Risk control strategy generation and evaluation method, device and equipment

Abstract

The invention relates to the technical field of data processing, in particular to a risk control strategy generation and evaluation method, device and equipment. The method comprises the following steps: acquiring a plurality of risk control strategies and different versions thereof, comparing, acquiring a real-time key index of each version, and pre-evaluating the risk control strategies according to the indexes; then obtaining business scene feature data, performing selection according to the pre-evaluation result of each version risk control strategy, then screening out appropriate strategy splitting, recombination and iteration, and generating a target risk control strategy of each business scene; and finally, storing the target risk control strategy, and performing evaluation according to the data set of the service scene to generate a visual generation evaluation report. According to the embodiment of the invention, the method solves a problem that the risk control strategy lacks flexibility and accuracy in a rapidly changing market environment, and improves the flexibility and accuracy of the risk control strategy.

Description

Technical Field

[0001] This specification relates to the field of data processing technology, and in particular to a method, device and equipment for generating and evaluating a risk control strategy. Background Art

[0002] At present, with the rapid development of network technology, the effectiveness of risk control strategies in modern banking business has become a core factor that determines the bank's risk management level and business income. At present, the construction and optimization of traditional risk control strategies still mainly rely on manual experience judgment and manual adjustment. Although they are highly flexible, they lack a scientific and systematic management framework. Among the common evaluation methods, the predefined strategies based on the rule engine can quickly respond to basic risk scenarios, but they are difficult to adapt to the dynamic market environment and complex risk forms due to the rigid rules; and the intelligent methods using machine learning models can improve the accuracy of risk assessment through algorithms, but their model training process is complex, the iteration cycle is long, and the strategy version update lacks a standardized traceability mechanism, resulting in a vague model optimization path. More prominently, existing technologies generally have systemic defects: strategy adjustments rely on manual intervention, resulting in strong subjectivity and insufficient reliability of evaluation results; the lack of version control makes it difficult to trace the history of strategy changes and chaotic management; there is a lack of efficient comparison tools between multiple versions of strategies, and it is impossible to quickly screen the optimal solution to support decision-making. These problems not only reduce the iteration efficiency of risk control strategies, but are also likely to cause risk exposure to expand due to deviations in strategy execution.

[0003] Therefore, there is an urgent need for a risk control strategy management technology that integrates precise iteration, version traceability and multi-dimensional evaluation. By building a systematic and quantifiable strategy lifecycle management and control system, the shortcomings of traditional methods in version management, evaluation objectivity and strategy optimization efficiency can be addressed, thereby improving the transparency of bank risk prevention and control and the reliability of business decisions. Summary of the invention

[0004] In order to solve the problem that traditional risk control strategies in the prior art are often difficult to adapt to the rapid changes in the market, they need to rely on the experience of risk control experts and manually adjust the strategies, which lacks scientificity and systematicity. The problem that multiple strategy versions cannot be compared and analyzed efficiently, this specification embodiment provides a method, device and equipment for generating and evaluating risk control strategies. By introducing a version control system and a scientific evaluation method, accurate and efficient iteration and evaluation of strategies are achieved, thereby improving the flexibility and accuracy of risk control strategies. This solves the problem that bank risk control strategies lack flexibility and accuracy in a rapidly changing market environment.

[0005] In order to solve the above technical problems, this specification provides a method for generating and evaluating a risk control strategy, the method comprising:

[0006] Obtain a policy set containing multiple risk control policies and their different versions, and generate a comparison chart of the multiple risk control policies;

[0007] Obtain the feature data of at least one business scenario and establish a data set. According to the pre-evaluation results of each version of the risk control policy and the data set, screen out at least one risk control policy suitable for the business scenario, split and reorganize the at least one risk control policy and iterate it to generate the target risk control policy for each business scenario;

[0008] Obtain the feature data of at least one business scenario and establish a data set. According to the pre-evaluation results of each version of the risk control policy and the data set, select at least one risk control policy suitable for the business scenario for iteration to generate the target risk control policy for each business scenario;

[0009] Store the target risk control policy through a distributed version control system and a relational database, and generate a visualization evaluation report according to the data set of the business scenario.

[0010] Furthermore, obtaining a policy set containing multiple risk control policies and their different versions includes,

[0011] Obtain multiple risk control policies, the historical versions of each risk control policy and their change records to generate the policy set.

[0012] Furthermore, performing policy trial calculations on the policy set to obtain real-time key indicators in each version of the risk control policy further includes,

[0013] Apply the same test data to simulate the operation of different historical versions of the same policy in the policy set;

[0014] Calculate and analyze according to the results of each simulation operation to obtain the real-time key indicators in each version of the risk control policy.

[0015] Furthermore, the real-time key indicators include,

[0016] Policy manual review rate, policy passing rate, policy rejection rate, approved quantity, number of application pens, processing time, approval amount range, approval accuracy rate, approval error rate.

[0017] Furthermore, the data set further includes,

[0018] Weight parameters corresponding to preset feature dimensions of different business scenarios;

[0019] The preset feature dimension is a feature dimension pre-divided based on the feature data of each business scenario;

[0020] The weight parameter includes a weight coefficient corresponding to the preset feature dimension determined according to prior rules.

[0021] Further, selecting at least one risk control strategy suitable for the business scenario for iteration according to the pre-evaluation results of the risk control strategies of each version and the dataset further includes

[0022] selecting at least one risk control strategy suitable for the business scenario according to the relationship between the strategy set and the dataset;

[0023] iterating at least one risk control strategy corresponding to the business scenario to convergence according to preset metrics and weight parameters corresponding to preset feature dimensions of the business scenario to generate the target risk control strategy.

[0024] On the other hand, an embodiment of this specification also provides a device for generating and evaluating risk control strategies, and the device includes

[0025] a strategy set acquisition module, configured to acquire a strategy set including multiple risk control strategies and their different versions, and generate a comparison chart of the multiple risk control strategies;

[0026] a strategy pre-evaluation module, configured to perform strategy trial calculations on the strategy set to obtain real-time key metrics in each version of the risk control strategy, and pre-evaluate each version of the risk control strategy in the strategy set according to the real-time key metrics and the comparison chart;

[0027] a target strategy generation module, configured to acquire feature data of at least one business scenario and establish a dataset, select at least one risk control strategy suitable for the business scenario for iteration according to the pre-evaluation results of each version of the risk control strategy and the dataset, and generate a target risk control strategy for each business scenario;

[0028] a strategy evaluation module, configured to store the target risk control strategy through a distributed version control system and a relational database, and generate a visual evaluation report according to the dataset of the business scenario.

[0029] On the other hand, an embodiment of this specification also provides a computer device, including a memory, a processor, and a computer program stored on the memory. When the processor executes the computer program, the above method is implemented.

[0030] On the other hand, an embodiment of this specification also provides a computer storage medium, on which a computer program is stored. When the computer program is run by a processor of a computer device, the above method is executed.

[0031] Finally, an embodiment of this specification also provides a computer program product, where the computer program product includes a computer program, and when the computer program is executed by a processor, the above method is implemented.

[0032] Using the embodiments of this specification, in order to accurately iterate and evaluate the policies of the policy service platform, it is first necessary to obtain a policy set containing multiple risk control policies and their different versions, generate a comparison chart containing multiple risk control policies and their different versions, and ensure the integrity of the change records and historical data of each policy version; then perform policy trial calculations on the policy set to obtain the real-time key indicators in each version of the risk control policy, ensure the timeliness and accuracy of the evaluation results, and pre-evaluate the risk control policies of each version in the policy set according to the real-time key indicators and the comparison chart, simulate the operation of different policy versions, improve the scientificity and accuracy of the evaluation, and enhance the effectiveness of the policy in actual applications. Next, obtain the feature data of at least one business scenario and establish a data set to ensure that each version of the policy can achieve the best effect in a specific scenario; finally, select at least one risk control policy suitable for the business scenario for iteration according to the pre-evaluation results of each version of the risk control policy and the data set, and generate the target risk control policy for each business scenario; store the target risk control policy through a distributed version control system and a relational database, and generate a visual evaluation report according to the data set of the business scenario. Ensure that the change history of each version is traceable, and enhance the transparency and reliability of policy management. At the same time, business personnel can achieve accurate version iteration and evaluation of policies through a visual interface. Through this method, the problems in the prior art of manually adjusting policies, lacking scientificity and systematicness, and being unable to efficiently compare and analyze multiple policy versions are solved, realizing the accurate iteration and evaluation of risk control policies, ensuring that each version achieves the best effect in a specific scenario, and enhancing the risk management level and business benefits of the bank. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] In order to more clearly illustrate the technical solutions in the embodiments of this specification or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of this specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0034] Figure 1 The figure shows a schematic diagram of a risk control policy generation and evaluation system according to an embodiment of this specification;

[0035] Figure 2 The figure shows a flowchart of a risk control policy generation and evaluation method according to an embodiment of this specification;

[0036] Figure 3 The figure shows a schematic diagram of the logical version management of risk policies based on a database according to an embodiment of this specification;

[0037] Figure 4The following is a schematic flowchart of the process of performing policy calculation on the policy set according to the embodiments of the present specification to obtain real-time key indicators in each risk control policy version;

[0038] Figure 5 The following is a schematic flowchart of the process of selecting at least one risk control policy suitable for the business scenario for iteration according to the embodiments of the present specification;

[0039] Figure 6 The following shows a device for generating and evaluating a risk control policy according to the embodiments of the present specification;

[0040] Figure 7 The following is a schematic structural diagram of a computer device according to the embodiments of the present specification.

[0041]

Explanation of the reference numerals

[0042] 101, terminal;

[0043] 102, server;

[0044] 301, database;

[0045] 302, file system;

[0046] 303, user terminal;

[0047] 702, computer device;

[0048] 704, processing device;

[0049] 706, storage resource;

[0050] 708, driving mechanism;

[0051] 710, input / output module;

[0052] 712, input device;

[0053] 714, output device;

[0054] 716, presentation device;

[0055] 718, graphical user interface;

[0056] 720, network interface;

[0057] 722, communication link;

[0058] 724, communication bus. Detailed implementation manners

[0059] Next, in combination with the accompanying drawings in the embodiments of this specification, the technical solutions in the embodiments of this specification will be clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of this specification, rather than all the embodiments. Based on the embodiments in this specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope protected by this specification.

[0060] It should be noted that the terms "first", "second", etc. in this specification and the above-mentioned accompanying drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of this specification described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product, or equipment that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products, or equipment.

[0061] As Figure 1 shown in the schematic diagram of a risk control strategy generation and evaluation system according to an embodiment of the present invention, it may include a terminal 101 and a server 102. A communication connection is established between the terminal 101 and the server 102 to enable data interaction. The terminal 101 can obtain a policy set containing multiple risk control strategies and their different versions and send them to the server 102. The server 102 generates a comparison chart of the policy set and evaluates each risk control strategy therein. In addition, for the purpose of simulating tests in different business scenarios, the server 102 can pre-evaluate different business scenarios and iterate the target risk control strategy for each scenario according to the pre-evaluation results. Then, the server 102 stores each target risk control strategy and generates a visual generation evaluation report and uploads it back to the terminal 101.

[0062] In the embodiments of this specification, the server 102 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery network (CDN), and big data and artificial intelligence platforms.

[0063] It should be noted that Figure 1The figure only shows an application environment provided by the present disclosure. In actual applications, other application environments may also be included, which are not limited in the embodiments of the present invention.

[0064] In the technical solution of this application, the acquisition, storage, use, processing, etc. of data all comply with the relevant regulations of relevant laws and regulations.

[0065] When the server 102 receives multiple risk control strategies and their different versions, it generates a policy set covering all versions and generates a comparison chart for comparison, so as to solve the problem of being unable to efficiently compare and analyze multiple policy versions. To solve the above problems, the embodiments of this article provide a method for generating and evaluating risk control strategies. According to multiple risk control strategies and their different versions, the target risk control strategy suitable for each business scenario is iteratively generated. Figure 2 The figure shows a flowchart of a method for generating and evaluating a risk control strategy according to an embodiment of this article. The generation and evaluation process of the risk control strategy is described in this figure, but based on routine or non-creative labor, more or fewer operation steps may be included. The order of steps listed in the embodiments is only one way among the execution orders of numerous steps and does not represent the only execution order. When the system or device product in reality executes, it can be executed in the order of the method shown in the embodiments or the figures, or executed in parallel.

[0066] Step 201: Obtain a policy set including multiple risk control strategies and their different versions, and generate a comparison chart of the multiple risk control strategies;

[0067] Step 202: Perform policy trial calculations on the policy set to obtain real-time key indicators in each version of the risk control strategy, and pre-evaluate each version of the risk control strategy in the policy set according to the real-time key indicators and the comparison chart;

[0068] Step 203: Obtain the characteristic data of at least one business scenario and establish a data set, screen out at least one risk control strategy suitable for the business scenario according to the pre-evaluation results of each version of the risk control strategy and the data set, split and reorganize the at least one risk control strategy and perform iteration to generate the target risk control strategy for each business scenario;

[0069] Step 204: Store the target risk control strategy through a distributed version control system and a relational database, and generate a visual evaluation report according to the data set of the business scenario.

[0070] Using the embodiments of this specification, in order to accurately iterate and evaluate the policies of the policy service platform, it is first necessary to obtain a policy set containing multiple risk control policies and their different versions, generate a comparison chart containing multiple risk control policies and their different versions, and ensure the integrity of the change records and historical data of each policy version; then perform policy trial calculations on the policy set to obtain the real-time key indicators in each version of the risk control policy, ensure the timeliness and accuracy of the evaluation results, and pre-evaluate each version of the risk control policy in the policy set according to the real-time key indicators and the comparison chart, simulate the operation of different policy versions, improve the scientificity and accuracy of the evaluation, and enhance the effectiveness of the policy in actual applications. Next, obtain the characteristic data of at least one business scenario and establish a data set to ensure that each version of the policy can achieve the best effect in a specific scenario; finally, select at least one risk control policy suitable for the business scenario for iteration according to the pre-evaluation results of each version of the risk control policy and the data set, and generate the target risk control policy for each business scenario; store the target risk control policy through a distributed version control system and a relational database, and generate a visual evaluation report according to the data set of the business scenario. Ensure that the change history of each version is traceable, and enhance the transparency and reliability of policy management. At the same time, business personnel can achieve accurate version iteration and evaluation of policies through a visual interface. Through this method, the problem in the prior art of manually adjusting policies, lacking scientificity and systematicness, and being unable to efficiently compare and analyze multiple policy versions is solved, realizing the accurate iteration and evaluation of risk control policies, ensuring that each version achieves the best effect in a specific scenario, and enhancing the risk management level and business benefits of the bank.

[0071] In the embodiments of this specification, to improve the technical level and market competitiveness in risk policy management, a precise iteration and evaluation tool for policies is implemented on the policy service platform. Through multi-version control of policies, the traceability and systematic management of policies are realized, ensuring the integrity of the change records and historical data of each policy version, and using historical data for trial calculation and evaluation, simulating the operation of different policy versions, improving the scientificity and accuracy of the evaluation, and enhancing the effectiveness of the policy in actual applications. Based on the policy configuration center, compare multiple policy versions, and generate detailed comparison reports and charts based on data visualization technologies such as Echarts to assist decision-makers in selecting the optimal version; in the policy operation center, realize the quasi-real-time calculation of key indicators through technologies such as window aggregation of Flink, ensure the timeliness and accuracy of the evaluation results, provide a reliable basis for policy optimization, and improve the scientificity and accuracy of the evaluation. At the same time, business personnel can achieve accurate version iteration and evaluation of policies through a visual interface, ensure that each version achieves the best effect in a specific scenario, and through a set of efficient management processes, realize the full-process onlineization from version control, experimental evaluation to performance index calculation of policy codes.

[0072] In one embodiment of the present specification, in order to achieve the traceability and systematic management of policies and ensure the integrity of the change records and historical data of each policy version, obtaining the policy set including multiple risk control policies and their different versions in step 201 includes,

[0073] Obtaining multiple risk control policies, the historical versions of each risk control policy, and their change records to generate the policy set.

[0074] Specifically, during the process of version control, it also includes the physical management of source code and the logical version management of risk policies based on the database. Among them, the physical management of source code uses a distributed version control system (such as Git) to manage the source code, record information such as change content, time, and author. Through branch management and merge strategies, it is ensured that different versions of the source code can be developed and tested in parallel, improving development efficiency and code quality. Among them, as Figure 3 shown, the version control includes the physical management of source code and the logical version management of risk policies based on the database, including:

[0075] Using an efficient database 301 and file system 302 to store version data; completing data verification through a hash algorithm to achieve data integrity; implementing distributed collaboration through network communication functions so that the client 303 can remotely obtain and push version data; through the command line and graphical interface, enabling the client 303 to perform various version control operations, such as commit, push, pull, etc., for the convenience of users.

[0076] In the embodiment of the present specification, the logical version management of risk policies uses a relational database to manage risk policy versions, recording metadata such as version identifiers, creation times, creators, description information, and change logs. Through the version control function of the database, the change history of each policy version is traced to ensure that all changes are traceable.

[0077] In another embodiment of the present specification, in order to ensure the timeliness and accuracy of evaluation results, obtaining real-time metrics for each policy version, such as Figure 4 shown, further including obtaining the real-time key metrics in each risk control policy version by performing a policy trial calculation on the policy set,

[0078] Step 401: Simulating the operation of different historical versions of the same policy in the policy set using the same test data;

[0079] Step 402: Performing trial calculation and analysis to obtain the real-time key metrics in each risk control policy version.

[0080] Among them, the real-time key metrics include,

[0081] Policy manual review rate, policy pass rate, policy rejection rate, approved quantity, application volume, processing time, approval amount range, approval accuracy rate, approval error rate.

[0082] Specifically, in the embodiments of this specification, by constructing a multi-version parallel trial calculation framework and a dynamic resource monitoring system, the accurate quantification of policy performance and version optimization are achieved. The multiple historical versions of the same policy are synchronously simulated and run through a standardized test data set to ensure the consistency of the triggering conditions of each version's logical branches. By driving the parallel trial calculation of multiple versions of the policy with the same test data, it helps the bank quickly screen the optimal risk control policy version. Exemplarily, when it is necessary to evaluate the accuracy of multiple versions of the policy, each policy version will independently run exactly the same test data (such as the time, amount, user information, credit information, historical offline features, etc. of processing the same batch of transactions), avoiding evaluation result deviations caused by data differences. At the same time, the system will monitor the running results of each policy version in real time, including: policy manual review rate, policy pass rate, policy rejection rate, approved quantity, application volume, processing time, approval amount range, approval accuracy rate, approval error rate. Based on these real-time metrics, the system will automatically compare the risk control effects of different versions and finally recommend different policy versions to be adopted in different situations.

[0083] In the embodiments of this specification, in order to ensure that the generated risk control policy can be adjusted according to the characteristic data of each business scenario, the data set further includes

[0084] Weight parameters corresponding to preset feature dimensions of different business scenarios;

[0085] The preset feature dimension is a feature dimension pre-divided based on the characteristic data of each business scenario;

[0086] The weight parameter includes a weight coefficient corresponding to the preset feature dimension determined according to prior rules.

[0087] In a specific embodiment of the present invention, the feature dimensions and their weight parameter configurations preset for different business scenarios further optimize the flexibility and accuracy of the risk control strategy. The preset feature dimensions are pre-divided based on the feature data of each business scenario. For example, in the "cross-border remittance" business scenario, the preset feature dimensions include transaction frequency, single transaction amount, risk level of the receiving region, and user identity authentication strength; while in the "small and micro enterprise loan approval" scenario, the feature dimensions are divided into enterprise credit rating, industry prosperity index, cash flow stability, and type of collateral assets. The weight parameters of each feature dimension are dynamically set by prior rules, and the prior rules are derived from historical risk event statistics, regulatory requirements, and industry expert experience. Customizing feature dimensions for different business requirements (such as cross-border remittance, loan approval) reflects the differential risk control logic. It solves the problem of rigid cross-scenario risk control strategies. Especially when dealing with complex scenarios such as sudden changes in regional risks in cross-border business and industry fluctuations in small and micro enterprise loans, it demonstrates significant technical advantages. At the same time, through systematic version control and scientific evaluation methods, precise iteration of the risk control strategy is achieved to ensure that each version achieves the best effect in a specific scenario.

[0088] According to another embodiment of this specification, in order to obtain the target risk control strategy corresponding to each business scenario, as Figure 5 shown, further including iterating at least one risk control strategy suitable for the business scenario according to the pre-evaluation results of each version of the risk control strategy and the data set,

[0089] Step 501: Select at least one risk control strategy suitable for the business scenario according to the relationship between the strategy set and the data set;

[0090] Step 502: Iterate at least one risk control strategy corresponding to the business scenario to convergence according to the preset metrics and the weight parameters corresponding to the preset feature dimensions of the business scenario to generate the target risk control strategy.

[0091] Specifically, by constructing a historical policy set and a dynamic scenario adaptation mechanism, cross-cycle iterative optimization of risk control policies is achieved. First, the historical risk control policy versions stored in the policy set are matched with the current business scenario dataset in terms of features. The policy set contains multiple historical policy versions, and the dataset consists of indicators such as customer credit characteristics, business transaction fluctuation patterns, and industry risks under each business scenario. Based on the distribution similarity between the feature data of the historical policy effective period and the current scenario data, the system filters out the historical policies with qualified adaptability to form a candidate policy pool with business continuity. Then, in the policy iteration and optimization stage, for the multiple business objectives of the current business scenario, the system decomposes the candidate risk control policies into recombinable basic rule units and model components, and recombines and evaluates the basic rule units and model components of each candidate policy according to the preset indicators and the weight parameters corresponding to the preset feature dimensions of the business scenario, and outputs the target risk control policy that integrates the advantages of multiple historical policies. By establishing a temporal binding mechanism between policy versions and historical scenario data, the digital inheritance of policy experience is realized; breaking through the limitations of traditional parameter tuning, supporting the deep recombination of cross-generation policy components, and avoiding the trap of local optimality. Compared with traditional risk control policy iteration methods, it can show stronger policy migration ability and risk adaptation flexibility.

[0092] Through the collaborative architecture of a distributed version control system and a relational database, the whole process tracking and historical backtracking of risk control policy codes and parameters are realized, automatically recording the complete information of each change and supporting differential comparison, solving the problems of traditional version management relying on manual work and being difficult to trace. At the same time, an automated test and evaluation system is constructed. Through the simulation of the policy pre-release observation mode and the analysis of quantitative indicators, multi-dimensional evaluation results are generated, providing a scientific basis for policy optimization and significantly improving the evaluation efficiency and objectivity. The system integrates multi-version comparison analysis and dynamic iteration mechanisms, identifies the optimal policy version through a visualization tool, and triggers parameter-directed optimization in combination with real-time market data and effect feedback to achieve precise policy iteration. Supported by a high-performance computing architecture for parallel evaluation and rapid response, it ensures the efficient adaptation of risk control policies in a complex market environment, forms a closed-loop management of development, testing, and iteration, and comprehensively improves the risk control ability and business adaptability.

[0093] Exemplarily, in the banking business scenario, a technical implementation system to support the iterative optimization of risk control policies is constructed. Based on a structured database system, the full life cycle management of test data is realized. By classifying and storing historical policy data, a label system associated with customer risk portraits and transaction behavior characteristics is established to form a traceable policy evolution map. The test dataset is converted into a CSV format file and stored in a bank-level object storage system through a secure transmission protocol to ensure access control and audit compliance of sensitive business data.

[0094] The system maintains the integrity and consistency of test data through the database transaction mechanism, and supports multi-dimensional data retrieval and dynamic update functions. In the policy trial calculation and evaluation stage, parallel simulation operations are performed on multiple policy versions using the same benchmark test data set. The evaluation engine developed based on the microservice framework performs simulation deductions of core business scenarios such as credit approval and risk prediction through automated scripts, and synchronously calculates the key performance indicators of policy versions in the real business environment. Finally, the evaluation dimensions cover core risk control indicators such as the risk identification accuracy rate (including precision and recall rate), the approval decision accuracy rate, and the business processing time deviation rate to construct a multi-dimensional policy evaluation matrix. By quantitatively analyzing the performance differences of different version policies in links such as customer access and risk pricing, it provides data support for the version switching decision. This technical system effectively guarantees the iterative optimization and precise deployment of credit risk control policies, and significantly improves the bank's risk resistance ability and operating efficiency in the complex market environment.

[0095] As Figure 6 described, this specification also discloses a device for generating and evaluating risk control strategies. The device includes,

[0096] A policy set acquisition module 601, configured to acquire a policy set including multiple risk control strategies and their different versions, and generate a comparison chart of the multiple risk control strategies;

[0097] A policy pre-evaluation module 602, configured to perform policy trial calculations on the policy set to obtain real-time key indicators in each version of the risk control strategy, and pre-evaluate each version of the risk control strategy in the policy set according to the real-time key indicators and the comparison chart;

[0098] A target policy generation module 603, configured to acquire feature data of at least one business scenario and establish a data set, and select at least one risk control strategy suitable for the business scenario for iteration according to the pre-evaluation results of each version of the risk control strategy and the data set, and generate a target risk control strategy for each business scenario;

[0099] A policy evaluation module 604, configured to store the target risk control strategy through a distributed version control system and a relational database, and generate a visual evaluation report according to the data set of the business scenario.

[0100] In the embodiments of this specification, through the deep combination of systematic version control and scientific evaluation methods, a risk control strategy iteration system that is traceable, quantifiable, and optimizable is constructed. Specifically, a distributed version control system is used to manage the entire life cycle of the strategy source code, ensuring that the submission records, version differences, and associated metadata of each strategy change are completely traceable. At the same time, based on a relational database, the logical versions of risk strategies are stored in a structured manner, recording the version effective time, modifying personnel, and change impact scope, realizing the transparent management of strategy evolution. In the strategy evaluation link, through the strategy pre-evaluation module 602, multi-version parallel tests are performed on historical or simulated transaction data, and the strategy effects are dynamically quantified in combination with preset key performance indicators (such as strategy manual review rate, strategy passing rate, strategy rejection rate, approved quantity, application number, processing time, approval amount range, etc.). The relational database is used to classify and store the test data and perform tagged association to ensure the integrity and reproducibility of the evaluation process. Further, through the multi-version strategy evaluation module 604, a difference report and visualization charts are automatically generated to intuitively present the performance of different versions in specific scenarios, driving the direction of strategy optimization. Finally, a new version is generated based on the feedback results and the next round of iteration is started, forming a full closed-loop optimization link covering strategy development, testing, deployment, and monitoring. This technical system solves the pain points of version chaos, strong subjectivity in evaluation, and insufficient basis for optimization in traditional risk control strategy iteration, ensuring that the strategy iteration process is both scientific and engineering controllable.

[0101] As Figure 7 shown is a schematic structural diagram of a computer device according to an embodiment of the present invention. The device in this specification may be the computer device in this embodiment, which executes the method in this specification. The computer device 702 may include one or more processing devices 704, such as one or more central processing units (CPUs), and each processing unit may implement one or more hardware threads. The computer device 702 may also include any storage resource 706 for storing any type of information such as code, settings, data, etc. Non-limitingly, for example, the storage resource 706 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory devices, hard disks, optical discs, etc. More generally, any storage resource may use any technology to store information. Further, any storage resource may provide volatile or non-volatile retention of information. Further, any storage resource may represent a fixed or removable component of the computer device 702. In one case, when the processing device 704 executes the associated instructions stored in any storage resource or combination of storage resources, the computer device 702 may perform any operation of the associated instructions. The computer device 702 also includes one or more drive mechanisms 708 for interacting with any storage resource, such as a hard disk drive mechanism, an optical disc drive mechanism, etc.

[0102] The computer device 702 may further include an input / output module 710 (I / O) for receiving various inputs (via the input device 712) and for providing various outputs (via the output device 714). A specific output mechanism may include a presentation device 716 and an associated graphical user interface (GUI) 718. In other embodiments, the input / output module 710 (I / O), the input device 712, and the output device 714 may not be included, and it may only be a computer device in the network. The computer device 702 may further include one or more network interfaces 720 for exchanging data with other devices via one or more communication links 722. One or more communication buses 724 couple the components described above together.

[0103] The communication link 722 may be implemented in any manner, for example, via a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication link 722 may include any combination of hardwired links, wireless links, routers, gateway functions, name servers, etc. governed by any protocol or combination of protocols.

[0104] Corresponding to Figures 2 to 5 In the embodiments of this specification, a computer-readable storage medium is further provided. A computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the steps of the above-described method are executed.

[0105] The embodiments of this specification further provide a computer-readable instruction. When the processor executes the instruction, the program causes the processor to execute the method as Figures 2 to 5 shown.

[0106] It should be understood that in the various embodiments of this specification, the magnitudes of the serial numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of this specification.

[0107] It should also be understood that in the embodiments of this specification, the term "and / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B may represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this specification generally represents an "or" relationship between the associated objects before and after.

[0108] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in combination with the embodiments disclosed in this specification can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled professionals can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this specification.

[0109] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0110] In the several embodiments provided in this specification, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed couplings or direct couplings or communication connections to each other can be indirect couplings or communication connections through some interfaces, devices, or units, and can also be electrical, mechanical, or other forms of connection.

[0111] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiments of this specification.

[0112] In addition, the functional units in the various embodiments of this specification can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0113] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of this specification, in essence, or the part that contributes to the prior art, or all or part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this specification. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs.

[0114] Specific embodiments are used in this specification to elaborate on the principles and implementation manners of this specification. The descriptions of the above embodiments are only used to help understand the method and its core idea of this specification; at the same time, for those of ordinary skill in the art, according to the idea of this specification, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to this specification.

Claims

1. A method for generating and evaluating a risk control strategy, characterized in that: The method comprises, Obtaining a policy set including multiple risk control policies and different versions thereof, and generating a comparison chart of the multiple risk control policies; Performing strategy trial calculation on the strategy set to obtain real-time key indicators in each version of the risk control strategy, and pre-evaluating each version of the risk control strategy in the strategy set based on the real-time key indicators and the comparison chart; Acquire characteristic data of at least one business scenario and establish a data set, filter out at least one risk control strategy suitable for the business scenario according to the pre-evaluation results of each version of the risk control strategy and the data set, split and reorganize the at least one risk control strategy and iterate to generate a target risk control strategy for each business scenario; The target risk control strategy is stored through a distributed version control system and a relational database, and an evaluation report is generated visually based on the data set of the business scenario.

2. The method for generating and evaluating risk control strategies according to claim 1, characterized in that: Get a policy set containing multiple risk control policies and their different versions, including: A plurality of risk control strategies, historical versions of each risk control strategy and change records thereof are obtained to generate the strategy set.

3. The method for generating and evaluating risk control strategies according to claim 1, characterized in that: Performing strategy calculation on the strategy set to obtain real-time key indicators in each risk control strategy version further includes: Applying the same test data to simulate different historical versions of the same strategy in the strategy set; Trial analysis obtains real-time key indicators in each risk control strategy version.

4. The method for generating and evaluating risk control strategies according to claim 3, characterized in that: The real-time key indicators include: Strategy manual review rate, strategy approval rate, strategy rejection rate, number of approvals, number of applications, processing time, approval amount range, precision rate, approval accuracy rate, and approval error rate.

5. The method for generating and evaluating risk control strategies according to claim 3, characterized in that: The data set further includes, Preset weight parameters corresponding to feature dimensions in different business scenarios; The preset feature dimension is a feature dimension obtained by pre-dividing the feature data of each business scenario; The weight parameter includes a weight coefficient corresponding to the preset feature dimension determined according to a priori rules.

6. The method for generating and evaluating risk control strategies according to claim 5, characterized in that: Selecting at least one risk control strategy suitable for the business scenario for iteration according to the pre-evaluation result of each version of the risk control strategy and the data set further includes: Selecting at least one risk control strategy suitable for the business scenario according to the relationship between the strategy set and the data set; According to preset indicators and weight parameters corresponding to preset feature dimensions of the business scenario, at least one risk control strategy corresponding to the business scenario is iterated until convergence to generate the target risk control strategy.

7. A device for generating and evaluating a risk control strategy, characterized in that: The device comprises, A strategy set acquisition module, used to acquire a strategy set including multiple risk control strategies and their different versions, and generate a comparison chart of the multiple risk control strategies; A strategy pre-evaluation module, used to perform strategy calculation on the strategy set to obtain real-time key indicators in each version of the risk control strategy, and pre-evaluate each version of the risk control strategy in the strategy set based on the real-time key indicators and the comparison chart; A target strategy generation module is used to obtain characteristic data of at least one business scenario and establish a data set, select at least one risk control strategy suitable for the business scenario according to the pre-evaluation results of each version of the risk control strategy and the data set, and iterate to generate a target risk control strategy for each business scenario; The strategy evaluation module is used to store the target risk control strategy through a distributed version control system and a relational database, and to evaluate and generate a visual evaluation report based on the data set of the business scenario.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images