Financial risk model optimization processing method and device, equipment and storage medium
By analyzing and screening historical financial risk reports and customer data, target financial risk models are generated, and the problems of single and similarity of financial risk reports caused by fixed templates are solved, achieving richness and accuracy of the report content.
Patent Information
- Application Number
- CN202510335298.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-20
- Publication Date
- 2025-06-20
AI Technical Summary
In the prior art, a unified financial risk report is produced for customers through fixed templates, resulting in a single report expression and a high similarity between different reports.
By obtaining multiple historical financial risk reports and customer data, conducting statement splitting, data analysis and screening, determining target information and suspicious statements, conducting various methods of model training and evaluation, and generating target financial risk models for generating financial risk reports.
It improves the richness and accuracy of the expression content of financial risk reports, reduces the similarity between different reports, and makes the reports more personalized and accurate.
Smart Images
Figure CN120182010A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of computer technology, and in particular, to an optimization processing method, device, equipment, and storage medium for a financial risk model. Background Art
[0002] A large model refers to a machine learning model with a large number of parameters and a complex computational structure, designed to improve the model's expressive ability and prediction performance so as to be able to handle more complex tasks and data. Among them, the large model can also be a specific model such as a financial risk model.
[0003] In the related art, when it is necessary for a customer to generate a corresponding financial risk report, a unified financial risk report is made for the customer through a fixed template, and the financial risk report is output as a reference for whether the customer has financial risks.
[0004] However, in the related art, making a unified financial risk report for customers through a fixed template results in a single expression of the financial risk report and a high similarity between different financial risk reports. Summary of the Invention
[0005] This application provides an optimization processing method, device, equipment, and storage medium for a financial risk model to solve the problem that making a unified financial risk report for customers through a fixed template results in a single expression of the financial risk report and a high similarity between different financial risk reports.
[0006] In a first aspect, this application provides an optimization processing method for a financial risk model, which is applied to a computer device and includes:
[0007] Obtain a plurality of historical financial risk reports and the corresponding historical customer data for each historical financial risk report;
[0008] Split the sentences of each historical financial risk report to obtain the corresponding suspicious sentences;
[0009] Perform data analysis on each historical customer data to obtain each analyzed historical customer data;
[0010] Determine the target information corresponding to the suspicious sentences according to the analyzed historical customer data;
[0011] Screen each target information and each suspicious sentence to obtain each screened target information and each screened suspicious sentence;
[0012] Train a preset financial risk model in multiple different ways according to the screened target information and the screened suspicious sentences to obtain multiple trained financial risk models;
[0013] Evaluate the reliability, professionalism, and integrity of each trained financial risk model to obtain corresponding evaluation values;
[0014] Determine the target financial risk model from the multiple trained financial risk models according to the respective evaluation values;
[0015] Output the target financial risk model, where the financial risk model is used to generate a financial risk report based on the input customer data to be processed.
[0016] In a possible design, where each of the suspicious statements corresponds to each of the target information; correspondingly, the screening of each target information and each suspicious statement to obtain each screened target information and each screened suspicious statement includes: inputting any target information and the corresponding suspicious statement into the preset financial risk model for processing to determine whether the target information matches the suspicious statement; if it is determined that the target information matches the suspicious statement, then determine the target information and the suspicious statement as the screened target information and the screened suspicious statement; if it is determined that the target information does not match the suspicious statement, then filter out the target information and the suspicious statement; traverse the remaining target information and suspicious statements to obtain each screened target information and each screened suspicious statement.
[0017] In a possible design, the evaluation of the reliability, professionalism, and integrity of each trained financial risk model to obtain corresponding evaluation values includes: evaluating the reliability of each trained financial risk model to obtain a corresponding first evaluation value; evaluating the professionalism of each trained financial risk model to obtain a corresponding second evaluation value; evaluating the integrity of each trained financial risk model to obtain a corresponding third evaluation value; determining the corresponding arithmetic mean according to the first evaluation value, the second evaluation value, and the third evaluation value; and determining the arithmetic mean as the corresponding evaluation value.
[0018] In a possible design, the evaluation of the reliability of each trained financial risk model to obtain a corresponding first evaluation value includes: obtaining any financial risk report and the corresponding customer data; determining a plurality of suspicious statements according to the financial risk report, and determining a corresponding plurality of known information according to the customer data; determining whether the trained financial risk model can generate corresponding suspicious statements according to each known information; if it is determined that the trained financial risk model can generate corresponding suspicious statements according to one or more known information, then obtain the first quantity corresponding to the one or more known information; and completing the reliability evaluation according to the first quantity to obtain a corresponding first evaluation value.
[0019] In a possible design, the financial risk report further includes a suspicious conclusion statement; correspondingly, the professional evaluation of each trained financial risk model to obtain the corresponding second evaluation value includes: determining whether the trained financial risk model can generate the suspicious conclusion statement based on each suspicious statement; if it is determined that the trained financial risk model can generate the suspicious conclusion statement based on one or more suspicious statements, obtaining the corresponding second quantity of the one or more suspicious statements; and completing the professional evaluation according to the second quantity to obtain the corresponding second evaluation value.
[0020] In a possible design, the integrity evaluation of each trained financial risk model to obtain the corresponding third evaluation value includes: determining whether the trained financial risk model can generate each suspicious statement based on the financial risk report; if it is determined that the trained financial risk model can generate one or more suspicious statements based on the financial risk report, obtaining the corresponding third quantity of the one or more suspicious statements; and completing the integrity evaluation according to the third quantity to obtain the corresponding third evaluation value.
[0021] In a possible design, after outputting the target financial risk model, it further includes: obtaining the to-be-processed customer data corresponding to any customer; inputting the to-be-processed customer data into the target financial risk model for processing to generate a financial risk report; and outputting the financial risk report corresponding to the customer.
[0022] In a second aspect, the present application provides an optimization processing device for a financial risk model, which is applied to a computer device and includes:
[0023] A first acquisition module, configured to acquire a plurality of historical financial risk reports and the corresponding historical customer data of each historical financial risk report;
[0024] A splitting module, configured to split the statements of each historical financial risk report to obtain the corresponding suspicious statements;
[0025] An analysis module, configured to perform data analysis on each historical customer data to obtain each analyzed historical customer data;
[0026] A first determination module, configured to determine the target information corresponding to the suspicious statement according to each analyzed historical customer data;
[0027] A screening module, configured to screen each target information and each suspicious statement to obtain each screened target information and each screened suspicious statement;
[0028] A training module, configured to train a preset financial risk model in multiple different ways according to the filtered target information and the filtered suspicious statements, so as to obtain multiple trained financial risk models;
[0029] An evaluation module, configured to evaluate the reliability, professionalism and integrity of each trained financial risk model to obtain corresponding evaluation values;
[0030] A second determination module, configured to determine a target financial risk model from the multiple trained financial risk models according to the evaluation values;
[0031] A first output module, configured to output the target financial risk model, where the financial risk model is used to generate a financial risk report according to the input customer data to be processed.
[0032] In a third aspect, the present application provides a computer device, including: at least one processor and a memory;
[0033] The memory stores computer execution instructions;
[0034] The at least one processor executes the computer execution instructions stored in the memory, so that the at least one processor executes the optimization processing method of the financial risk model described in the first aspect and various possible designs of the first aspect above.
[0035] In a fourth aspect, the present application provides a computer storage medium, in which computer execution instructions are stored, and when the processor executes the computer execution instructions, the optimization processing method of the financial risk model described in the first aspect and various possible designs of the first aspect above is implemented.
[0036] The optimization processing method, device, equipment and storage medium of the financial risk model provided by this application split the sentences of each obtained historical financial risk report to obtain corresponding suspicious sentences; perform data analysis on each obtained historical customer data to obtain each analyzed historical customer data; determine the target information corresponding to the suspicious sentences according to each analyzed historical customer data; screen each target information and each suspicious sentence to obtain each screened target information and each screened suspicious sentence; train the preset financial risk model in a variety of different ways according to each screened target information and each screened suspicious sentence to obtain multiple trained financial risk models; evaluate the reliability, professionalism and integrity of each trained financial risk model to obtain corresponding evaluation values; determine the target financial risk model from multiple trained financial risk models according to each evaluation value; output the target financial risk model, where the financial risk model is used to generate a financial risk report according to the input customer data to be processed. By using the target financial risk model instead of a fixed template, the content of the subsequent generated financial risk report is rich and accurate, and the similarity between different financial risk reports is reduced. Brief Description of the Drawings
[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0038] Figure 1 It is a schematic diagram of the application scenario of the optimization processing method of the financial risk model provided by the embodiment of the present application;
[0039] Figure 2 It is a flowchart of the optimization processing method of the financial risk model provided by the embodiment of the present application Figure 1 ;
[0040] Figure 3 It is a flowchart of the optimization processing method of the financial risk model provided by the embodiment of the present application Figure 2 ;
[0041] Figure 4 It is a schematic diagram of the structure of the optimization processing device of the financial risk model provided by the embodiment of the present application;
[0042] Figure 5 It is a schematic diagram of the hardware structure of the computer equipment provided by the embodiment of the present application. Detailed Description of the Embodiments
[0043] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts belong to the scope of protection of this application.
[0044] In the technical solutions of this application, the collection, storage, use, processing, transmission, provision, and disclosure of information such as financial data or user data all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.
[0045] It should be noted that in the embodiments of this application, some industry-existing solutions such as certain software, components, and models may be mentioned. They should be regarded as exemplary. The purpose is only to illustrate the feasibility in the implementation of the technical solutions of this application, but it does not mean that the applicant has already or necessarily used this solution.
[0046] A large model refers to a machine learning model with a large number of parameters and a complex computational structure. These models are usually constructed by deep neural networks and have billions or even hundreds of billions of parameters. The design purpose is to improve the expression ability and prediction performance of the model so that it can handle more complex tasks and data. Among them, the large model can also be a specific model such as a financial risk model. In related technologies, when a financial risk report corresponding to a customer needs to be generated, a unified financial risk report is made for the customer through a fixed template and the financial risk report is output as a reference for whether the customer has financial risks. However, in related technologies, making a unified financial risk report for customers through a fixed template results in a single expression of the financial risk report and a high similarity between different financial risk reports.
[0047] To solve the above technical problems, the embodiments of this application propose the following technical concept: The inventor considered multiple historical financial risk reports and corresponding historical customer data obtained, determined corresponding suspicious statements based on each historical financial risk report, determined corresponding target information based on each historical customer data, used each suspicious statement and each target information to train a preset financial risk model in different ways to obtain each trained financial risk model, and evaluated each trained financial risk model to determine the target financial risk model, so that the content of the financial risk report generated by the target financial risk model subsequently is rich and accurate in expression and the similarity between different financial risk reports is reduced.
[0048] Figure 1 It is a schematic diagram of the application scenario of the optimization processing method of the financial risk model provided by the embodiments of this application.
[0049] Such asFigure 1 As shown in the figure, this scenario includes a display terminal 101 and a computer device 102.
[0050] Among them, the display terminal 101 can be a display screen or a terminal such as a personal computer.
[0051] The computer device 102 can be an independent device or a cluster composed of multiple devices.
[0052] The computer device 102 obtains multiple historical financial risk reports and corresponding historical customer data; determines corresponding suspicious statements according to each historical financial risk report; determines corresponding target information according to each historical customer data; screens each target information and each suspicious statement to obtain each screened target information and each screened suspicious statement; trains a preset financial risk model in multiple different ways according to each screened target information and each screened suspicious statement to obtain multiple trained financial risk models; evaluates the reliability, professionalism, and integrity of each trained financial risk model to obtain corresponding evaluation values; determines and outputs a target financial risk model to the display terminal 101 according to each evaluation value. The following uses detailed embodiments for detailed description.
[0053] Figure 2 It is a schematic flowchart of the optimization processing method for the financial risk model provided by the embodiment of the present application Figure 1 , and the execution subject of this embodiment can be Figure 1 the computer device in the embodiment shown, and no special limitation is made here in this embodiment. As Figure 2 shown, this method includes:
[0054] S201: Obtain multiple historical financial risk reports and the corresponding historical customer data for each historical financial risk report.
[0055] In this embodiment, the historical financial risk report is a written discussion on financial risks.
[0056] In this embodiment, the historical customer data is customer information, transaction description, transaction statement, due diligence report, investigation information, transaction warning, and customer transaction characteristics.
[0057] S202: Split each historical financial risk report into statements to obtain corresponding suspicious statements.
[0058] Exemplarily, the suspicious statements are large - amount transfer statements, multiple - transfer statements, or other suspicious statements.
[0059] S203: Perform data analysis on each historical customer data to obtain each analyzed historical customer data.
[0060] In this embodiment, the analyzed historical customers are for transaction interval analysis, upstream / downstream transaction analysis, suspicious period transactions, upstream / downstream transaction period analysis, and upstream / downstream fund usage analysis.
[0061] S204: Determine the target information corresponding to the suspicious statement according to each piece of analyzed historical customer data.
[0062] In this embodiment, the target information can be customer information or transaction information; the target information can include one piece of information or multiple pieces of information.
[0063] Exemplarily, according to any piece of analyzed historical customer data, such as upstream / downstream transaction analysis, determine the transaction information corresponding to the large amount transfer statement.
[0064] S205: Screen each target information and each suspicious statement to obtain each screened target information and each screened suspicious statement.
[0065] In this embodiment, each suspicious statement corresponds to each target information one by one; correspondingly, step S205 specifically includes:
[0066] S2051: Input any target information and the corresponding suspicious statement into a preset financial risk model for processing to determine whether the target information matches the suspicious statement.
[0067] S2052: If it is determined that the target information matches the suspicious statement, determine the target information and the suspicious statement as the screened target information and the screened suspicious statement; if it is determined that the target information does not match the suspicious statement, filter out the target information and the suspicious statement.
[0068] S2053: Traverse the remaining target information and suspicious statements to obtain each screened target information and each screened suspicious statement.
[0069] S206: Train the preset financial risk model in multiple different ways according to each screened target information and each screened suspicious statement to obtain multiple trained financial risk models.
[0070] In this embodiment, multiple different ways can be set according to multiple different requirements.
[0071] S207: Evaluate the reliability, professionalism, and integrity of each trained financial risk model to obtain corresponding evaluation values.
[0072] Specifically, step S207 specifically includes:
[0073] S2071: Evaluate the reliability of each trained financial risk model to obtain the corresponding first evaluation value.
[0074] Specifically, step S2071 specifically includes:
[0075] S20711: Obtain any financial risk report and corresponding customer data.
[0076] S20712: Determine multiple suspicious statements based on the financial risk report, and determine corresponding multiple known information based on the customer data.
[0077] In this embodiment, the suspicious statements may be large-amount transfer statements, multiple transfer statements, or other suspicious statements.
[0078] In this embodiment, the known information may be customer information, transaction information, or other information.
[0079] Exemplarily, there are 3 suspicious statements and 3 known information.
[0080] S20713: Determine whether the trained financial risk model can generate corresponding suspicious statements based on each known information.
[0081] S20714: If it is determined that the trained financial risk model can generate corresponding suspicious statements based on one or more known information, obtain the first quantity corresponding to the one or more known information.
[0082] Exemplarily, if it is determined that the trained financial risk model can generate corresponding suspicious statements based on 2 known information, the obtained corresponding first quantity is 2.
[0083] S20715: Complete the reliability evaluation according to the first quantity to obtain the corresponding first evaluation value.
[0084] Specifically, complete the reliability evaluation according to the first quantity and all quantities of the known information to obtain the corresponding first evaluation value.
[0085] Exemplarily, complete the reliability evaluation according to the first quantity and all quantities of the known information to obtain the corresponding first evaluation value of 2 / 3.
[0086] S2072: Conduct a professional evaluation on each trained financial risk model to obtain the corresponding second evaluation value.
[0087] In this embodiment, the financial risk report further includes suspicious conclusion statements; correspondingly, step S2072 specifically includes:
[0088] S20721: Determine whether the trained financial risk model can generate suspicious conclusion statements based on each suspicious statement.
[0089] S20722: If it is determined that the trained financial risk model can generate suspicious conclusion statements based on one or more suspicious statements, obtain the second quantity corresponding to the one or more suspicious statements.
[0090] Exemplarily, if it is determined that the trained financial risk model can generate suspicious conclusion statements based on 2 suspicious statements, obtain the corresponding second quantity as 2.
[0091] S20723: Complete the professionalism evaluation according to the second quantity to obtain the corresponding second evaluation value.
[0092] Specifically, complete the professionalism evaluation according to the second quantity and the total quantity of suspicious statements to obtain the corresponding second evaluation value.
[0093] Exemplarily, complete the professionalism evaluation according to the second quantity and the total quantity of suspicious statements to obtain the corresponding second evaluation value as 2 / 3.
[0094] S2073: Conduct an integrity evaluation on each trained financial risk model to obtain the corresponding third evaluation value.
[0095] Specifically, step S2073 specifically includes:
[0096] S20731: Determine whether the trained financial risk model can generate each suspicious statement based on the financial risk report.
[0097] S20732: If it is determined that the trained financial risk model can generate one or more suspicious statements based on the financial risk report, obtain the third quantity corresponding to the one or more suspicious statements.
[0098] Exemplarily, if it is determined that the trained financial risk model can generate 2 suspicious statements based on the financial risk report, obtain the corresponding third quantity as 2.
[0099] S20733: Complete the integrity evaluation according to the third quantity to obtain the corresponding third evaluation value.
[0100] Specifically, complete the integrity evaluation according to the third quantity and the total quantity of suspicious statements to obtain the corresponding third evaluation value.
[0101] Exemplarily, complete the integrity evaluation according to the third quantity and the total quantity of suspicious statements to obtain the corresponding third evaluation value as 2 / 3.
[0102] S2074: Determine the corresponding arithmetic mean according to the first evaluation value, the second evaluation value, and the third evaluation value.
[0103] Exemplarily, determine the corresponding arithmetic mean as 2 / 3 according to the first evaluation value, the second evaluation value, and the third evaluation value.
[0104] S2075: Determine the arithmetic mean as the corresponding evaluation value.
[0105] Exemplarily, determining the arithmetic mean as the corresponding evaluation value is 2 / 3.
[0106] S208: Determine the target financial risk model from multiple trained financial risk models according to each evaluation value.
[0107] Exemplarily, the above steps are illustrated by taking the calculation of one evaluation value as an example. Now, according to each evaluation value, determine the trained financial risk model with the highest evaluation value as the target financial risk model.
[0108] S209: Output the target financial risk model, where the financial risk model is used to generate a financial risk report according to the input customer data to be processed.
[0109] In summary, the optimization processing method of the financial risk model provided in this embodiment obtains corresponding suspicious statements by splitting sentences of each obtained historical financial risk report; performs data analysis on each obtained historical customer data to obtain each analyzed historical customer data; determines the target information corresponding to the suspicious statements according to each analyzed historical customer data; screens each target information and each suspicious statement to obtain each screened target information and each screened suspicious statement; trains the preset financial risk model in multiple different ways according to each screened target information and each screened suspicious statement to obtain multiple trained financial risk models; evaluates the reliability, professionalism, and integrity of each trained financial risk model to obtain the corresponding evaluation values; determines the target financial risk model from multiple trained financial risk models according to each evaluation value; outputs the target financial risk model, where the financial risk model is used to generate a financial risk report according to the input customer data to be processed. By using the target financial risk model instead of a fixed template, the content of the subsequent generated financial risk report is rich and accurate, and the similarity between different financial risk reports is reduced.
[0110] Figure 3 Schematic flow of the optimization processing method of the financial risk model provided in the embodiment of the present application Figure 2 . In the embodiment of the present application, on the basis of the Figure 2 provided embodiment, a detailed description is given of the specific implementation method for generating a financial risk report after step S209. As Figure 3 shown, the method includes:
[0111] S301: Obtain multiple historical financial risk reports and the historical customer data corresponding to each historical financial risk report.
[0112] In this embodiment, the discussion about the historical financial risk reports and historical customer data has been elaborated in detail in step S201, and will not be repeated here.
[0113] S302: Split the sentences of each historical financial risk report to obtain corresponding suspicious sentences.
[0114] In this embodiment, the discussion about the suspicious sentences has been elaborated in detail in step S202, and will not be repeated here.
[0115] S303: Conduct data analysis on each historical customer data to obtain the analyzed historical customer data.
[0116] In this embodiment, the discussion about the analyzed historical customer data has been elaborated in detail in step S203, and will not be repeated here.
[0117] S304: Determine the target information corresponding to the suspicious sentences according to the analyzed historical customer data.
[0118] In this embodiment, the discussion about the target information has been elaborated in detail in step S204, and will not be repeated here.
[0119] S305: Screen each target information and each suspicious sentence to obtain the screened target information and the screened suspicious sentences.
[0120] In this embodiment, the discussion about the screening has been elaborated in detail in step S205, and will not be repeated here.
[0121] S306: Train the preset financial risk model in multiple different ways according to the screened target information and the screened suspicious sentences to obtain multiple trained financial risk models.
[0122] S307: Evaluate the reliability, professionalism and integrity of each trained financial risk model to obtain corresponding evaluation values.
[0123] In this embodiment, the discussion about the evaluation has been elaborated in detail in step S207, and will not be repeated here.
[0124] S308: Determine the target financial risk model from multiple trained financial risk models according to the evaluation values.
[0125] S309: Output the target financial risk model, where the financial risk model is used to generate a financial risk report according to the input customer data to be processed.
[0126] S310: Obtain the customer data to be processed corresponding to any customer.
[0127] S311: Input the customer data to be processed into the target financial risk model for processing to generate a financial risk report.
[0128] S312: Output the financial risk report corresponding to the customer.
[0129] In summary, for the optimization processing method of the financial risk model provided in this embodiment, by obtaining the customer data to be processed corresponding to any customer; inputting the customer data to be processed into the target financial risk model for processing to generate a financial risk report; and outputting the financial risk report corresponding to the customer, the content of the generated financial risk report is rich and accurate, and the similarity with other financial risk reports is reduced.
[0130] Figure 4 This is a schematic structural diagram of the optimization processing device for the financial risk model provided in the embodiment of the present application. As Figure 4 shown, the optimization processing device for the financial risk model includes: a first acquisition module 401, a splitting module 402, an analysis module 403, a first determination module 404, a screening module 405, a training module 406, an evaluation module 407, a second determination module 408, and a first output module 409.
[0131] The first acquisition module 401 is configured to acquire a plurality of historical financial risk reports and the historical customer data corresponding to each historical financial risk report;
[0132] The splitting module 402 is configured to split the sentences of each historical financial risk report to obtain corresponding suspicious sentences;
[0133] The analysis module 403 is configured to perform data analysis on each historical customer data to obtain each analyzed historical customer data;
[0134] The first determination module 404 is configured to determine the target information corresponding to the suspicious sentences according to the analyzed historical customer data;
[0135] The screening module 405 is configured to screen each target information and each suspicious sentence to obtain each screened target information and each screened suspicious sentence;
[0136] The training module 406 is configured to train the preset financial risk model in multiple different ways according to the screened target information and the screened suspicious sentences to obtain a plurality of trained financial risk models;
[0137] The evaluation module 407 is configured to evaluate the reliability, professionalism, and integrity of each trained financial risk model to obtain corresponding evaluation values;
[0138] The second determination module 408 is configured to determine a target financial risk model from the multiple trained financial risk models according to each evaluation value;
[0139] The first output module 409 is configured to output the target financial risk model, where the financial risk model is used to generate a financial risk report according to the input customer data to be processed.
[0140] In a possible implementation manner, each of the suspicious statements corresponds to each of the target information; correspondingly, the screening module 405 specifically includes:
[0141] The processing unit is configured to input any target information and the corresponding suspicious statement into the preset financial risk model for processing to determine whether the target information matches the suspicious statement;
[0142] The filtering unit is configured to, if it is determined that the target information matches the suspicious statement, determine the target information and the suspicious statement as the screened target information and the screened suspicious statement; if it is determined that the target information does not match the suspicious statement, filter out the target information and the suspicious statement;
[0143] The traversing unit is configured to traverse the remaining target information and suspicious statements to obtain the screened target information and the screened suspicious statements.
[0144] In a possible implementation manner, the evaluation module 407 specifically includes:
[0145] The first evaluation unit is configured to perform a reliability evaluation on each trained financial risk model to obtain a corresponding first evaluation value;
[0146] The second evaluation unit is configured to perform a professionalism evaluation on each trained financial risk model to obtain a corresponding second evaluation value;
[0147] The third evaluation unit is configured to perform a completeness evaluation on each trained financial risk model to obtain a corresponding third evaluation value;
[0148] The first determination unit is configured to determine a corresponding arithmetic mean according to the first evaluation value, the second evaluation value, and the third evaluation value;
[0149] The second determination unit is configured to determine the arithmetic mean as the corresponding evaluation value.
[0150] In a possible implementation manner, the first evaluation unit specifically includes:
[0151] The first acquisition unit is configured to acquire any financial risk report and the corresponding customer data;
[0152] A determination unit, configured to determine a plurality of suspicious statements according to the financial risk report, and determine a corresponding plurality of known information according to the customer data;
[0153] A judgment unit, configured to judge whether the trained financial risk model can generate corresponding suspicious statements according to each known information;
[0154] A second acquisition unit, configured to, if it is determined that the trained financial risk model can generate corresponding suspicious statements according to one or more known information, acquire a first quantity corresponding to the one or more known information;
[0155] An evaluation unit, configured to complete the reliability evaluation according to the first quantity to obtain a corresponding first evaluation value.
[0156] In a possible implementation manner, the financial risk report further includes a suspicious conclusion statement; correspondingly, the second evaluation unit specifically includes:
[0157] A judgment unit, configured to judge whether the trained financial risk model can generate the suspicious conclusion statement according to each suspicious statement;
[0158] An acquisition unit, configured to, if it is determined that the trained financial risk model can generate the suspicious conclusion statement according to one or more suspicious statements, acquire a second quantity corresponding to the one or more suspicious statements;
[0159] An evaluation unit, configured to complete the professionalism evaluation according to the second quantity to obtain a corresponding second evaluation value.
[0160] In a possible implementation manner, the third evaluation unit specifically includes:
[0161] A judgment unit, configured to judge whether the trained financial risk model can generate each suspicious statement according to the financial risk report;
[0162] An acquisition unit, configured to, if it is determined that the trained financial risk model can generate one or more suspicious statements according to the financial risk report, acquire a third quantity corresponding to the one or more suspicious statements;
[0163] An evaluation unit, configured to complete the integrity evaluation according to the third quantity to obtain a corresponding third evaluation value.
[0164] In a possible implementation manner, the apparatus further includes:
[0165] A second acquisition module, configured to acquire to-be-processed customer data corresponding to any customer;
[0166] A processing module for inputting the customer data to be processed into the target financial risk model for processing to generate a financial risk report;
[0167] A second output module for outputting the financial risk report corresponding to the customer.
[0168] The device provided in this embodiment can be used to execute the technical solutions of the above method embodiments. The implementation principles and technical effects are similar, and will not be elaborated here.
[0169] Figure 5 It is a schematic hardware structure diagram of a computer device provided in an embodiment of the present application. As Figure 5 shown, the computer device in this embodiment includes: a processor 501 and a memory 502; the memory stores computer execution instructions; at least one processor executes the computer execution instructions stored in the memory, so that at least one processor executes the optimization processing method of the financial risk model as above.
[0170] Optionally, the memory 502 can be either independent or integrated with the processor 501.
[0171] When the memory 502 is independently provided, the computer device further includes a bus 503 for connecting the memory 502 and the processor 501.
[0172] An embodiment of the present application further provides a computer storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the optimization processing method of the financial risk model as above is implemented.
[0173] An embodiment of the present application further provides a computer program product, including a computer program. When the computer program is executed by a processor, the optimization processing method of the financial risk model as above is implemented.
[0174] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of devices or modules can be in electrical, mechanical or other forms.
[0175] The module described as a separation component may or may not be physically separated. The component shown as a module may or may not be a physical unit, that is, it may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to implement the solution of this embodiment.
[0176] In addition, in each embodiment of this application, each functional module can be integrated in a processing unit, or each module can exist physically alone, or two or more modules can be integrated in one unit. The unit formed by the above modules can be implemented in the form of hardware or in the form of a combination of hardware and software functional units.
[0177] The integrated module implemented in the form of a software functional module can be stored in a computer-readable storage medium. The above software functional module is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute some steps of the methods described in each embodiment of this application.
[0178] It should be understood that the above processor can be a Central Processing Unit (CPU for short), and can also be other general-purpose processors, Digital Signal Processors (DSP for short), Application Specific Integrated Circuits (ASIC for short), etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly embodied as being executed and completed by a hardware processor, or executed and completed by a combination of hardware and software modules in the processor.
[0179] The memory may include high-speed RAM memory, and may also include non-volatile storage NVM, such as at least one disk memory, and can also be a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk, or an optical disc, etc.
[0180] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the buses in the drawings of this application are not limited to only one bus or one type of bus.
[0181] The above storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic memory, flash memory, a magnetic disk or an optical disc. The storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer.
[0182] An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the storage medium can also exist as discrete components in an electronic device or a master control device.
[0183] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including the above method embodiments; and the foregoing storage medium includes various media that can store program codes, such as ROM, RAM, magnetic disks, or optical discs.
[0184] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.
Claims
1. A method for optimizing a financial risk model, characterized in that: Applicable to computer equipment, including: Obtain multiple historical financial risk reports and historical customer data corresponding to each historical financial risk report; Split each historical financial risk report into sentences to obtain the corresponding suspicious sentences; Performing data analysis on each historical customer data to obtain each analyzed historical customer data; Determining target information corresponding to the suspicious statement according to the analyzed historical customer data; Screening each target information and each suspicious sentence to obtain each screened target information and each screened suspicious sentence; Training the preset financial risk model in a plurality of different ways according to the screened target information and the screened suspicious statements to obtain a plurality of trained financial risk models; Evaluate the reliability, professionalism and integrity of each trained financial risk model to obtain the corresponding evaluation value; Determining a target financial risk model from the plurality of trained financial risk models according to each evaluation value; The target financial risk model is output, wherein the financial risk model is used to generate a financial risk report based on the input customer data to be processed.
2. The method according to claim 1, characterized in that wherein each of the suspicious sentences corresponds to each of the target information; Accordingly, the target information and the suspicious statements are screened to obtain the screened target information and the screened suspicious statements, including: Input any target information and the corresponding suspicious statement into the preset financial risk model for processing to determine whether the target information matches the suspicious statement; If it is determined that the target information matches the suspicious statement, the target information and the suspicious statement are determined as the filtered target information and the filtered suspicious statement; if it is determined that the target information does not match the suspicious statement, the target information and the suspicious statement are filtered out; The remaining target information and suspicious sentences are traversed to obtain the filtered target information and the filtered suspicious sentences.
3. The method according to claim 1, characterized in that: The reliability, professionalism and integrity evaluation of each trained financial risk model to obtain a corresponding evaluation value includes: Performing reliability evaluation on each trained financial risk model to obtain a corresponding first evaluation value; Performing a professional evaluation on each trained financial risk model to obtain a corresponding second evaluation value; Conduct integrity evaluation on each trained financial risk model to obtain a corresponding third evaluation value; Determine a corresponding arithmetic mean according to the first evaluation value, the second evaluation value and the third evaluation value; The arithmetic mean is determined as the corresponding evaluation value.
4. The method according to claim 3, characterized in that: The reliability evaluation of each trained financial risk model to obtain a corresponding first evaluation value includes: Obtain any financial risk report and corresponding customer data; Determining a plurality of suspicious statements according to the financial risk report, and determining a plurality of corresponding known information according to the customer data; Determine whether the trained financial risk model can generate corresponding suspicious statements based on each known information; If it is determined that the trained financial risk model can generate a corresponding suspicious statement according to one or more known information, obtaining a first quantity corresponding to the one or more known information; The reliability evaluation is completed according to the first quantity to obtain a corresponding first evaluation value.
5. The method according to claim 4, characterized in that The financial risk report also includes suspicious conclusion statements; Accordingly, the professional evaluation of each trained financial risk model to obtain a corresponding second evaluation value includes: Determining whether the trained financial risk model can generate the suspicious conclusion statement based on each suspicious statement; If it is determined that the trained financial risk model can generate the suspicious conclusion statement according to one or more suspicious statements, obtaining a second quantity corresponding to the one or more suspicious statements; Complete the professional evaluation according to the second quantity to obtain a corresponding second evaluation value.
6. The method according to claim 5, characterized in that The integrity evaluation of each trained financial risk model to obtain a corresponding third evaluation value includes: Determining whether the trained financial risk model can generate each suspicious statement according to the financial risk report; If it is determined that the trained financial risk model can generate one or more suspicious statements according to the financial risk report, obtaining a third quantity corresponding to the one or more suspicious statements; The integrity evaluation is completed according to the third quantity to obtain a corresponding third evaluation value.
7. The method according to any one of claims 1 to 6, characterized in that: After outputting the target financial risk model, the method further includes: Get the pending customer data corresponding to any customer; Inputting the to-be-processed customer data into the target financial risk model for processing to generate a financial risk report; Output the financial risk report corresponding to the client.
8. An optimization processing device for a financial risk model, characterized in that: Applicable to computer equipment, including: A first acquisition module is used to acquire multiple historical financial risk reports and historical customer data corresponding to each historical financial risk report; A splitting module is used to split the statements of each historical financial risk report to obtain the corresponding suspicious statements; An analysis module, used for performing data analysis on each historical customer data to obtain each analyzed historical customer data; A first determination module, configured to determine target information corresponding to the suspicious statement according to the analyzed historical customer data; A screening module, used for screening each target information and each suspicious sentence to obtain each screened target information and each screened suspicious sentence; A training module, used for training the preset financial risk model in a variety of different ways according to the screened target information and the screened suspicious statements, so as to obtain a plurality of trained financial risk models; Evaluation module, used to evaluate the reliability, professionalism and integrity of each trained financial risk model to obtain the corresponding evaluation value; A second determination module, configured to determine a target financial risk model from the plurality of trained financial risk models according to each evaluation value; The first output module is used to output the target financial risk model, wherein the financial risk model is used to generate a financial risk report based on the input customer data to be processed.
9. A computer device, characterized in that: include: at least one processor and memory; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the optimization processing method of the financial risk model as described in any one of claims 1 to 7.
10. A computer storage medium, characterized in that: The computer storage medium stores computer execution instructions. When the processor executes the computer execution instructions, the optimization processing method of the financial risk model as described in any one of claims 1 to 7 is implemented.