Method, system and device for generating and deploying risk control strategy and storage medium

Abstract

The invention provides a method, system and device for generating and deploying risk control strategies and a storage medium, and relates to the technical field of computer application. The method comprises the following steps: deploying a to-be-tested strategy in a laboratory environment; generating a first test result in the laboratory environment according to historical data in a preset range in the production environment and the to-be-tested strategy; when the first test result meets a preset standard, generating a second test result according to real-time production data in the production environment and the to-be-tested strategy; when the second test result meets the preset standard, generating a third test result according to the online data and the to-be-tested strategy in a preset proportion in the production environment; and when the third test result meets the preset standard, deploying the to-be-tested strategy to the production environment. By adopting the method disclosed by the invention, the strategy can be progressively operated step by step in three stages in a laboratory environment, the result is analyzed step by step, and the strategy is adjusted, so that the strategy effect is accurately controlled, and the risk loss is reduced.

Description

Technical Field

[0001] The present invention relates to the field of computer application technology, and more specifically, to a method, system, device, and storage medium for generating and deploying risk control strategies. Background Art

[0002] With the continuous development and evolution of intelligent risk control, the industry's requirements for the precision and granularity of risk control strategies are becoming increasingly stringent. However, the current simple case testing model dominated by manual testing is no longer able to meet the increasingly complex business needs, exposing the following limitations:

[0003] 1. Manual testing can only barely handle basic connectivity testing. It is unable to cope with the complex multi-scenario testing involved in risk control strategies and is difficult to achieve comprehensive coverage.

[0004] 2. Risk control strategies themselves involve numerous variables, each of which is complex and mutually influential, undoubtedly adding immense difficulty and challenges to manual testing.

[0005] 3. Statistical testing requires a large amount of nearly real data for input and verification. However, manual testing has obvious shortcomings in terms of data volume, test efficiency, and result accuracy, and cannot meet the needs of statistical testing.

[0006] It can be seen that the existing manual testing methods cannot meet the increasingly precise granularity requirements for risk control strategies, and there is an urgent need to seek more efficient and intelligent testing methods. Summary of the Invention

[0007] In order to solve the problems or at least part of the problems existing in the above-mentioned prior art, the embodiments of the present invention provide a method, system, storage medium, device and computer program product for generating and deploying risk control strategies. The strategy is run step by step through three stages in a laboratory environment, and the results are gradually analyzed and the strategy is adjusted, so as to accurately control the effect of the strategy, reduce risk losses, and ensure that the business achieves expectations.

[0008] According to a first aspect of the present invention, an embodiment of the present invention provides a method for generating and deploying risk control strategies, which includes: obtaining a strategy to be tested and deploying the strategy to be tested in a laboratory environment; obtaining historical data within a preset range in a production environment, and generating a first test result in the laboratory environment based on the historical data and the strategy to be tested; when the first test result meets the preset standard, obtaining real-time production data in the production environment, and generating a second test result in the laboratory environment based on the real-time production data and the strategy to be tested; when the second test result meets the preset standard, obtaining a preset proportion of online data in the production environment, and generating a third test result in the laboratory environment based on the online data and the strategy to be tested; when the third test result meets the preset standard, deploying the strategy to be tested to the production environment.

[0009] According to the above-described embodiment of the present invention, the strategy to be tested is run in a laboratory environment using production data. Through a three-stage progressive process, the results are gradually analyzed and the strategy is adjusted. This allows for precise control of the strategy's effectiveness, reduces risk losses, and ensures that business expectations are met. Furthermore, running the strategy with a large amount of real data can more realistically and accurately reflect the results of the strategy after it is launched. This addresses the problem that manual testing of strategies can only simply test connectivity and cannot address scenarios where a large number of variables interact. It also improves the accuracy of strategy iteration and reduces asset losses caused by inaccurate risk control.

[0010] In some embodiments of the present invention, the method for generating and deploying risk control strategies further includes: when any one of the first test result, the second test result, and the third test result does not meet the preset standard, optimizing and updating the strategy to be tested.

[0011] In some embodiments of the present invention, the method for generating and deploying risk control strategies also includes: obtaining first comparison data between the first test result and the historical results in the production environment based on preset comparison variables; obtaining second comparison data between the second test result and the line results in the production environment based on the preset comparison variables; and extracting variable record information of the third test result based on preset monitoring variables.

[0012] In some embodiments of the present invention, the first comparison data and the second comparison data include: pass rate, rejection rate, and difference variables; the method for generating and deploying risk control strategies also includes: generating a test result report based on the third test result, and the test result report includes: pass rate, rejection rate, and distribution result statistical information of variable values.

[0013] According to the second aspect of the present invention, an embodiment of the present invention provides a system for generating and deploying risk control strategies, which includes: a strategy acquisition module, used to acquire the strategy to be tested and deploy the strategy to be tested in a laboratory environment; a first test module, used to acquire historical data within a preset range in the production environment, and generate a first test result in the laboratory environment based on the historical data and the strategy to be tested; a second test module, used to acquire real-time production data in the production environment when the first test result meets the preset standard, and generate a second test result in the laboratory environment based on the real-time production data and the strategy to be tested; a third test module, used to acquire a preset proportion of online data in the production environment when the second test result meets the preset standard, and generate a third test result in the laboratory environment based on the online data and the strategy to be tested; a strategy online module, used to deploy the strategy to be tested to the production environment when the third test result meets the preset standard.

[0014] According to the above-described embodiment of the present invention, the strategy to be tested is run in a laboratory environment using production data. Through a three-stage progressive process, the results are gradually analyzed and the strategy is adjusted. This allows for precise control of the strategy's effectiveness, reduces risk losses, and ensures that business expectations are met. Furthermore, running the strategy with a large amount of real data can more realistically and accurately reflect the results of the strategy after it is launched. This addresses the problem that manual testing of strategies can only simply test connectivity and cannot address scenarios where a large number of variables interact. It also improves the accuracy of strategy iteration and reduces asset losses caused by inaccurate risk control.

[0015] In some embodiments of the present invention, when any one of the first test result, the second test result, and the third test result does not meet the preset standard, the strategy acquisition module optimizes the strategy to be tested and updates the strategy to be tested.

[0016] In some embodiments of the present invention, the first test module is further used to obtain first comparison data between the first test result and the historical results in the production environment based on preset comparison variables; the second test module is further used to obtain second comparison data between the second test result and the line results in the production environment based on the preset comparison variables; the third test module is further used to extract variable recording information of the third test result based on preset monitoring variables.

[0017] In some embodiments of the present invention, the first comparison data and the second comparison data include: pass rate, rejection rate, and difference variables; the first test module is also used to generate a first comparison result report based on the first comparison data; the second test module is also used to generate a second comparison result report based on the second comparison data; the third test module is also used to generate a test result report based on the third test result, and the test result report includes: pass rate, rejection rate, and distribution result statistical information of variable values.

[0018] According to the third aspect of the present invention, an embodiment of the present invention provides a computer-readable storage medium having computer-readable instructions stored thereon. When the computer-readable instructions are executed by a processor, the computer performs the following operations: the operations include the steps included in the method for generating and deploying risk control strategies as described in any of the above embodiments.

[0019] According to the fourth aspect of the present invention, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory is used to store one or more computer-readable instructions, wherein the one or more computer-readable instructions, when executed by the processor, can implement the method for generating and deploying risk control strategies as described in any of the above embodiments.

[0020] According to a fifth aspect of the present invention, an embodiment of the present invention provides a computer program product comprising a computer program, which, when executed by a processor, implements the method for generating and deploying risk control strategies as described in any one of the above embodiments.

[0021] As can be seen from the foregoing, the method, system, storage medium, device, and computer program product for generating and deploying risk control strategies provided by the embodiments of the present invention run the strategy to be tested in a laboratory environment, using production data. Through a three-stage progressive process, the results are gradually analyzed and the strategy is adjusted. This allows for precise control of the strategy's effectiveness, reduces risk losses, and ensures that business expectations are met. Furthermore, by running the strategy using a large amount of real data, the results of the strategy after launch can be more realistically and accurately reflected, while also improving the accuracy of strategy iteration. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 is a flowchart of a method for generating and deploying a risk control strategy according to embodiment 1 of the present invention;

[0023] Figure 2 is a flowchart of a method for generating and deploying a risk control strategy according to embodiment 3 of the present invention;

[0024] Figure 3 2. It is a schematic diagram of the task execution process in the historical data backtracking phase according to an embodiment of the present invention;

[0025] Figure 4 This is a schematic diagram of the task execution process of the real-time data accompanying running stage according to an embodiment of the present invention;

[0026] Figure 5 This is a schematic diagram of the task execution process in the online data diversion stage according to an embodiment of the present invention;

[0027] Figure 6 2 is a schematic diagram of the architecture of a system for generating and deploying risk control strategies according to embodiment 4 of the present invention. DETAILED DESCRIPTION

[0028] The various aspects of the present invention are described in detail below in conjunction with the accompanying drawings and specific embodiments. Among them, well-known modules, units and their connections, links, communications or operations are not shown or described in detail. In addition, the described features, architectures or functions can be combined in any manner in one or more embodiments. It should be understood by those skilled in the art that the various embodiments described below are only for illustration and are not intended to limit the scope of protection of the present invention. It can also be easily understood that the modules or units or processing methods in the various embodiments described herein and shown in the drawings can be combined and designed in various different configurations.

[0029] [Example 1]

[0030] Figure 1 4 is a flow chart of a method for generating and deploying risk control strategies according to embodiment 1 of the present invention.

[0031] like Figure 1 As shown, in embodiment 1 of the present invention, the method for generating and deploying risk control strategies may include at least: step S11, step S12, step S13, step S14, and step S15. These steps are described in detail below.

[0032] In step S11, a policy to be tested is obtained and deployed in a laboratory environment, wherein the laboratory environment is a laboratory scenario that is highly similar to a production environment.

[0033] In step S12, historical data within a preset range in the production environment is acquired, and a first test result is generated in the laboratory environment according to the historical data and the strategy to be tested.

[0034] In step S13, if the first test result meets the preset criteria, real-time production data in the production environment is obtained, and a second test result is generated in the laboratory environment based on the real-time production data and the strategy to be tested. If the first test result does not meet the preset criteria, the strategy to be tested is optimized and updated.

[0035] In some embodiments, the preset criteria include but are not limited to whether the pass rate, rejection rate, hit code value, etc. meet the requirements.

[0036] In step S14, if the second test result meets the preset criteria, a preset proportion of online data in the production environment is obtained, and a third test result is generated in the laboratory environment based on the online data and the strategy to be tested. If the second test result does not meet the preset criteria, the strategy to be tested is optimized and updated.

[0037] In step S15, when the third test result meets the preset standard, the strategy to be tested is deployed to the production environment. When the third test result does not meet the preset standard, the strategy to be tested is optimized and updated.

[0038] By using the above-mentioned method for generating and deploying risk control strategies of Example 1 of the present invention, the strategy to be tested is run in a laboratory environment similar to the production environment, using production data, and progressing step by step through three stages. The results are gradually analyzed and the strategy is adjusted, thereby accurately controlling the effectiveness of the strategy, reducing risk losses, and ensuring that the business meets expectations. In addition, using a large amount of real data to run the strategy can more realistically and accurately reflect the results after the strategy is launched. This can solve the problem that manual testing of strategies can only simply test connectivity and cannot meet the scenarios where a large number of variables affect each other. At the same time, it improves the accuracy of strategy iteration and reduces asset losses caused by inaccurate risk control.

[0039] In a further embodiment, the method for generating and deploying a risk control strategy further includes: obtaining first comparative data between the first test result and historical results in the production environment based on a preset comparative variable; obtaining second comparative data between the second test result and online results in the production environment based on the preset comparative variable; and extracting variable record information of the third test result based on a preset monitoring variable. The first comparative data and the second comparative data may include, but are not limited to, one or more of the following: pass rate, rejection rate, and difference variable.

[0040] In a further embodiment, the method for generating and deploying risk control strategies also includes: generating a comparison result report based on the first comparison data and the second comparison data; generating a test result report based on the third test result, and the test result report includes: pass rate, rejection rate and distribution result statistical information of variable values.

[0041] [Example 2]

[0042] Embodiment 2 of the present invention provides a specific example of generating an online strategy. In this example, a method for generating and deploying a risk control strategy may include the following steps:

[0043] Step 1: Configure the experiment phase, product type, and experimental data time range to generate the task.

[0044] In some embodiments, the experimental phase includes a historical data backtracking phase, a real-time data running phase, and an online data diversion phase. In an exemplary embodiment, the configured product type is a certain business type, for example, the product type of JD.com refers to an incoming application related to JD.com.

[0045] The experimental data time range specifies the time period of historical data to be used for the experiment. This configuration allows you to select historical data within a specific time period as the data for strategy testing. Once the aforementioned task parameters (experimental phase, product type, and experimental data time range) are configured, one or more tasks are generated to perform strategy testing based on the selected experimental data.

[0046] Step 2: Configure the variables that need to be compared or monitored (such as pass rate, hit code value, etc.) in the task generated in step 1. This is the data used to generate a report at the end of the experiment.

[0047] Step 3: Deploy the policy to be tested in a lab environment similar to the production environment.

[0048] Step 4: Execute the task generated in step 1.

[0049] Step 5: Acquire test data from different stages and input the test data into the laboratory environment where the strategy to be tested is deployed for strategy testing. Specifically, in the historical data backtracking stage, query the incoming data within the task setting range and the corresponding third-party data of the strategy to be tested to conduct experimental strategy testing; further, in the real-time data accompanying stage, send the online real-time data that has passed the strategy and the corresponding third-party data to the laboratory strategy to be tested for testing; further, in the online data diversion stage, send the online incoming data to the laboratory strategy to be tested in a preset ratio for testing and processing. At the same time, the required third-party data is obtained by directly calling the third-party interface.

[0050] Among them, the application data is data used for experiments. For example, the application data may include basic customer information (name, ID card, phone number, etc.) and business-related information (product number, channel number, credit amount, interest rate, etc.); the third-party data corresponding to the strategy refers to the third-party data queried during the current customer's historical approval. For example, the third-party data may include credit data queried by calling a third-party interface (such as anti-fraud scores, credit scores, etc.); the online real-time completion strategy means that the application data first obtains the final decision result through the approval strategy online, and then conducts experiments after the application process is completed.

[0051] Step 6: Comparative analysis of results. After each piece of incoming data is processed by the strategy to be tested, the data comparison results are generated in real time and recorded in the database. Specifically, in the historical data backtracking stage, according to the comparison variables configured in the task, the test results of the historical data backtracking stage are obtained and compared with the historical results, and recorded in the database; further, in the real-time data accompanying stage, according to the comparison variables configured in the task, the test results of the real-time data accompanying stage are obtained and compared with the current results, and recorded in the database; further, in the online data diversion stage, according to the monitoring variables configured in the task, the variables of the test results of the online data diversion stage are extracted and recorded in the database.

[0052] Among them, the configured comparison variable refers to the variable in the experimental result data that needs to be compared with the historical result data in the experiment. By comparing whether the values ​​of the comparison variable in the experimental results and the historical results are consistent, it is judged whether the strategy to be tested meets the requirements; historical results refer to the data used to execute the experiment, which have been obtained by online execution; current results refer to the results of online execution of the application data executed in the experimental environment after the online execution is completed; configured monitoring variables refer to the variables that need to be monitored and used to generate statistical reports when the experiment is completed. Business personnel can analyze whether the strategy to be tested meets expectations through the distribution of variable values ​​(corresponding to the monitoring variables) in the statistical report; the variables of the test results in the online data diversion stage, that is, the experimental variables, refer to the results of the experiment output at the end of the application (for example, hit code value, rejection reason code value, credit limit, interest rate, etc.), and the configured monitoring variables are a subset of the variables of the test results.

[0053] Step 7: Export the test result report after the experimental task is completed. Specifically, in the historical data backtracking stage, a comparison result report is generated, wherein the comparison result report includes but is not limited to one or more of the following: pass rate, rejection rate and difference variables (variables that are different between test results and historical results); further, in the real-time data accompanying stage, a comparison result report is generated, wherein the comparison result report includes but is not limited to one or more of the following: pass rate, rejection rate and percentage of difference variables and other statistics and comparison data; further, in the online data diversion stage, a test result report is exported, wherein the test result report includes but is not limited to: pass rate, rejection rate and distribution result statistics of variable values. Among them, the pass rate is the percentage of experimental approvals that passed (number of passes / total number of experiments); the rejection rate is the percentage of experimental approvals that rejected (number of rejections / total number of experiments); the difference variables are the configured comparison variables and the percentage of experimental results that are inconsistent with historical results (number of differences / total number of experiments); the distribution result statistics of variable values ​​refer to the number of times each value of the statistical variable is hit after the experiment is completed, and the distribution of variable values ​​is drawn.

[0054] Step 8: Analyze and process the test results of the strategy to be tested, comprehensively evaluate the strategy effect, and optimize it according to the test results. If the test results do not meet the preset standards, optimize and update the strategy and retest it until the strategy to be tested meets the online requirements.

[0055] Step 9: Release the policy to be tested. When the policy to be tested meets the launch requirements, it will be released and deployed to the production environment for production.

[0056] The above-mentioned method for generating and deploying risk control strategies of Example 2 of the present invention is adopted. By creating a laboratory environment similar to the production environment, the strategy to be tested is run in the laboratory environment. Using production data, the system progresses step by step through three stages, gradually analyzing the results and adjusting the strategy to achieve the expected business results. In addition, by feeding a large amount of real data, the results after the strategy is launched can be more realistic and accurate, solving the problem that the current manual testing strategy can only simply test connectivity and cannot meet the scenarios where a large number of variables affect each other. At the same time, it also improves the accuracy of strategy iteration and reduces asset losses caused by inaccurate risk control.

[0057] [Example 3]

[0058] Example 3 of the present invention provides a specific example of generating an online strategy based on Java 8 or above. Figures 2 to 5 As shown, in this example, the method for generating and deploying a risk control strategy may include the following steps:

[0059] Step 1: Configure the historical data backtracking task on the page. This includes the following steps: 1.1. Create a historical data backtracking task on the interactive interface; 1.2. Select the time period and product; 1.3. Configure the variables to be compared.

[0060] Step 2: On the interactive page, deploy the experimental strategy (the strategy to be tested) and publish it to the lab environment. This includes the following steps: 2.1. Upload the strategy to be tested; 2.2. Deploy the tested strategy to the lab environment; 2.3. Publish the strategy to be tested to the lab environment. The lab environment is highly similar to the production environment and provides a foundation for strategy testing.

[0061] Step 3: Execute the historical data backtracking phase. This specifically includes the following steps: 3.1. Start the task and retrieve the incoming data one by one according to the configured time interval; 3.2. From the online third-party data cache, retrieve the variable data corresponding to the incoming data; 3.3. Enter the acquired incoming data and third-party variable data into the strategy to be tested for execution; 3.4. Compare the test results with the historical results based on the configured comparison variables; 3.5. Save the comparison results to the database.

[0062] Step 4, generate a test report and analyze the results. Specifically include the following steps: 4.1, download the report; 4.2, count all the data results of the corresponding task batch; 4.3, the statistical dimensions include pass rate, rejection rate, decision variable result distribution and the difference ratio with historical comparison results, etc.; 4.4, generate a report for the comparison results; 4.5, after comparing the report, multi-party evaluation and optimization, when the test results do not meet the preset standards, perform strategy optimization and update and retest until it passes and enters the next stage. In an exemplary embodiment, the decision variable result distribution can include the percentage of decision passes and rejections, as well as the distribution range of some hit code values; the difference ratio with historical comparison results refers to the comparison of the experimental result data obtained by the laboratory environment test of the same piece of data with the historical result data (for example, pass rate, hit code value, etc.) and the statistical proportion of inconsistent results (number of differences / total number of experiments).

[0063] During the historical data retrospective phase, various historical data from the production environment are collected, strategies are applied, and key indicators are recorded. After generating a comparison report, multi-faceted evaluation and optimization are conducted. When the test results do not meet the standards, the strategy is optimized and updated and retested until it passes and enters the next phase. This allows the ability of the strategy to be tested to cope with complex past business operations to be verified.

[0064] Step 5: Enter the real-time data comparison phase. This includes the following steps: 5.1. Create a real-time data comparison task in the interactive interface; 5.2. Select the time range and product; 5.3. Configure the variables to be compared.

[0065] Step 6: Execute the tasks during the real-time data phase. This specifically includes the following steps: 6.1. Start execution. Send a copy of the online application data to the lab environment via RabbitMQ (an open source message broker middleware); 6.2. Obtain the variable data corresponding to the application from the online third-party data cache; 6.3. Enter the obtained application data and third-party variable data into the strategy to be tested for execution; 6.4. Compare the test results with the online results based on the configured comparison variables; 6.5. Save the comparison results to the database.

[0066] Step 7: Generate a test report and analyze the results. The specific implementation steps are the same as those in Step 4 above and will not be repeated here.

[0067] During the real-time data run-in phase, production data is collected in real time so that the strategy to be tested is processed synchronously with the current strategy. After multi-dimensional comparative analysis, evaluation and optimization are performed. When the test results do not meet the standards, the strategy is optimized and updated and retested until it passes and enters the next phase. In this way, the effectiveness of the strategy to be tested in real-time business can be verified.

[0068] Step 8: Enter the online data diversion phase. This includes the following steps: 8.1. Create an online data diversion task in the interactive interface; 8.2. Select a time range and product; 8.3. Configure the variables to be monitored and the traffic ratio allocated to the lab.

[0069] Step 9: Execute the online data diversion phase. This includes the following steps: 9.1. Start the task, and send the online incoming data to the lab environment via RabbitMQ according to the configured preset ratio; 9.2. Directly query the variable data of the third party; 9.3. Put the acquired incoming data and third-party variable data into the strategy to be tested for execution; 9.4. Parse and store the test results according to the configured monitoring variables;

[0070] Step 10: Generate a test report and analyze the results. This includes the following steps: 10.1. Download the report; 10.2. Compile statistics for all incoming data for the corresponding task batch; 10.3. Statistical dimensions include pass rate, rejection rate, and distribution of decision variable results; 10.4. Generate a report based on the comparison results; 10.5. After comparing the reports, conduct multi-faceted evaluation and optimization. If the test results do not meet the preset standards, optimize the strategy and update it, then retest until it passes the test and is released to the production environment.

[0071] During the online data diversion stage, we sample online data for strategy testing to comprehensively evaluate its effectiveness and stability. We optimize based on the results. When the test results do not meet the standards, we optimize and update the strategy and retest it until the strategy meets the requirements and goes online. This allows us to accurately control the effectiveness of the strategy, reduce risk losses, and ensure that the business meets expectations.

[0072] Step 11: The tested policy that has passed the test will be released to the production environment and replace the original production policy.

[0073] The above-mentioned method for generating and deploying risk control strategies of Example 3 of the present invention is adopted. In the historical data backtracking stage, the historical data feeding strategy for the time period to be tested is selected, the strategy test results are compared with the historical strategy results in multiple dimensions, and key indicators are recorded. After generating the comparison report, multi-party evaluation and optimization are performed. When the test results do not meet the standards, the strategy is optimized and updated and retested; in the real-time data running-in stage, production data is obtained in real time, so that the test strategy and the current strategy are processed approximately synchronously (the strategy to be tested is executed after the current strategy is processed), and the optimization is evaluated after multi-dimensional comparative analysis. When the test results do not meet the standards, the strategy is optimized and updated and retested; in the online data diversion stage, the online data test strategy is sampled, and part of the online data traffic is allocated to the laboratory environment. The strategy effect is comprehensively evaluated and optimized according to the results. When the test results do not meet the standards, the strategy is optimized and updated and retested until the requirements are met and the strategy is put online, so as to accurately control the strategy effect, reduce risk losses, and ensure that the business meets expectations.

[0074] [Example 4]

[0075] Figure 6 2 is a schematic diagram of the architecture of a system for generating and deploying risk control strategies according to embodiment 4 of the present invention.

[0076] like Figure 6 As shown, the system for generating and deploying risk control strategies includes: a strategy acquisition module 210 , a first testing module 220 , a second testing module 230 , a third testing module 240 , and a strategy online module 250 .

[0077] The policy acquisition module 210 is used to acquire the policy to be tested and deploy the policy to be tested in a laboratory environment.

[0078] The first testing module 220 is used to obtain historical data within a preset range in a production environment, and generate a first test result in a laboratory environment according to the historical data and the strategy to be tested.

[0079] In some embodiments, the first test module is further configured to obtain first comparison data between the first test result and historical results in the production environment based on a preset comparison variable. Further, the first test module is further configured to generate a first comparison result report based on the first comparison data.

[0080] The second test module 230 is configured to obtain real-time production data in the production environment when the first test result meets a preset standard, and generate a second test result in a laboratory environment according to the real-time production data and the strategy to be tested.

[0081] In some embodiments, the second testing module is further configured to obtain second comparison data between the second test result and the line result in the production environment based on the preset comparison variable. Further, the second testing module is further configured to generate a second comparison result report based on the second comparison data.

[0082] The third test module 240 is configured to obtain a preset proportion of online data in the production environment when the second test result meets the preset standard, and generate a third test result in the laboratory environment based on the online data and the strategy to be tested.

[0083] In some embodiments, the third test module is further configured to extract variable record information of the third test result based on preset monitoring variables. Furthermore, the third test module is further configured to generate a test result report based on the third test result, the test result report including statistical information on a pass rate, a rejection rate, and a distribution of variable values.

[0084] In this embodiment, when any one of the first test result, the second test result, and the third test result does not meet the preset standard, the policy acquisition module optimizes the policy to be tested and updates the policy to be tested.

[0085] The policy online module 250 is configured to deploy the policy to be tested to the production environment when the third test result meets the preset standard.

[0086] The system for generating and deploying risk control strategies described in Example 4 of the present invention is used to run the strategy to be tested in a laboratory environment. Using production data, the strategy is progressively analyzed through three stages, and the results are gradually analyzed and the strategy is adjusted. This allows for precise control of the strategy's effectiveness, reduces risk losses, and ensures that the business meets expectations. Furthermore, running the strategy using a large amount of real data can more realistically and accurately reflect the results after the strategy is launched. This addresses the problem that manually testing strategies can only simply test connectivity and cannot meet the requirements of scenarios where a large number of variables interact. It also improves the accuracy of strategy iteration and reduces asset losses caused by inaccurate risk control.

[0087] Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by combining software with a hardware platform. Based on this understanding, all or part of the contribution of the technical solution of the present invention to the background art can be embodied in the form of a software product. This computer software product can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention or certain parts of the embodiments.

[0088] Correspondingly, embodiments of the present invention further provide a computer-readable storage medium having computer-readable instructions or a program stored thereon. When executed by a processor, the computer-readable instructions or program causes the computer to perform the following operations: the operations include the steps included in the method for generating and deploying risk control strategies described in any of the above embodiments, which are not further described here. The storage medium may include, for example, an optical disk, a hard disk, a floppy disk, a flash memory, a magnetic tape, and the like.

[0089] In addition, embodiments of the present invention further provide a computer device comprising a memory and a processor, wherein the memory is configured to store one or more computer-readable instructions or programs, wherein the one or more computer-readable instructions or programs, when executed by the processor, can implement the method for generating and deploying risk control strategies as described in any of the above embodiments. The computer device may be, for example, a server, a desktop computer, a laptop computer, a tablet computer, or the like.

[0090] Embodiments of the present invention also provide a computer program product comprising a computer program containing program code for executing the method for generating and deploying risk control strategies shown in the flowchart. When the computer program product is executed in a computer system, the program code causes the computer system to implement the method for generating and deploying risk control strategies provided in an embodiment or implementation of the present disclosure.

[0091] According to an embodiment of the present disclosure, the program code for executing the computer program provided by the embodiment of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, python, "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect via the Internet).

[0092] Finally, it should be noted that the above embodiments are intended only to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments may be modified or some of the technical features thereof may be replaced with equivalents. Such modifications or replacements do not deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A method for generating and deploying a risk control strategy, characterized in that: The method comprises: Obtaining a policy to be tested and deploying the policy to be tested in a laboratory environment; Acquire historical data within a preset range in the production environment, and generate a first test result in the laboratory environment based on the historical data and the strategy to be tested; When the first test result meets the preset standard, obtaining real-time production data in the production environment, and generating a second test result in the laboratory environment according to the real-time production data and the strategy to be tested; When the second test result meets the preset standard, obtaining a preset proportion of online data in the production environment, and generating a third test result in the laboratory environment based on the online data and the strategy to be tested; When the third test result meets the preset standard, the policy to be tested is deployed to the production environment.

2. The method according to claim 1, wherein The method further includes: when any one of the first test result, the second test result, and the third test result does not meet the preset standard, optimizing the strategy to be tested and updating the strategy to be tested.

3. The method according to claim 2, wherein The method further comprises: Obtaining first comparison data between the first test result and historical results in the production environment according to a preset comparison variable; Obtaining second comparison data between the second test result and the line result in the production environment according to the preset comparison variable; Variable record information of the third test result is extracted according to preset monitoring variables.

4. The method according to claim 3, wherein The first comparison data and the second comparison data include: pass rate, rejection rate, and difference variable; The method further includes: generating a comparison result report based on the first comparison data and the second comparison data; and generating a test result report based on the third test result, wherein the test result report includes statistical information on a pass rate, a rejection rate, and a distribution result of a variable value.

5. A system for generating and deploying risk control strategies, characterized in that: The system comprises: A policy acquisition module, configured to acquire a policy to be tested and deploy the policy to be tested in a laboratory environment; A first testing module is configured to obtain historical data within a preset range in a production environment, and generate a first test result in a laboratory environment based on the historical data and the strategy to be tested; a second testing module configured to, when the first test result meets a preset standard, obtain real-time production data in the production environment, and generate a second test result in a laboratory environment based on the real-time production data and the strategy to be tested; a third test module, configured to, when the second test result meets the preset standard, obtain a preset proportion of online data in the production environment, and generate a third test result in the laboratory environment based on the online data and the strategy to be tested; A strategy online module is used to deploy the strategy to be tested to the production environment when the third test result meets the preset standard.

6. The system according to claim 5, wherein: When any one of the first test result, the second test result, and the third test result does not meet the preset standard, the strategy acquisition module optimizes the strategy to be tested and updates the strategy to be tested.

7. The system according to claim 6, wherein: The first test module is further configured to obtain first comparison data between the first test result and historical results in the production environment based on a preset comparison variable; The second testing module is further configured to obtain second comparison data between the second test result and the line result in the production environment according to the preset comparison variable; The third test module is further configured to extract variable record information of the third test result based on preset monitoring variables.

8. A computer-readable storage medium storing computer-readable instructions, characterized in that: The computer-readable instructions are executed by a processor to implement the method for generating and deploying risk control strategies according to any one of claims 1 to 4.

9. A computer device comprising a memory and a processor, The memory stores computer-readable instructions, characterized in that: The processor executes the computer-readable instructions to implement the method for generating and deploying risk control strategies according to any one of claims 1 to 4.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method for generating and deploying a risk control strategy according to any one of claims 1 to 4 is implemented.

Images