Model-based data processing method and device, equipment, medium and product

Abstract

The embodiment of the invention provides a model-based data processing method and device, equipment, a medium and a product, and belongs to the technical field of data processing. The method comprises the following steps: acquiring mechanism characteristic data of a plurality of insurance mechanisms; inputting the mechanism characteristic data into a pre-trained risk bearing capability evaluation model to obtain a risk bearing capability index corresponding to each insurance mechanism; according to the risk bearing capability index, performing differential threshold setting on the plurality of insurance institutions to obtain a reinsurance over-claim threshold corresponding to each insurance institution; the reinsurance over-claim threshold value is used for representing the maximum claim payment resource quantity of the insurance institution in the preset insurance claim payment scene; and determining the compensation proportion of each insurance institution in a preset insurance compensation scene according to the reinsurance over-claim threshold value. According to the embodiment of the invention, the accuracy of determining the reinsurance over-claim scheme can be improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a model-based data processing method and apparatus, equipment, medium, and product. Background Art

[0002] Within the insurance industry's risk management system, the excess-of-loss reinsurance mechanism has long been a crucial tool for risk diversification. Related technologies primarily focus on risk transfer between insurance companies and reinsurers, typically achieving this external risk transfer by setting a uniform compensation ratio. However, these technologies often rely on manual experience to determine the compensation ratio, resulting in low efficiency. Furthermore, due to the subjectivity and limitations of manual experience, it's difficult to fully consider all relevant factors, resulting in a low accuracy in the final compensation ratio. Summary of the Invention

[0003] The main purpose of the embodiments of the present application is to propose a model-based data processing method and device, equipment, medium and product, aiming to improve the accuracy of determining the compensation ratio.

[0004] To achieve the above objectives, a first aspect of an embodiment of the present application proposes a model-based data processing method, the method comprising:

[0005] Obtaining institutional characteristic data of multiple insurance institutions, where there is a dependency relationship between the multiple insurance institutions;

[0006] Input the institutional characteristic data into the pre-trained risk-bearing capacity assessment model to obtain the corresponding risk-bearing capacity index of each insurance institution. The risk-bearing capacity index is positively correlated with the insurance institution's ability to bear risks.

[0007] Based on the risk-bearing capacity index, differentiated thresholds are set for multiple insurance institutions to obtain the corresponding reinsurance excess loss thresholds for each insurance institution. The reinsurance excess loss thresholds are used to represent the maximum compensation resources of an insurance institution under a preset insurance compensation scenario.

[0008] Based on the reinsurance excess loss threshold, the compensation ratio of each insurance institution under the preset insurance compensation scenario is determined.

[0009] To achieve the above-mentioned purpose, a second aspect of an embodiment of the present application provides a model-based data processing device, the device comprising:

[0010] An acquisition module, configured to acquire institutional characteristic data of a plurality of insurance institutions, wherein the plurality of insurance institutions have a dependency relationship between them;

[0011] A first determination module is configured to input the institution characteristic data into a pre-trained risk-bearing capacity assessment model to obtain a risk-bearing capacity index corresponding to each insurance institution, wherein the risk-bearing capacity index is positively correlated with the insurance institution's ability to bear risk;

[0012] a setting module for setting differentiated thresholds for the multiple insurance institutions based on the risk-bearing capacity index to obtain a corresponding reinsurance excess loss threshold for each insurance institution; the reinsurance excess loss threshold is used to represent the maximum compensation resource amount of the insurance institution under a preset insurance compensation scenario;

[0013] The second determining module is used to determine the compensation ratio of each insurance institution under the preset insurance compensation scenario according to the reinsurance excess loss threshold.

[0014] To achieve the above-mentioned purpose, the third aspect of an embodiment of the present application proposes an electronic device, which includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the method described in the first aspect when executing the computer program.

[0015] To achieve the above-mentioned purpose, the fourth aspect of the embodiments of the present application proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method described in the first aspect.

[0016] To achieve the above objectives, the fifth aspect of the embodiments of the present application proposes a computer program product. When the instructions in the computer program product are executed by the processor of an electronic device, the electronic device executes the method described in the first aspect above.

[0017] The model-based data processing method, apparatus, device, medium, and product proposed in this application obtains institutional characteristic data for multiple dependent insurance institutions. This institutional characteristic data is input into a pre-trained risk-bearing capacity assessment model to obtain a risk-bearing capacity index for each insurance institution. This risk-bearing capacity index is positively correlated with the insurance institution's risk tolerance. Based on this risk-bearing capacity index, differentiated thresholds are set for multiple insurance institutions to obtain a corresponding reinsurance excess loss threshold for each insurance institution, thereby accurately reflecting each institution's actual risk tolerance. Furthermore, based on these reinsurance excess loss thresholds, the compensation ratio for each insurance institution under pre-set insurance claims scenarios is determined, ensuring that the compensation ratio fully adapts to the differentiated risk characteristics of different institutions within the insurance company and achieving reasonable risk sharing. This model-based assessment method avoids or reduces the subjectivity and limitations caused by reliance on manual experience in related technologies, thereby significantly improving the accuracy of compensation ratio determination. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is one of the flow charts of the model-based data processing method provided in the embodiments of the present application;

[0019] Figure 2 This is the second flowchart of the model-based data processing method provided in the embodiment of the present application;

[0020] Figure 3 This is the third flow chart of the model-based data processing method provided in the embodiment of the present application;

[0021] Figure 4 This is the fourth flow chart of the model-based data processing method provided in the embodiment of the present application;

[0022] Figure 5 is a schematic structural diagram of a model-based data processing device provided in an embodiment of the present application;

[0023] Figure 6 This is a schematic diagram of the hardware structure of the electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0025] It should be noted that although the device schematics illustrate functional module divisions and the flowcharts illustrate logical sequences, in certain circumstances, the steps shown or described may be performed in a sequence that differs from the module divisions in the device or the sequence in the flowcharts. The terms "first," "second," and so on, in the specification, claims, and drawings, are used to distinguish similar items and are not necessarily used to describe a specific sequence or precedence.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein are for the purpose of describing the embodiments of this application only and are not intended to limit this application.

[0027] First, let’s analyze some of the terms used in this application:

[0028] Artificial intelligence (AI) is a new technical discipline that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. A branch of computer science, AI seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. AI can simulate the information processes of human consciousness and thinking. It also encompasses the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, to perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results.

[0029] Based on this, embodiments of the present application provide a model-based data processing method and apparatus, equipment, medium, and product, aiming to improve the accuracy of determining reinsurance excess loss plans.

[0030] The model-based data processing method, apparatus, equipment, medium, and product provided in the embodiments of the present application are specifically illustrated through the following embodiments. First, the model-based data processing method in the embodiments of the present application is described.

[0031] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use knowledge to achieve optimal results.

[0032] Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0033] The model-based data processing method provided in the embodiment of the present application relates to the field of artificial intelligence technology. The model-based data processing method provided in the embodiment of the present application can be applied to a terminal, can be applied to a server side, or can be software running in a terminal or a server side. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc.; the server side can be configured as an independent physical server, or as a server cluster or distributed system composed of multiple physical servers, or as 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, CDN, and big data and artificial intelligence platforms; the software can be an application that implements a model-based data processing method, etc., but is not limited to the above forms.

[0034] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments in which tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media, including storage devices.

[0035] It should be noted that in each specific embodiment of the present application, when it comes to the need to perform relevant processing based on data related to the user's identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first, and the collection, use, and processing of such data will comply with relevant laws, regulations, and standards. In addition, when the embodiment of the present application needs to obtain the user's sensitive personal information, the user's separate permission or consent will be obtained through a pop-up window or by jumping to a confirmation page. After clearly obtaining the user's separate permission or consent, the necessary user-related data for the normal operation of the embodiment of the present application will be obtained.

[0036] Figure 1 is an optional flowchart of the model-based data processing method provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S104.

[0037] S101, obtaining institutional characteristic data of a plurality of insurance institutions, wherein the plurality of insurance institutions have a dependency relationship;

[0038] S102: Inputting the institution characteristic data into a pre-trained risk-bearing capacity assessment model to obtain a risk-bearing capacity index corresponding to each insurance institution, wherein the risk-bearing capacity index is positively correlated with the insurance institution's risk tolerance;

[0039] S103: Setting differentiated thresholds for the multiple insurance institutions based on the risk-bearing capacity index to obtain a corresponding reinsurance excess loss threshold for each insurance institution; the reinsurance excess loss threshold is used to represent the maximum compensation resource amount of the insurance institution under a preset insurance compensation scenario;

[0040] S104: Determine the compensation ratio of each insurance institution under the preset insurance compensation scenario based on the reinsurance excess loss threshold.

[0041] Steps S101 to S104 shown in the embodiment of the present application are performed by obtaining the institutional characteristic data of multiple insurance institutions with a dependent relationship. These institutional characteristic data are input into the pre-trained risk-bearing capacity assessment model to obtain the risk-bearing capacity index of each insurance institution, and the risk-bearing capacity index is positively correlated with the insurance institution's ability to bear risks. Based on the risk-bearing capacity index, differentiated thresholds are set for multiple insurance institutions to obtain the reinsurance excess loss threshold corresponding to each insurance institution, thereby accurately reflecting the actual risk-bearing level of each institution. Furthermore, the compensation ratio of each insurance institution under the preset insurance compensation scenario is determined based on these reinsurance excess loss thresholds, ensuring that the compensation ratio can fully adapt to the differentiated risk characteristics of different institutions within the insurance company and realize the reasonable sharing of risks. This model-based evaluation method avoids or reduces the subjectivity and limitations caused by reliance on manual experience in related technologies, thereby significantly improving the accuracy of determining the compensation ratio.

[0042] In step S101 of some embodiments, the insurance institution may be a company that has passed the review of an insurance regulatory authority, obtained a license to operate insurance business, and registered with the industrial and commercial administration authorities to operate insurance business. In this embodiment of the present application, the insurance institution includes the insurance company headquarters and multiple subordinate branches with close business ties to the headquarters. These branches are interdependent and mutually influential with the headquarters in terms of business operations, risk assumption, and management.

[0043] Specifically, a dependency relationship between multiple insurance institutions refers to the mutual dependence and influence of insurance institutions in terms of business operations, risk assumption, and management. For example, between an insurance company's headquarters and its subordinate branches, the headquarters is responsible for overall business planning and major business decisions, while the branches are responsible for the development and execution of specific businesses. The business performance of the branches directly affects the headquarters' performance. In terms of risk assumption, the risks faced by branches may affect the risk profile of the entire company, and the headquarters needs to comprehensively manage and coordinate the risks of the branches. In terms of management, the management policies and systems formulated by the headquarters must be complied with and implemented by the branches. At the same time, the branches' operational data and risk information must be promptly fed back to the headquarters for effective management and decision-making.

[0044] Optionally, in the embodiment of the present application, the institutional characteristic data is multi-dimensional information such as historical operating data, financial data, and risk data that comprehensively characterizes the insurance institution's operating capabilities and risk management capabilities. Specifically, the institutional characteristic data may include:

[0045] Data related to institutional size: such as total transaction resources and number of branches (reflecting the business volume and economic strength of insurance institutions).

[0046] Financial health: such as actual resources held, solvency ratio, and net profit (measures the amount of trading resources held and profitability of insurance institutions);

[0047] Risk performance: claims ratio, distribution of claims resources, and risk event records (to assess the insurance institution's historical risk exposure and management effectiveness);

[0048] Management capabilities: maturity of the risk control system, compliance record, and management risk awareness (reflecting the insurance institution's risk control level);

[0049] External environment: economic growth rate, regulatory policies, and frequency of natural disasters in the region (influencing regional risk characteristics).

[0050] Other auxiliary data: Customer satisfaction and complaint rate reflect service quality and customer relationship management capabilities; management risk awareness and capabilities include the risk management experience and training records of the senior management team.

[0051] Optionally, in a feasible implementation of the present application, obtaining the institutional characteristic data of multiple insurance institutions with dependencies is the basis of the entire model-based data processing method. This process first requires clarifying the source of the data, which mainly covers the insurance company's internal operating system, financial statements, risk management department, and external industry databases and market research reports. From the internal operating system, the daily business data and operational details of the insurance institution can be obtained. When collecting data, data extraction tools can be used to integrate and clean data scattered in different systems and sources to ensure the consistency and accuracy of the data. The data extraction tool can be a specially developed program or script used to extract the required institutional characteristic data from various data sources.

[0052] To ensure data quality and relevance, the collected data can also be preprocessed. This can include data cleaning, which removes erroneous and duplicate data records; data integration, which combines data from multiple sources into a consistent dataset; data transformation, such as standardizing or normalizing data to enable comparison and analysis of data of different dimensions within the model; and data reduction, which reduces data volume and improves data processing efficiency while preserving the original characteristics of the data as much as possible.

[0053] After data collection, extraction, and preprocessing, a structured dataset will be generated, containing multi-dimensional institutional characteristics that characterize insurance institutions' operational and risk management capabilities. This data will serve as input to the risk-bearing capacity assessment model in step 102, laying a solid foundation for accurately evaluating the risk-bearing capacity of each insurance institution.

[0054] In step S102 of some embodiments, the pre-trained risk-bearing capacity assessment model is a tool for assessing an insurance institution's risk-bearing capacity, pre-trained using a large amount of historical data and a machine learning algorithm. The risk-bearing capacity assessment model takes the insurance institution's institutional characteristic data as input. By learning the relationship between each characteristic in the historical data and the risk-bearing capacity index, the risk-bearing capacity assessment model can assess the characteristic data of a new insurance institution and output a corresponding Risk Bearing Capacity Index (RBCI).

[0055] The Risk Bearing Capacity Index (RBI) is a quantitative indicator output by a pre-trained risk bearing capacity assessment model that comprehensively reflects an insurance institution's risk tolerance. The index is derived through weighted calculations and other processing of an insurance institution's multi-dimensional characteristic data, transforming complex institutional characteristic data into an easily understandable and comparable numerical value. The RBI can be used to rank and categorize the risk bearing capacity of different insurance institutions, providing a basis for subsequent setting of differentiated thresholds and developing reinsurance excess loss plans. For example, an insurance institution with a high RBI typically demonstrates strong performance in terms of transaction resource adequacy and risk management, and is therefore able to withstand higher levels of risk.

[0056] The positive correlation between the risk-bearing capacity index and the insurance institution's risk tolerance means that there is a positive linear relationship between the value of the risk-bearing capacity index and the insurance institution's actual risk tolerance. Specifically, the higher the risk-bearing capacity index, the greater the loss the insurance institution can withstand when faced with a risk; conversely, the lower the risk-bearing capacity index, the smaller the loss the insurance institution can withstand. This positive correlation enables the risk-bearing capacity index to accurately and intuitively reflect the insurance institution's risk tolerance level, providing reliable data support for the formulation of reasonable reinsurance excess loss plans. For example, when an insurance institution's risk-bearing capacity index increases from 60 to 80, it indicates that the insurance institution's maximum compensation resources that can be independently borne have also increased, and its compensation ratio in the reinsurance excess loss plan may also increase accordingly.

[0057] Optionally, in an embodiment of the present application, the output of the risk-bearing capacity assessment model mainly includes a risk-bearing capacity index and related risk assessment reports and recommended measures. RBCI is a comprehensive scoring indicator obtained through quantitative analysis and weighted calculation, which is used to measure the insurance institution's ability to bear risks. The calculation method can be to standardize indicators of multiple dimensions such as capital adequacy, profitability, institutional size, historical claims stability and risk management level, multiply them by corresponding weights and add them up. For example, RBCI = w1×capital adequacy + w2×profitability + w3×institutional size + w4×historical claims stability + w5×risk management level, where the sum of each weight is 1, and the weight is set according to business strategy and risk preference. The value range of RBCI can be standardized to 0-100 or 0-1. The higher the value, the stronger the risk-bearing capacity of the insurance institution.

[0058] In addition to the RBCI, the model's outputs also include a risk-bearing capacity assessment report and risk improvement recommendations. The risk assessment report provides more detailed information, including basic information about the insurance institution, its RBCI score and its scores and rankings for each dimension, a risk analysis (where the institution performs well and where it has weaknesses), a trend analysis (analyzing the changing trends in the institution's risk-bearing capacity compared with historical data), and risk warning signals (such as insufficient trading resource holdings, abnormally high losses ratios, and weak risk management mechanisms). The risk improvement recommendations, based on the assessment results, propose specific measures to enhance risk-bearing capacity, such as increasing trading resource holdings or optimizing the trading resource structure, improving profitability or controlling costs, strengthening risk management systems, optimizing reinsurance arrangements, and adjusting business structures or risk exposures. These recommendations are prioritized by importance and implementation difficulty to facilitate decision-making. These outputs provide insurance institutions with a scientific, comprehensive, and actionable basis for developing internal reinsurance excess-of-loss plans. They help them accurately understand the risk-bearing capacity of each insurance institution, thereby ensuring the rational allocation and management of risk, optimizing trading resource allocation, and improving overall risk control and profitability.

[0059] Optionally, in one feasible implementation of the present application, the collection, cleaning, and preprocessing of the institution's characteristic data are first ensured, so that its format is consistent with the model input requirements. The pre-trained risk-bearing capacity assessment model is pre-trained using a large amount of historical data and machine learning algorithms, and is capable of learning the complex relationship between the various characteristics of an insurance institution and its risk-bearing capacity. The risk-bearing capacity assessment model takes the insurance institution's institutional characteristic data as input and processes the input data through complex internal operations, such as weighted calculations and nonlinear transformations. The risk-bearing capacity index can be calculated using a weighted scoring method, where the indicators of each dimension are standardized, multiplied by the corresponding weights, and finally summed. A higher value indicates a stronger risk tolerance of the insurance institution, i.e., the risk-bearing capacity index is positively correlated with the insurance institution's risk tolerance. These indices will serve as the basis for setting differentiated thresholds for multiple insurance institutions in the subsequent step S103, thereby providing key support for developing a reasonable reinsurance excess loss plan.

[0060] In step S103 of some embodiments, the reinsurance excess loss threshold is a key indicator used to define the maximum compensation resources an insurance institution can independently bear when handling an insurance claim event. A higher risk-bearing capacity index indicates stronger insurance institutions and better risk management practices, and thus a higher corresponding reinsurance excess loss threshold.

[0061] In some optional implementations, first, institutions can be divided into three risk levels based on the RBCI values: high (≥80), medium (60-79), and low (<60), and a basic "capability-threshold" mapping relationship can be established. Subsequently, the basic threshold is determined based on the risk level. Specifically, the basic threshold can be set based on the risk-bearing capacity index (RBCI), and high, medium, and low are respectively corresponding to the differentiated calculation rules of "trading resource holdings × (0.3 to 0.5)", "trading resource holdings × (0.2 to 0.3)", and "trading resource holdings × (0.1 to 0.2)", realizing the basic logic of "the stronger the capability, the higher the threshold for self-assessment".

[0062] Secondly, by introducing business type and regional risk adjustment factors, the basic threshold is dynamically revised to obtain the final reinsurance excess loss threshold. For example, due to the better risk diversification of health insurance business, the reinsurance excess loss threshold can be increased by 20% from the basic value. However, for auto insurance institutions in high-risk areas (such as earthquake-prone areas), the reinsurance excess loss threshold needs to be reduced by 15%. At the same time, trigger conditions are set based on the real-time operating data of insurance institutions: if an institution's loss ratio exceeds 70% for two consecutive months, regardless of its RBCI level, the reinsurance excess loss threshold is automatically lowered to the transaction resource holdings × 0.1; if the transaction resource holdings increase by more than 20%, the reinsurance excess loss threshold is recalculated based on the new transaction resource holdings.

[0063] In step S104 of some embodiments, pre-set compensation scenarios refer to various possible insurance compensation situations or scenarios pre-defined when formulating a reinsurance excess of loss plan. These scenarios can be defined based on historical data, industry experience, and specific business needs, such as different types of insurance accidents (e.g., natural disasters, traffic accidents), different compensation amount ranges, and different business lines or products. The purpose of pre-set compensation scenarios is to clarify the compensation liability of each insurance institution in different situations.

[0064] The compensation ratio refers to the proportion of the total compensation payable by each insurance institution, based on its reinsurance excess loss threshold, under these pre-defined compensation scenarios. This compensation ratio ensures fair risk sharing, preventing any single institution from bearing excessive compensation pressure, and improving the stability and risk resilience of the entire insurance system.

[0065] A reinsurance excess loss plan is a detailed plan developed by an insurance company for multiple dependent insurance institutions, which comprehensively stipulates the compensation ratio and fund allocation process of each insurance institution when an insurance claim event occurs.

[0066] The transaction resource allocation mechanism specifies how each insurance institution will allocate the compensation resources it should bear to the insurance institution that needs compensation after a compensation event occurs. It involves details such as the allocation method (such as internal transfer or post-settlement), the allocation time limit (such as within a few working days after the review is passed) and the allocation account to ensure that funds are in place in a timely and accurate manner.

[0067] Optionally, in a feasible implementation of the present application, first, the system automatically identifies the type of excess loss scenario (single / cumulative / special event excess loss) based on the reinsurance excess loss threshold. For example, when a branch insurance institution's single loss reaches 600,000 yuan (reinsurance excess loss threshold is 500,000 yuan), the regular excess loss process is triggered; if the quarterly cumulative loss exceeds the threshold by 200% due to a natural disaster, it is determined to be a special event excess loss (i.e., a preset loss scenario), and the emergency apportionment rule is activated. Among them, the regular excess loss process refers to a standardized apportionment process executed according to a preset compensation ratio (such as dynamic apportionment between the headquarters and branches according to RBCI) for general situations where single or cumulative losses exceed the threshold; the emergency apportionment rule refers to a special apportionment mechanism for excess losses caused by special events such as natural disasters (such as exceeding the threshold by 200%) by increasing the total share ratio (such as from 60% to 70%) and introducing external reinsurance resources.

[0068] The compensation ratio can be dynamically adjusted according to the insurance institution's risk bearing capacity index (RBCI) and the excess loss level: for conventional excess loss scenarios, the part exceeding the threshold is shared by the headquarters and branches according to the principle of "inverse correlation of RBCI weights". For example, the formula can be: headquarters' share = 60% - (RBCI × 0.5%). For example, for an insurance institution with an RBCI of 70, the headquarters bears 60% - 35% = 25%, and the remaining part is shared by other institutions according to the proportion of transaction resources held (for example, if institution A holds 40% of transaction resources, it will bear 40%); for special event excess loss, the headquarters' share is increased to 70% for the part exceeding the threshold by more than 150%, and external reinsurance resources are introduced to share 20% to disperse systemic risks.

[0069] The transaction resource allocation mechanism can be automated by connecting the internal financial system with the risk management platform: after the over-compensation event is reviewed and confirmed by the headquarters, the system automatically generates a transaction resource allocation instruction, and transfers transaction resources from the headquarters transaction resource pool and branch accounts to the compensation institution according to the apportionment ratio. For example, if Institution B has an over-compensation of 100,000 yuan, the headquarters will bear 60,000 yuan and Institution A will bear 40,000 yuan. The system will automatically transfer 60,000 yuan from the headquarters account to Institution B and send a deduction notice to Institution A. At the same time, the solution has a built-in dispute arbitration module. If an institution has any objection to the apportionment ratio, it can submit evidence through the platform. The headquarters risk committee will recalculate and force an update of the results through the model within 10 working days to ensure that the process is transparent and traceable.

[0070] Furthermore, this application supports dynamic version management, updating compensation ratios and allocation parameters annually based on risk-bearing capacity assessments. For example, if an institution's RBCI increases from 65 to 75, its compensation ratio in conventional excess-of-loss insurance will decrease from 35% to 25%. By digitally integrating thresholds, scenarios, ratios, and processes, S104 achieves automated management of the entire risk chain, from risk event triggering to fund settlement. This significantly improves the execution efficiency and risk diversification effectiveness of reinsurance excess-of-loss programs, providing a precise and efficient operational framework for insurance companies' internal risk management.

[0071] In one embodiment, the institution characteristic data includes the insurance institution's historical average single claim payment amount;

[0072] The step of setting differentiated thresholds for the multiple insurance institutions based on the risk-bearing capacity index to obtain a corresponding reinsurance excess loss threshold for each insurance institution includes at least one of the following:

[0073] Determining the risk level corresponding to each insurance institution based on the risk-bearing capacity index; setting differentiated thresholds for the multiple insurance institutions based on the risk level to obtain a reinsurance excess loss threshold corresponding to each insurance institution; the risk level is positively correlated with the risk-bearing capacity of the insurance institution;

[0074] For any insurance institution, the compensation ratio is determined based on the risk-bearing capacity index corresponding to the insurance institution; the product of the compensation ratio and the historical average single compensation amount of the insurance institution is determined as the reinsurance excess loss threshold corresponding to the insurance institution.

[0075] Optionally, in this embodiment of the present application, the risk level is a hierarchical classification based on the insurance institution's risk tolerance index (RBCI), which is used to intuitively reflect the institution's risk tolerance and is positively correlated with risk tolerance. By dividing the risk levels, a basic framework is provided for setting differentiated thresholds.

[0076] The historical average single claim payment refers to the average amount paid per claim case by an insurance institution over a past period (e.g., 3-5 years). This data, extracted from the institution's characteristic data, provides a quantitative summary of the institution's historical claims performance, reflecting the institution's typical risk exposure level in daily operations. For example, if an institution's historical average single claim payment is 500,000 yuan, it indicates that its single claims payments in the past have tended to be concentrated around this amount. This can serve as a reference for measuring future claims volumes and is an important foundational data for calculating reinsurance excess loss thresholds.

[0077] The claims ratio is a multiplier determined by an insurance institution's risk tolerance index or risk rating. It is used to amplify or reduce the historical average single claim payment to determine a reinsurance excess loss threshold that matches the institution's risk tolerance. Higher risk ratings or larger RBCI values ​​correspond to higher claims ratios, providing a quantitative link between risk tolerance and claims liability.

[0078] In these optional embodiments, by incorporating the insurance institution's historical average single claim amount into its institutional characteristic data and combining it with a risk-bearing capacity index to set differentiated thresholds, a more precise match can be achieved between each insurance institution's actual risk tolerance and claims level. On the one hand, by setting differentiated excess-of-loss thresholds based on risk level, institutions with high risk tolerance bear higher claims liability, thereby optimizing risk allocation. On the other hand, by introducing the product of the claims ratio and the historical average single claim amount to determine the excess-of-loss threshold, the threshold dynamically adapts to the insurance institution's business scale and risk characteristics, enhancing the sophistication of risk assessment and management, and strengthening the scientific and rationale of reinsurance excess-of-loss programs.

[0079] In one embodiment, if Figure 2 As shown, the institutional characteristic data includes the actual holdings of the insurance institution;

[0080] The step of determining the risk level corresponding to each insurance institution based on the risk-bearing capacity index; and setting differentiated thresholds for the multiple insurance institutions based on the risk levels to obtain a reinsurance excess loss threshold corresponding to each insurance institution, including:

[0081] For any insurance institution, execute the following:

[0082] S201, based on a preset grading rule, grading the insurance institution according to the risk bearing capacity index corresponding to the insurance institution to obtain the risk grade corresponding to the insurance institution;

[0083] S202: Determine the product of the actual resources held by the insurance institution and the multiplication coefficient corresponding to the risk level of the insurance institution as the reinsurance excess loss threshold corresponding to the insurance institution; the multiplication coefficient is positively correlated with the risk level.

[0084] Optionally, in a specific implementation of the present application, first, the risk level classification of insurance institutions based on the risk-bearing capacity index is the basis of the entire process. The system automatically maps the RBCI values ​​of each insurance institution to the corresponding risk level through preset level classification rules (such as RBCI ≥ 80 is classified as high risk level, 60 ≤ RBCI < 80 is classified as medium risk level, and RBCI < 60 is classified as low risk level), forming a three-level classification system of "high-medium-low". This process can be automatically completed by an algorithm, such as using a decision tree model to cut the RBCI interval to ensure the consistency and traceability of the level classification.

[0085] After obtaining the risk level, differentiated threshold calculation rules can be used for institutional characteristic data in different scenarios:

[0086] For example, a threshold calculation can be based on the historical average single claim amount. Specifically, for any insurance institution, the system first extracts its historical average single claim amount from the institution's characteristic data (e.g., the average over the past three or five years). This data reflects the typical risk exposure level of the insurance institution in its daily operations. Next, the corresponding loss ratio is determined based on the insurance institution's risk level. The higher the risk level, the larger the loss ratio (e.g., 5-10 times for high risk, 3-5 times for medium risk, and 1-3 times for low risk). For example, if a high-risk institution has a historical average single claim amount of 1 million yuan and a loss ratio of 8, its reinsurance excess loss threshold is 8 million yuan (1 million x 8). This calculation method aligns the reinsurance excess loss threshold with the insurance institution's actual claims-paying capacity. High-risk institutions, due to their greater risk tolerance, can assume a higher multiple of their historical average claims, while low-risk institutions limit their risk exposure through lower ratios.

[0087] Calculations can also be based on a threshold of actual resources held. Specifically, actual resources held can include core indicators such as an institution's trading resource holdings and retained earnings, and serve as the financial foundation for an insurance institution's risk mitigation efforts. The system sets a predefined multiplication factor for each risk level (e.g., 0.4-0.5 for high risk, 0.2-0.3 for medium risk, and 0.1-0.2 for low risk). This multiplication factor is positively correlated with risk level, meaning that higher risk levels indicate a higher proportion of trading resource holdings available for risk mitigation. For example, if an institution with a medium risk rating has actual resources held of 80 million yuan and a multiplication factor of 0.25, its reinsurance excess loss threshold is 20 million yuan (80 million x 0.25). This calculation method, based directly on actual resources held, ensures that the reinsurance excess loss threshold is within the insurance institution's financial tolerance. Furthermore, differentiated multiplication factors across different risk levels prevent idle or overloaded actual resources.

[0088] In its implementation, the system automatically retrieves an insurance institution's historical average single claim amount and actual held resource data through a data interface, dynamically matching it with its risk level. For example, when an insurance institution's RBCI increases from 70 to 85 (risk level rises from medium to high), the system automatically adjusts its claim ratio from 4x to 8x, and simultaneously adjusts the multiplication factor from 0.25 to 0.4. The recalculated reinsurance excess loss thresholds increase from 4 million yuan (1 million x 4) and 20 million yuan (80 million x 0.25) to 8 million yuan and 32 million yuan (80 million x 0.4), respectively. This reflects the dynamic release of independent underwriting authority as risk-bearing capacity increases.

[0089] In addition, the system supports a threshold calculation model that integrates multiple rules. For example, it simultaneously considers historical claims data and actual holdings, taking the lower value of the two as the final threshold to further ensure that risks are controllable. The entire process, from data acquisition and grading to threshold calculation, can be implemented through an automated rules engine. The entire process, from data acquisition and grading to threshold calculation, requires no human intervention and supports real-time updates. When the institution's characteristic data or risk level changes, the reinsurance excess loss threshold will be adjusted synchronously and trigger an update to the reinsurance plan. This ensures that the reinsurance excess loss mechanism always matches the institution's actual risk profile, improving the accuracy and timeliness of the insurance company's overall risk management.

[0090] In these optional embodiments, by incorporating the actual resources held by insurance institutions into their characteristic data and combining them with a risk-bearing capacity index to categorize risk levels, each insurance institution's actual risk tolerance can be more accurately reflected. A multiplication coefficient is determined based on the risk level, and the reinsurance excess loss threshold is calculated by multiplying the actual resources held by this coefficient. This allows for differentiated threshold settings for institutions of different risk levels, ensuring that higher-risk institutions bear higher claims liability and enabling more rational risk allocation.

[0091] In one embodiment, if Figure 3 As shown, the institutional characteristic data includes the actual holdings and loss ratio of the insurance institution;

[0092] The plurality of insurance institutions include a first institution and subordinate institutions of the first institution;

[0093] The method further comprises:

[0094] S301, re-evaluating the reinsurance excess loss threshold of each insurance institution according to a preset update frequency or when a preset update condition is triggered;

[0095] S302, updating the compensation ratio according to the reassessed reinsurance excess loss threshold of each insurance institution to obtain an updated compensation ratio;

[0096] The preset update condition includes at least one of the following:

[0097] The change in the risk-bearing capacity index of the subordinate institution is greater than or equal to the first threshold;

[0098] The decrease in the actual resources held by the subordinate institution is greater than or equal to the second threshold;

[0099] The decrease in the claims ratio of the subordinate institution is greater than or equal to a third threshold.

[0100] Optionally, in the embodiments of the present application, the claims ratio is a key indicator for measuring an insurance institution's claims payment situation. It refers to the ratio of the insurance institution's claims resources paid to its premium income over a certain period (e.g., annually). For example, if an institution's premium income is 10 million yuan and its claims resources are 6 million yuan, its claims ratio is 60%. As one of the institution's characteristic data, the claims ratio reflects the insurance institution's actual claims costs and risk profile and can be used to assess its risk-bearing capacity and reinsurance needs.

[0101] A primary institution is the dominant or top-level management organization within an insurance system. It can be a group headquarters, parent company, or a higher-level organization with management functions. For example, an insurance group company, as the primary institution, is responsible for coordinating the business strategies, risk control, and reinsurance arrangements of its subsidiaries (subordinate institutions), and has management responsibility for the operations and risk profile of its subordinate institutions.

[0102] A subordinate institution is a subordinate organization under the primary institution and can be a subsidiary, branch, or sub-branch of the primary institution. For example, provincial branches and regional sub-branches under the primary institution (head office) are all subordinate institutions. Subsidiaries directly participate in the underwriting and claims settlement of specific insurance businesses. Their risk-bearing capacity, resource availability, and loss ratios influence the primary institution's formulation and adjustment of reinsurance plans.

[0103] Optionally, in a specific implementation of the present application, a closed-loop mechanism of dynamic monitoring - threshold reassessment - plan update can be used to achieve real-time adaptation of the reinsurance excess loss system to the institution's risk profile. The specific implementation is as follows:

[0104] First, the system collects core indicators such as the risk-bearing capacity index, actual holdings, and loss ratio of subordinate institutions in real time through a data interface, and compares them with preset update conditions. The preset update conditions use a dual trigger mechanism of "absolute value + ratio".

[0105] For example, if the change in RBCI is ≥ the first threshold (such as ±15%), when the RBCI of a subsidiary institution rises from 70 to 85 (an increase of 21%) or falls from 80 to 65 (a decrease of 18.75%), a threshold reassessment is triggered; if the decrease in actual held resources is ≥ the second threshold (such as 20%), if the capital of a subsidiary institution falls from 100 million yuan to 75 million yuan (a decrease of 25%), it indicates that its risk-resistant capital has been weakened and needs to be reassessed; if the decrease in the losses ratio is ≥ the third threshold (such as -30%), a decrease in the losses ratio from 60% to 40% (a decrease of 33.3%) may reflect business structure optimization or data anomalies, and the threshold needs to be adjusted after verification.

[0106] When any condition is met or the preset update frequency (such as the first day of the first month of each quarter) is reached, the system automatically initiates the threshold reassessment process. Specifically, the risk level of the subsidiary institution is reclassified based on the latest RBCI (for example, if the RBCI of a medium-risk institution rises to 82, it is adjusted to a high-risk level); the latest actual holdings (such as 120 million yuan after capital changes) and loss ratio (such as 55% in the latest quarter) are extracted. If the "actual holdings × multiplication coefficient" rule is used, the multiplication coefficient corresponding to the high-risk level is adjusted from 0.3 to 0.4; and the reinsurance excess loss threshold is recalculated based on the updated level and data. For example, if a subsidiary institution has actual holdings of 80 million yuan and its risk level rises from medium (multiplication coefficient 0.25) to high (0.4), the threshold will increase from 20 million yuan (80 million yuan × 0.25) to 32 million yuan (80 million yuan × 0.4).

[0107] After the threshold reassessment is completed, the system automatically updates the reinsurance excess loss plan based on the new threshold. Specifically, if the threshold of a subsidiary institution is increased from 5 million yuan to 8 million yuan, the upper limit of the amount it can independently bear in conventional excess loss scenarios will be increased, and the headquarters' compensation ratio for the excess amount will be reduced accordingly (for example, from 60% to 50%). The updated threshold is synchronized with the financial system. When an excess loss event occurs, the system calculates the transaction resources that each institution should bear according to the new compensation ratio (for example, if the excess loss is 1 million yuan, the subsidiary institution will bear 500,000 yuan and the headquarters will bear 500,000 yuan). If the subsidiary institution triggers an update due to a decrease in the loss ratio, it is necessary to verify whether its business type has changed (for example, the increase in the proportion of health insurance has led to a decrease in the loss ratio), and adjust the threshold calculation rules for the corresponding insurance type (for example, the health insurance loss ratio will increase from 5 times to 6 times).

[0108] The entire process can be implemented without human intervention through an automated engine: data collection, conditional judgment, threshold calculation, and plan updates are all executed by the system according to preset logic, and an audit log is generated to record the details of the changes. For threshold updates involving the primary institution (headquarters), an additional management approval process can be triggered (such as electronic approval within two business days) to ensure the compliance of major adjustments. Through this dynamic mechanism, the reinsurance excess loss plan can respond to changes in institutional risks in real time, avoiding the risk control lag caused by static rules, and improving the insurance company's overall risk resilience and capital operation efficiency.

[0109] In these optional embodiments, by incorporating the actual resources held and loss ratios of insurance institutions into institutional profile data and setting preset update frequencies and conditions for the first institution and its subordinate institutions, the method can promptly capture changes in the risk profiles of insurance institutions. When a subordinate institution's risk-bearing capacity index fluctuates significantly, actual resources held decline significantly, or the loss ratio declines significantly, a reassessment of each insurance institution's reinsurance excess loss threshold is triggered. Based on the reassessed threshold, the compensation ratio is updated, thereby updating the reinsurance excess loss plan. This ensures that the plan dynamically adapts to the latest risk characteristics of each institution, maintains the rationality of risk allocation, and safeguards the stability of the insurance company's overall operations and risk control capabilities.

[0110] In one embodiment, the plurality of insurance institutions include a first institution and at least one subordinate institution of the first institution;

[0111] After determining the compensation ratio of each insurance institution under the preset insurance compensation scenario based on the reinsurance excess loss threshold, the method further includes:

[0112] In the event that the amount of compensation resources of the second institution exceeds the reinsurance excess loss threshold of the second institution, the amount of compensation resources corresponding to each of the first institution and the third institution shall be determined based on the compensation ratio; the second institution shall be any subordinate institution of the at least one subordinate institution, and the third institution shall be a subordinate institution of the at least one subordinate institution other than the second institution. The amount of compensation resources corresponding to each of the first institution and the third institution shall be the excess of the amount of compensation resources.

[0113] Optionally, in an embodiment of the present application, the multiple insurance institutions include a first institution serving as the headquarters or superior management entity, and its subordinate branches. The logic for setting the compensation ratio is as follows: when an insurance claim is incurred by a subordinate institution and the claim resources exceed its own reinsurance excess loss threshold, the excess claim resources will be shared by the first institution and the other subordinate institutions. The specific compensation ratio is determined based on the risk-bearing capacity index of each insurance institution and institutional characteristic data (such as actual held resources, loss ratio, etc.). For example, if the reinsurance excess loss threshold of a subsidiary institution A is 5 million yuan and a claim requires an expenditure of 8 million yuan, the excess 3 million yuan will be apportioned according to preset rules. If the first institution has a high risk-bearing capacity index and strong capital, it may bear 40% (i.e., 1.2 million yuan); another subsidiary institution B with weaker risk-bearing capacity may bear 10% (i.e., 300,000 yuan); the remaining portion will be shared by other institutions according to the rules, thereby achieving a reasonable distribution of risk among the internal institutions.

[0114] Optionally, in a specific implementation of the present application, the compensation ratio is determined based on the risk bearing capacity index (RBCI) and institutional characteristic data of each institution, and differentiated distribution is achieved through a quantitative model. Specifically, the RBCI value can be converted into an apportionment weight, and the formula is: the compensation ratio of a certain insurance institution = 1-(RBCI / 100), that is, the higher the RBCI, the lower the proportion of the excess. For example, the RBCI of the first institution is 90, and the compensation ratio is 10% (1-90 / 100); the RBCI of the subordinate institution A is 70, and the compensation ratio is 30%; the RBCI of the subordinate institution B is 50, and the compensation ratio is 50%.

[0115] Compensation can also be reversed based on the proportion of actual resources held by the institutions. The formula is: Compensation ratio for a specific insurance institution = (Headquarters Actual Resources / Total Actual Resources) × Adjustment Coefficient. Assuming total actual resources held are 200 million yuan (Headquarters 120 million yuan, Institution A 50 million yuan, Institution B 30 million yuan), and headquarters' actual resources account for 60%, with an adjustment coefficient of 0.8. The total share is 48% (60% × 0.8), Institution A 20% (25% × 0.8), and Institution B 12% (15% × 0.8). Compensation ratios can also be adjusted based on real-time data such as claims ratios. For example, if a subsidiary's claims ratio has increased by 15% over the past three months, its compensation ratio will automatically increase by 5%.

[0116] Optionally, in other real-time methods of the present application, multiple insurance institutions are first determined, including a first institution and at least one of its subordinate institutions. Then, the compensation ratio of each insurance institution under the preset compensation scenario is determined based on the reinsurance excess loss threshold. Then, when the compensation resource amount of the subordinate institution (second institution) exceeds its reinsurance excess loss threshold, the compensation resource amount to be borne by the first institution and other subordinate institutions (third institution) is calculated based on the previously determined compensation ratio. Specifically, the excess compensation resources will be distributed between the first institution and other subordinate institutions according to the preset compensation ratio. The third institution is the remaining subordinate institution, and the third institution and the first institution share the excess compensation resources. In this way, each insurance institution can reasonably share the compensation liability according to its own risk-bearing capacity and compensation capacity, avoid individual institutions from bearing excessive compensation pressure, and achieve effective dispersion of risks among multiple institutions.

[0117] In these optional embodiments, when an insurance institution's claims resources exceed its reinsurance excess loss threshold, the corresponding claims resources of the first and other insurance institutions are determined based on a preset compensation ratio. This mechanism ensures that the excess claims are reasonably distributed among multiple insurance institutions. Each insurance institution shares the claims responsibility based on its own risk-bearing capacity and institutional characteristics. This not only ensures the timeliness and accuracy of claims payments, but also avoids the financial pressure on a single insurance institution due to excessive claims, thereby improving the ability of the entire insurance system to cope with large claims events and enhancing the flexibility and adaptability of the risk-sharing mechanism.

[0118] In one embodiment, if Figure 4 As shown, the risk-bearing capacity assessment model training process includes:

[0119] S401, determining a plurality of training samples based on historical operating data of a plurality of insurance institutions, wherein the historical operating data includes at least one of institution size, profitability, historical claims data, risk management capabilities, and external environmental factors;

[0120] S402, labeling each training sample with a corresponding risk-bearing capacity index to obtain a labeled training sample;

[0121] S403, based on the preset learning algorithm, with the characteristic data of the training sample as the input variable and the labeled risk-bearing capacity index as the target variable, the preset learning algorithm is trained until the preset training stop condition is met, thereby obtaining a trained risk-bearing capacity assessment model.

[0122] Optionally, in a specific implementation of the present application, the training process of the risk-bearing capacity assessment model can be implemented through a standardized process of "data collection - annotation system - algorithm training - verification and optimization", and the specific steps are as follows:

[0123] First, multi-dimensional features are extracted from the historical operating data of insurance institutions to construct training samples. The data covers institutional size, profitability, historical claims data, risk management capabilities and external environmental factors. Through data cleaning, outliers are eliminated and standardized (such as Z-score normalization) to form a multi-dimensional feature matrix.

[0124] Subsequently, a labeling system is established to label training samples with a Risk Bearing Capacity Index (RBCI). This approach employs a hybrid labeling approach combining expert scoring and data verification: risk management experts assess risk capacity based on information such as annual reports and audit reports. Historical risk events (such as reinsurance triggering and changes in capital adequacy ratios) are also incorporated as verification factors, ensuring label consistency through cross-validation. Once labeling is complete, the training, validation, and test sets can be divided in a 7:2:1 ratio to ensure balanced sample distribution.

[0125] During the model training phase, the gradient boosting tree can be used as the preset learning algorithm, and the XGBoost framework can be used to achieve automatic feature selection and nonlinear relationship fitting. The root mean square error is used as the loss function during training, and the training is automatically terminated when the validation set does not decrease for five consecutive iterations. After the training is completed, the model is analyzed for interpretability, and the importance of each feature can be quantified by the SHAP value. For example, the influence weight of core indicators such as the scale of trading resource holdings (contribution rate 25%), the volatility of the claims ratio in the past three years (20%), and the maturity score of the risk control system (18%) is significantly higher than that of other features. At the same time, a model monitoring mechanism is established, and incremental training is performed using the latest data every month. When the proportion of new samples exceeds 10% or there are major changes in industry benchmark data (such as regulatory policy adjustments), full retraining is triggered to ensure that the model continues to adapt to industry dynamics.

[0126] In these optional embodiments, by incorporating historical operating data from insurance institutions into training samples and combining key characteristics such as their size, profitability, historical claims data, risk management capabilities, and external environmental factors, each sample is accurately labeled with a risk-bearing capacity index. A pre-set learning algorithm is then trained using these characteristic data as input variables and the labeled risk-bearing capacity index as the target variable. The algorithm is trained until pre-set training termination conditions are met, resulting in a trained risk-bearing capacity assessment model. Such a model accurately reflects the risk-bearing capacity of each insurance institution, providing a scientific basis for the subsequent determination of compensation ratios.

[0127] It should be noted that the various optional implementation methods introduced in the embodiments of the present application can be implemented in combination with each other or separately if they do not conflict with each other, and the embodiments of the present application do not limit this.

[0128] To facilitate understanding of the model-based data processing method provided by the above embodiment, the above model-based data processing method is described below using a specific scenario embodiment.

[0129] This application provides an internal, model-based data processing method designed to design flexible and efficient reinsurance solutions tailored to the risk-bearing capacity and specific business needs of an insurance company's headquarters and branches. By comprehensively considering factors such as risk exposure, capital adequacy ratios, and business structure, this method optimizes resource allocation and enhances overall risk control and profitability.

[0130] The specific steps include:

[0131] Data collection and analysis: First, collect diversified information such as historical claims data, business scale, financial status, risk assessment reports, etc. from the insurance company headquarters and its subsidiaries, and use big data technology to conduct in-depth mining and analysis to quantitatively assess the risk-bearing capacity of each institution.

[0132] Risk-bearing capacity assessment model construction: Based on the collected data, a risk-bearing capacity assessment model is constructed. This model comprehensively considers multiple dimensions such as institutional size, profitability, capital adequacy ratio, historical loss ratio, and risk management level, and uses weighted calculation to derive each institution's risk-bearing capacity index.

[0133] Internal reinsurance excess loss threshold setting: Based on the assessment results, appropriate internal reinsurance excess loss thresholds are set for the insurance company's headquarters and subsidiaries. These thresholds should reflect the principle of differentiation, with institutions with greater risk-bearing capacity bearing higher initial claims liabilities, while institutions with weaker risk-bearing capacity can trigger the internal reinsurance excess loss mechanism at lower claims limits.

[0134] Develop an internal reinsurance excess loss agreement: Based on the above threshold settings, develop a detailed internal reinsurance excess loss agreement to clarify the compensation ratio, compensation process, fund allocation mechanism, etc. of each institution under different compensation scenarios to ensure the effective operation of the internal reinsurance excess loss mechanism.

[0135] Dynamic adjustment and optimization: Regularly review and adjust the risk-bearing capacity assessment model and internal reinsurance excess loss agreement. Implement dynamic optimization based on changes in the market environment, adjustments in insurance companies' business strategies, and the actual operations of each institution to ensure the adaptability and effectiveness of the internal reinsurance excess loss mechanism.

[0136] See also Figure 5 The present application also provides a model-based data processing device that can implement the above-mentioned model-based data processing method. The device includes:

[0137] An acquisition module, configured to acquire institutional characteristic data of a plurality of insurance institutions, wherein the plurality of insurance institutions have a dependency relationship between them;

[0138] A first determination module is configured to input the institution characteristic data into a pre-trained risk-bearing capacity assessment model to obtain a risk-bearing capacity index corresponding to each insurance institution, wherein the risk-bearing capacity index is positively correlated with the insurance institution's ability to bear risk;

[0139] a setting module for setting differentiated thresholds for the multiple insurance institutions based on the risk-bearing capacity index to obtain a corresponding reinsurance excess loss threshold for each insurance institution; the reinsurance excess loss threshold is used to represent the maximum compensation resource amount of the insurance institution under a preset insurance compensation scenario;

[0140] The second determining module is used to determine the compensation ratio of each insurance institution under the preset insurance compensation scenario according to the reinsurance excess loss threshold.

[0141] The specific implementation of the model-based data processing device is basically the same as the specific embodiment of the above-mentioned model-based data processing method, and will not be repeated here.

[0142] The present application also provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-described model-based data processing method when executing the computer program. The electronic device can be any smart terminal, such as a tablet computer or an in-vehicle computer.

[0143] See also Figure 6 , Figure 6 The hardware structure of an electronic device according to another embodiment is shown. The electronic device includes:

[0144] The processor 601 may be implemented as a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application.

[0145] The memory 602 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 602 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 602 and is called by the processor 601 to execute the model-based data processing method of the embodiments of this application.

[0146] Input / output interface 603, used to implement information input and output;

[0147] Communication interface 604, used to implement communication interaction between this device and other devices, which can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WiFi, Bluetooth, etc.);

[0148] Bus 605 , which transmits information between various components of the device (e.g., processor 601 , memory 602 , input / output interface 603 , and communication interface 604 );

[0149] The processor 601 , the memory 602 , the input / output interface 603 and the communication interface 604 are connected to each other in communication within the device via a bus 605 .

[0150] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the above-mentioned model-based data processing method.

[0151] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0152] The embodiments described in the embodiments of this application are intended to more clearly illustrate the technical solutions of the embodiments of this application and do not constitute a limitation on the technical solutions provided by the embodiments of this application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0153] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0154] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, i.e., they may be located in one place or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of this embodiment.

[0155] Those skilled in the art will appreciate that all or some of the steps in the methods, systems, and functional modules / units in the devices disclosed above may be implemented as software, firmware, hardware, or appropriate combinations thereof.

[0156] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0157] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0158] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the above-mentioned units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0159] The units described above as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0161] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes multiple instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: various media that can store programs, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0162] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but are not intended to limit the scope of the present invention. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and essence of the present invention should be within the scope of the present invention.

Claims

1. A data processing method based on a model, characterized in that: The method comprises: Acquiring institutional characteristic data of a plurality of insurance institutions, wherein the plurality of insurance institutions have a dependency relationship; Inputting the institutional characteristic data into a pre-trained risk-bearing capacity assessment model to obtain a risk-bearing capacity index corresponding to each insurance institution, wherein the risk-bearing capacity index is positively correlated with the insurance institution's ability to bear risk; According to the risk-bearing capacity index, differentiated thresholds are set for the multiple insurance institutions to obtain a corresponding reinsurance excess loss threshold for each insurance institution; the reinsurance excess loss threshold is used to represent the maximum compensation resource amount of the insurance institution under a preset insurance compensation scenario; Based on the reinsurance excess loss threshold, the compensation ratio of each insurance institution under the preset insurance compensation scenario is determined.

2. The method according to claim 1, characterized in that The institutional characteristic data include the historical average single claim amount of the insurance institution; The step of setting differentiated thresholds for the multiple insurance institutions based on the risk-bearing capacity index to obtain a corresponding reinsurance excess loss threshold for each insurance institution includes at least one of the following: Determining the risk level corresponding to each insurance institution based on the risk-bearing capacity index; setting differentiated thresholds for the multiple insurance institutions based on the risk level to obtain a reinsurance excess loss threshold corresponding to each insurance institution; the risk level is positively correlated with the risk-bearing capacity of the insurance institution; For any insurance institution, the compensation ratio is determined based on the risk-bearing capacity index corresponding to the insurance institution; the product of the compensation ratio and the historical average single compensation amount of the insurance institution is determined as the reinsurance excess loss threshold corresponding to the insurance institution.

3. The method according to claim 2, characterized in that The institutional characteristic data includes the actual holdings of the insurance institution; Determining the risk level corresponding to each insurance institution based on the risk bearing capacity index; According to the risk level, differentiated thresholds are set for the multiple insurance institutions to obtain the reinsurance excess loss threshold corresponding to each insurance institution, including: For any insurance institution, execute the following: Based on the preset grading rules, the insurance institution is graded according to the risk bearing capacity index corresponding to the insurance institution to obtain the risk grade corresponding to the insurance institution; The product of the actual resources held by the insurance institution and the multiplication coefficient corresponding to the risk level of the insurance institution is determined as the reinsurance excess loss threshold corresponding to the insurance institution; the multiplication coefficient is positively correlated with the risk level.

4. The method according to claim 1, wherein The institutional characteristic data include the actual holdings and loss ratio of the insurance institution; The plurality of insurance institutions include a first institution and subordinate institutions of the first institution; The method further comprises: Reassess each insurer's excess loss threshold at a pre-set update frequency or upon triggering of pre-set update conditions; updating the compensation ratio according to the reassessed reinsurance excess loss thresholds of the respective insurance institutions to obtain an updated compensation ratio; The preset update condition includes at least one of the following: The change in the risk-bearing capacity index of the subordinate institution is greater than or equal to the first threshold; The decrease in the actual resources held by the subordinate institution is greater than or equal to the second threshold; The decrease in the claims ratio of the subordinate institution is greater than or equal to a third threshold.

5. The method according to claim 1, wherein The plurality of insurance institutions include a first institution and at least one subordinate institution of the first institution; After determining the compensation ratio of each insurance institution under the preset insurance compensation scenario based on the reinsurance excess loss threshold, the method further includes: In the event that the amount of compensation resources of the second institution exceeds the reinsurance excess loss threshold of the second institution, the amount of compensation resources corresponding to each of the first institution and the third institution shall be determined based on the compensation ratio; the second institution shall be any subordinate institution of the at least one subordinate institution, and the third institution shall be a subordinate institution of the at least one subordinate institution other than the second institution. The amount of compensation resources corresponding to each of the first institution and the third institution shall be the excess of the amount of compensation resources.

6. The method according to claim 1, characterized in that The risk-bearing capacity assessment model training process includes: determining a plurality of training samples based on the historical operating data of the plurality of insurance institutions; Label each training sample with the corresponding risk-bearing capacity index to obtain the labeled training sample; Based on a preset learning algorithm, the characteristic data of the training sample is used as an input variable, and the labeled risk-bearing capacity index is used as a target variable. The preset learning algorithm is trained until a preset training stop condition is met, thereby obtaining the trained risk-bearing capacity assessment model.

7. A data processing device based on a model, characterized in that: The device comprises: An acquisition module, configured to acquire institutional characteristic data of a plurality of insurance institutions, wherein the plurality of insurance institutions have a dependency relationship between them; A first determination module is configured to input the institution characteristic data into a pre-trained risk-bearing capacity assessment model to obtain a risk-bearing capacity index corresponding to each insurance institution, wherein the risk-bearing capacity index is positively correlated with the insurance institution's ability to bear risk; a setting module for setting differentiated thresholds for the multiple insurance institutions based on the risk-bearing capacity index to obtain a corresponding reinsurance excess loss threshold for each insurance institution; the reinsurance excess loss threshold is used to represent the maximum compensation resource amount of the insurance institution under a preset insurance compensation scenario; The second determining module is used to determine the compensation ratio of each insurance institution under the preset insurance compensation scenario according to the reinsurance excess loss threshold.

8. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the model-based data processing method according to any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the model-based data processing method according to any one of claims 1 to 6 is implemented.

10. A computer program product, characterized in that When the instructions in the computer program product are executed by an electronic device, the electronic device is caused to perform the model-based data processing method according to any one of claims 1 to 6.

Images