk scores of all customers in the grid and the time dimension weight of the application. The overall risk score of the grid is calculated as follows:

[0020]

[0021] in is the calculated risk score of customers in the grid, is the time dimension risk weight of the corresponding customer, and count(cust) is the number of customers in the grid.

[0022] Furthermore, the distribution of risky customers within the grid and the distribution of the overall risk grid are analyzed, the grid is divided by thresholds, and the risk attributes of the new application grid are obtained. The risk grid screening includes: for new applications added daily, the application address is grid-coded, compared with the risk grid, and the risk grid code and the risk situation of the grid are output.

[0023] Furthermore, the risk transmission to customers in the grid and output of high social risk customers in the risk grid include: based on medium and high risk grids, risk correlation analysis is performed on new and old customers in the risk grid, and risk customers with higher social risk scores are screened out from the daily new applications in the risk grid for early warning.

[0024] In a second aspect, the present invention also provides a bank customer risk prediction system, including: a coding module, a calculation module, a judgment module, a risk transmission module, and an early warning module;

[0025] Coding module: used to obtain the address information of historical application documents for analysis, and to perform grid division according to the analysis information;

[0026] Calculation module: used to cluster the historical applications into grid codes, calculate the number of applications in the same grid, and perform label analysis on customers in the grid;

[0027] Judgment module: used to analyze the distribution of risky customers in the grid and the distribution of the overall risk grid, divide the grid into thresholds, obtain the risk attributes of the newly added application grid, and screen the risk grid;

[0028] Risk transmission module: used to obtain basic customer information and social relationships from the basic customer information, and build a customer-customer homomediary map;

[0029] Early warning module: used to transmit risks to customers in the grid and output high social risk customers in the risk grid.

[0030] In a third aspect, the present invention provides a computing device, comprising:

[0031] A memory for storing program instructions;

[0032] The processor is used to call the program instructions stored in the memory and execute the above-mentioned bank customer risk prediction method in whole or in part according to the obtained program instructions.

[0033] In a fourth aspect, the present invention provides a computer-readable storage medium, including computer-readable instructions. When a computer reads and executes the computer-readable instructions, the above-mentioned bank customer risk prediction method is fully or partially implemented. The readable medium may be multiple, and the multiple readable media can operate independently of each other.

[0034] In a fifth aspect, the present invention provides a computer program product, which includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the above-mentioned bank customer risk prediction method is fully or partially executed.

[0035] Compared with the existing technology, this solution constructs a geographic grid in the customer dimension and constructs risk feature data in the grid dimension and combines it with the social graph to divide different types of customer risk features and geographic grid codes from the customer dimension, effectively improving the accuracy of the model. It summarizes various risk situations within the grid and effectively improves the interpretability of risk grid prediction. At the same time, the grid feature is used as a new edge relationship, and the social relationship between the new application and the existing application and the new application is used to predict the risk level of the new application, thereby improving the accuracy of the risk prediction of the application from multiple dimensions. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0037] Figure 1 A schematic diagram of a flow chart of an embodiment of a bank customer risk prediction method provided by the present invention;

[0038] Figure 2 A structural block diagram of an embodiment of a bank customer risk prediction system provided by the present invention;

[0039] Figure 3 The present invention provides a schematic structural diagram of an electronic device. DETAILED DESCRIPTION

[0040] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be interpreted as limiting the present application.

[0041] It will be understood by those skilled in the art that, unless otherwise stated, the singular forms "a", "an" and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. The term "and / or" used herein includes all or any unit and all combinations of one or more associated listed items.

[0042] At present, there are many fraud methods in the credit card industry. For ordinary big data risk control, there is no way to timely intercept the risk characteristics of customer incoming address clustering. The present invention will cluster the application and address grid through address resolution and address encoding algorithms, combine with knowledge graphs, and use risk transmission to predict the risk of applications in the address grid.

[0043] In order to make the objectives, technical solutions and advantages of the present application more clear, the specific implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0044] The core of the present invention is to provide a method for predicting bank customer risk. Figure 1 The flowchart of the bank customer risk prediction method provided by this embodiment is shown in FIG. Figure 1 The technical solution of this embodiment is described in detail:

[0045] Step S1: Obtain address information of historical application documents for analysis; perform grid division according to the analysis information;

[0046] In the specific implementation of this specification, the unstructured address data is converted into longitude and latitude data through address standardization technology and address resolution technology.

[0047] Step S2: cluster the historical applications by grid codes, calculate the number of applications in the same grid, and perform label analysis on the customers in the grid;

[0048] In the specific implementation of this specification, Geohash coding technology is used to calculate the longitude and latitude data into a 5-bit grid code as a grid reference; when the distribution of applications in the geographic grid is uneven, the grid is adaptively lowered; for areas where the number of applications is lower than the set value, the calculation is directly based on the 5-bit grid code; for areas with a high level of application density, the grid code is adaptively lowered.

[0049] Step S3: Analyze the distribution of risky customers in the grid and the distribution of the overall risk grid, divide the grid into thresholds, obtain the risk attributes of the newly added application grid, and screen the risk grid;

[0050] In a specific implementation of the present specification, for each new application added daily, the address of the application is grid-coded, compared with the risk grid, and the risk grid code and the risk status of the grid are output.

[0051] Step S4: Obtaining customer basic information and social relationships from the customer basic information, and constructing a customer-customer same-medium graph;

[0052] Step S5: Conduct risk transmission for customers in the grid and output high social risk customers in the risk grid.

[0053] In the specific implementation of this specification, based on the medium and high risk grids, risk association analysis is performed on new and old customers in the risk grid, and risk customers with higher social risk scores are screened out from the daily new applications in the risk grid for early warning.

[0054] According to one aspect of the present invention, unstructured address data is converted into longitude and latitude data through address standardization technology and address parsing technology.

[0055] The longitude and latitude data are calculated into a 5-bit grid code using Geohash encoding technology as the grid reference.

[0056] The historical applications are clustered with grid codes to calculate the number of applications in the same grid. Considering the uneven distribution of applications in the geographic grid, the present invention proposes to adaptively probe the grid. For areas with a small number of applications, the calculation can be performed directly according to the 5-bit grid code. For areas with a high application density, the grid code is adaptively probed to improve the grid scale and the accuracy of the analysis.

[0057] Perform label analysis on customers in the grid and calculate customer risk scores based on risk labels. Risk scores in the customer dimension:

[0058] score cust =∑ i a i w i ; [1]

[0059] where a i Indicates the risk label determined by the business end, w i Indicates the weight of the corresponding risk label.

[0060] Combine the risk scores of all customers in the grid with the time dimension weight of the application to calculate the overall risk score of the grid. The overall risk score of the grid dimension is:

[0061]

[0062] in, The risk score of customers in the grid calculated by [1], is the time dimension risk weight of the corresponding customer, and count(cust) is the number of customers in the grid.

[0063] Analyze the distribution of risky customers within the grid and the distribution of the overall risk grid, and divide the grid into thresholds.

[0064] For new applications added daily, the application address is grid-coded and compared with the risk grid, and the risk grid code and risk status of the grid are output.

[0065] Obtain customer basic information and social relationships from customer basic information and build a customer-customer homomediary graph. isomo { i |u j >,w ij},in i |u j > represents u i and u j There is information or social relationship, w ij Indicates u i and u j The relationship weight can be used to determine the relationship weight of u i and u j The closeness of the relationship between them.

[0066] Based on the medium and high risk grids, risk correlation analysis is conducted on new and old customers in the risk grid, and risk customers with higher social risk scores are screened out from the daily new applications in the black grid for early warning.

[0067] The solution of this embodiment uses geographic coding and customer characteristics to screen out risky customers and their related customers from a large number of customers, and then uses the knowledge graph system to further predict the risk of suspicious customers using social risk relationships, which greatly reduces the amount of data calculation and improves the performance of the overall model. Construct a geographic grid in the customer dimension, divide different types of customer risk characteristics and geographic grid codes from the customer dimension, and improve the accuracy of the model. Construct risk feature data in the grid dimension, summarize various risk situations within the grid, and improve the interpretability of risk grid prediction. Combined with the social graph, the grid feature is used as a new edge relationship, and the social relationship between the new application and the existing application and the new application is used to predict the risk level of the new application, thereby improving the accuracy of the risk prediction of the application from multiple dimensions.

[0068] This method has been tested many times, and the top 10 black grids were selected for offline verification and control. The 10 black grids that were recently checked showed that the overall abnormal incoming cases accounted for 22.10%, the overall malicious cases accounted for 3.26%, and the abnormal coverage rate reached 88%.​​

[0069] In one embodiment, the present invention provides a bank customer risk prediction system. Figure 2 The figure shows a structural block diagram of a bank customer risk prediction system provided by this embodiment, wherein the system comprises: an encoding module 10, a calculation module 20, a judgment module 30, a risk transmission module 40, and an early warning module 50;

[0070] Coding module 10: used to obtain historical application address information for analysis, and to perform grid division according to the analysis information;

[0071] Calculation module 20: used to cluster the historical applications into grid codes, calculate the number of applications in the same grid, and perform label analysis on customers in the grid;

[0072] The judgment module 30 is used to analyze the distribution of risky customers in the grid and the distribution of the overall risk grid, divide the grid into thresholds, obtain the risk attributes of the newly added application grid, and screen the risk grid;

[0073] Risk transmission module 40: used to obtain customer basic information and social relationships from the customer basic information, and build a customer-customer same-medium map;

[0074] Early warning module 50: used to conduct risk transmission to customers in the grid and output high social risk customers in the risk grid.

[0075] The solution of this embodiment uses geographic coding and customer characteristics to screen out risky customers and their related customers from a large number of customers, and then uses the knowledge graph system to further predict the risk of suspicious customers using social risk relationships, which greatly reduces the amount of data calculation and improves the performance of the overall model. Construct a geographic grid in the customer dimension, divide different types of customer risk characteristics and geographic grid codes from the customer dimension, and improve the accuracy of the model. Construct risk feature data in the grid dimension, summarize various risk situations within the grid, and improve the interpretability of risk grid prediction. Combined with the social graph, the grid feature is used as a new edge relationship, and the social relationship between the new application and the existing application and the new application is used to predict the risk level of the new application, thereby improving the accuracy of the risk prediction of the application from multiple dimensions.

[0076] The bank customer risk prediction system of the present embodiment is used to implement the aforementioned bank customer risk prediction method. Therefore, the specific implementation methods of the bank customer risk prediction system can be seen in the specific implementation methods of the bank customer risk prediction method in the previous text. For example, the encoding module 10 is used to implement step S1 in the aforementioned bank customer risk prediction method, the calculation module 20 is used to implement step S2 in the aforementioned bank customer risk prediction method, the judgment module 30 is used to implement step S3 in the aforementioned bank customer risk prediction method, the risk transmission module 40 is used to implement step S4 in the aforementioned bank customer risk prediction method, and the early warning module 50 is used to implement step S5 in the aforementioned bank customer risk prediction method. Therefore, its specific implementation methods can refer to the descriptions of the corresponding embodiments of each part. The system provided by the present invention is attached to the method provided by the present invention, and its specific effects can refer to the above method, which will not be repeated here.

[0077] In one embodiment, a computing device is provided, comprising:

[0078] A memory for storing program instructions;

[0079] A processor is used to implement the steps of any of the above-mentioned bank customer risk prediction methods when executing the computer program.

[0080] As an example, Figure 3 A schematic diagram of the structure of an electronic device applicable to an embodiment of the present application is shown. Figure 3 As shown, the electronic device 2000 includes: a processor 2001 and a memory 2003. The processor 2001 and the memory 2003 are connected, such as through a bus 2002. Optionally, the electronic device 2000 may also include a transceiver 2004. It should be noted that in actual applications, the transceiver 2004 is not limited to one, and the structure of the electronic device 2000 does not constitute a limitation on the embodiments of the present application.

[0081] The processor 2001 is applied in the embodiment of the present application to implement the method shown in the above method embodiment. The transceiver 2004 may include a receiver and a transmitter. The transceiver 2004 is applied in the embodiment of the present application to implement the function of the electronic device of the embodiment of the present application communicating with other devices when executed.

[0082] Processor 2001 may be a CPU (Central Processing Unit), a general purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 2001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0083] The bus 2002 may include a path to transmit information between the above components. The bus 2002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 2002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0084] The memory 2003 can be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to these.

[0085] Optionally, the memory 2003 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 2001. The processor 2001 is used to execute the application code stored in the memory 2003 to implement the investment and financing project recommendation method provided in any embodiment of the present application.

[0086] The electronic device provided in the embodiment of the present application is applicable to any embodiment of the above method and will not be described in detail here.

[0087] The present invention provides a bank customer risk prediction method, which has obvious advantages and beneficial effects compared with the prior art. By using geographic coding and customer characteristics, risky customers and related customers are screened out from a large number of customers, and then the knowledge graph system is used to further predict the risk of suspicious customers using social risk relationships, which greatly reduces the amount of data calculation and improves the performance of the overall model. A geographic grid of the customer dimension is constructed, and the risk characteristics and geographic grid codes of different types of customers are divided from the customer dimension to improve the accuracy of the model. Risk feature data of the grid dimension is constructed to summarize various risk situations within the grid and improve the interpretability of risk grid prediction. Combined with the social graph, the grid feature is used as a new edge relationship, and the social relationship between the new application and the existing application and the new application is used to predict the risk level of the new application, thereby improving the accuracy of the risk prediction of the application from multiple dimensions.

[0088] In one embodiment, a computer-readable storage medium is provided, including computer-readable instructions. When a computer reads and executes the computer-readable instructions, the steps of the above-mentioned method for predicting bank customer risks are implemented. The method for predicting bank customer risks provided by the present invention has obvious advantages and beneficial effects compared with the prior art. By using geographic coding and customer characteristics, risk customers and related customers are screened out from a large number of customers, and then the social risk relationship is used through the knowledge graph system to further predict the risk of suspicious customers, which greatly reduces the amount of data calculation and improves the performance of the overall model. A geographic grid of the customer dimension is constructed, and different types of customer risk characteristics and geographic grid codes are divided from the customer dimension to improve the accuracy of the model. Risk feature data of the grid dimension is constructed to summarize various risk situations in the grid and improve the interpretability of risk grid prediction. Combined with the social graph, the grid feature is used as a new edge relationship, and the social relationship between the new application and the existing application and the new application is used to predict the risk level of the new application, thereby improving the accuracy of the risk prediction of the application from multiple dimensions.

[0089] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the specific system can be divided into different functional modules to complete all or part of the functions described above.

[0090] Those of ordinary skill in the art will appreciate that the modules 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 composition and steps of each example have been generally described in terms of function in the above description. 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 to be beyond the scope of this application.

[0091] In several embodiments covered by the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the division of modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules 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 modules, or it can be an electrical, mechanical or other form of connection.

[0092] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiments of the present application.

[0093] In addition, each functional module in each embodiment of the present application can be implemented in the form of hardware or in the form of software functional modules. If these functional modules are implemented in the form of software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application is essentially or part of the prior art that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a computer program product, which includes one or more computer instructions. When loading and executing computer program instructions on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer program product is stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, a computer instruction can be transmitted from a website site, a computer, a server or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrations. The available media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state disks (SSDs)).

[0094] The above contents are merely examples and explanations of the structure of the present invention. The technicians in this technical field may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the structure of the invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.

Claims

1. A bank customer risk prediction method, characterized in that: include: Obtain address information of historical application documents for analysis; Performing grid division according to the analytical information; Cluster the historical applications by grid codes, calculate the number of applications in the same grid, and perform label analysis on the customers in the grid; Analyze the distribution of risky customers in the grid and the distribution of the overall risk grid, divide the grid by threshold, obtain the risk attributes of the newly added application grid, and screen the risk grid; Obtaining basic customer information and social relationships from the basic customer information, and constructing a customer-customer homomedia graph; Conduct risk transmission to customers within the grid and output high social risk customers within the risk grid.

2. The bank customer risk prediction method according to claim 1, characterized in that: The obtaining and parsing of the address information of the historical application documents includes: converting the unstructured address data into longitude and latitude data through address standardization technology and address parsing technology.

3. The bank customer risk prediction method according to claim 1, characterized in that: The grid division according to the parsed information includes: using Geohash coding technology to calculate the latitude and longitude data into a 5-bit grid code as a grid reference; when the application files are unevenly distributed within the geographic grid, the grid is adaptively lowered; for areas where the number of applications is lower than the set value, the 5-bit grid code is calculated; for areas with a high application density, the grid code is adaptively lowered.

4. The bank customer risk prediction method according to claim 1, characterized in that: The tag analysis of customers in the grid includes: Calculate customer risk points based on risk tags. Customer risks are divided into: score cust =∑ i a i w i ; [1] where a i Indicates the risk label determined by the business end, w i Indicates the weight of the corresponding risk label; The overall risk score of the grid is calculated by combining the risk scores of all customers in the grid and the time dimension weight of the application. The overall risk score of the grid is calculated as follows: in is the calculated risk score of customers in the grid, is the time dimension risk weight of the corresponding customer, and count(cust) is the number of customers in the grid.

5. The bank customer risk prediction method according to any one of claim 1, characterized in that: The distribution of risky customers in the grid and the distribution of the overall risk grid are analyzed, the grid is divided by thresholds, and the risk attributes of the new application grid are obtained. The risk grid screening includes: for new applications added daily, the application address is grid-coded, compared with the risk grid, and the risk grid code and the risk situation of the grid are output.

6. The bank customer risk prediction method according to claim 1, characterized in that: The risk transmission for customers in the grid and output of high social risk customers in the risk grid include: based on the medium and high risk grids, risk correlation analysis is performed on new and old customers in the risk grid, and risk customers with higher social risk scores are screened out from the daily new applications in the risk grid for early warning.

7. A bank customer risk prediction system, characterized in that: The system includes: a coding module, a calculation module, a judgment module, a risk transmission module, and an early warning module; Coding module: used to obtain the address information of historical application documents for analysis, and to perform grid division according to the analysis information; Calculation module: used to cluster the historical applications into grid codes, calculate the number of applications in the same grid, and perform label analysis on customers in the grid; Judgment module: used to analyze the distribution of risky customers in the grid and the distribution of the overall risk grid, divide the grid into thresholds, obtain the risk attributes of the newly added application grid, and screen the risk grid; Risk transmission module: used to obtain basic customer information and social relationships from the basic customer information, and build a customer-customer homomediary map; Early warning module: used to transmit risks to customers in the grid and output high social risk customers in the risk grid.

8. A computing device, characterized in that include: A memory for storing program instructions; A processor is used to call the program instructions stored in the memory, and execute the method according to any one of claims 1 to 6 according to the obtained program instructions.

9. A computer-readable storage medium, characterized in that: It includes computer-readable instructions. When a computer reads and executes the computer-readable instructions, it implements the method according to any one of claims 1 to 6. There may be multiple readable media, and the multiple readable media can operate independently of each other.

10. A computer program product, characterized in that The computer program product comprises one or more computer instructions, and when the computer program instructions are loaded and executed on a computer, the method according to any one of claims 1 to 6 is fully or partially executed.

Images