Intelligent underwriting management method and platform for special operation insurance supervision

Abstract

The application provides a kind of special operation insurance supervision-oriented intelligent underwriting management method and platform, belongs to underwriting management technical field, specifically includes: the risk hidden danger point identification result change of different underwriting analysis management target is determined using underwriting analysis management method, based on change, the risk identification process of the risk hidden danger point identification result consistent is divided into a group, according to the group data in different underwriting analysis management target and the new data of risk hidden danger point, determine the intervention management method of artificial underwriting of enterprise, to intervene management method for artificial underwriting intervention processing of different enterprises, according to the artificial underwriting intervention data of different enterprises, determine the underwriting analysis management method of enterprise except underwriting analysis management target, improve the reliability of underwriting processing.

Description

Technical Field

[0001] This invention belongs to the field of underwriting management technology, and in particular relates to an intelligent underwriting management method and platform for special operations insurance supervision. Background Technology

[0002] Special operations refer to operations that are prone to accidents causing personal injury or death, and may pose significant hazards to the safety of the operator, others, and surrounding facilities. Personnel engaged in special operations are required to hold a special operations certificate, and their employers are required to provide them with appropriate accident insurance and employer's liability insurance.

[0003] In existing technical solutions, the invention patent application CN202410008108.9, "A Method, Equipment and Medium for Underwriting Business Execution Based on Unified Interface Management," establishes a unified interface management platform to process all interfaces uniformly through configuration and other means. This enables accurate identification of the source of problems, reduces the degree of interface control chaos, and reduces excessive interface permission control when each service is provided externally, thus reducing repetitive development tasks. However, the following technical problems exist: When using models for underwriting special operations insurance, due to differences in enterprise data, there may be omissions and biases in the identification of various risk points in different enterprises. Therefore, it is an urgent technical problem to determine the underwriting management method for enterprises that have completed underwriting based on the risk of omissions in the identification of risk points in different enterprises, so as to improve the reliability of the model in identifying biases and omissions, and then carry out targeted updates.

[0004] Specifically, this application provides an intelligent underwriting management method and platform for special operations insurance supervision. Summary of the Invention

[0005] To achieve the objectives of this invention, the following technical solution is adopted: Specifically, this application provides an intelligent underwriting management method for special operations insurance supervision, which includes: S1 uses underwriting data of special operation insurance from different enterprises to identify risk hazard points in different types of enterprises. Based on the identification data of risk hazard points in various enterprises and the degree of similarity of the identification data, the underwriting analysis management objectives of the model are determined. S2 uses the updated results of the underwriting analysis and management objectives to determine the risk hazard point data for different underwriting analysis and management objectives, and uses the risk hazard point data for different underwriting analysis and management objectives to determine the underwriting analysis and management method for the underwriting analysis and management objectives; S3 uses the underwriting analysis and management method to determine the changes in the identification results of risk points for different underwriting analysis and management objectives. Based on the changes, risk identification processes with consistent risk point identification results are grouped into the same group. Based on the group data in different underwriting analysis and management objectives and the newly added data of risk points, the company's manual underwriting intervention management method is determined. S4 uses the aforementioned intervention management method to process manual underwriting intervention for different enterprises. Based on the manual underwriting intervention data of different enterprises, it determines the underwriting analysis management method for enterprises other than those with underwriting analysis management objectives.

[0006] The beneficial effects of this invention are as follows: By using the identification data of risk points in various enterprises, the potential identification risks of the model in each enterprise are determined. Based on the potential identification risks of the model in each enterprise, the underwriting analysis management objectives of the model are determined. This lays the foundation for determining the identification deviation risk of the model in the underwriting process based on the changes in the underwriting analysis results of the model in the underwriting analysis management objectives.

[0007] Based on the consistency between manual underwriting intervention data in different enterprises and the results of manual underwriting and the underwriting processing results of the model, the degree of deviation risk of the model's underwriting processing results is determined. Based on the degree of deviation risk of the model's underwriting processing results, the underwriting analysis and management methods of enterprises are determined. That is, how to conduct underwriting analysis and processing for enterprises that have completed underwriting, except for those with underwriting analysis and management objectives, thereby improving the efficiency of the model in accumulating and identifying deviation data, and thus improving the reliability of the model's update processing.

[0008] Furthermore, the underwriting data for the enterprise's special operations insurance is determined based on the number of special operations insurance policies underwritten during the underwriting process and the results of identifying potential risks during the underwriting process.

[0009] Furthermore, the types of enterprises are classified according to their products, specifically, enterprises with the same products are classified into the same enterprise type.

[0010] Furthermore, the risk hazard identification data includes the number of risk hazard points in different enterprises.

[0011] Furthermore, the method for determining the underwriting analysis management objectives of the model is as follows: Based on the identification data of risk points in various enterprises, enterprises with risk points are identified and used as matching enterprises for identification. Based on the degree of similarity of the identification data of risk points between different identification matching enterprises, the proportion of the number of identification matching enterprises in different risk points is determined, and it is used as the identification matching coefficient of the risk point. By using the identification and matching coefficients of the identified and matched enterprises and different risk points, the underwriting analysis and management objectives of the model are determined.

[0012] Furthermore, the method for determining the underwriting analysis management method of the enterprise is as follows: Based on the manual underwriting intervention data of different enterprises, identify the enterprises that have engaged in manual underwriting among the aforementioned enterprises; The degree of consistency between the manual underwriting results of enterprises with manual underwriting involvement and the underwriting processing results of the model is used to determine the number of deviations of risk points in different enterprises with manual underwriting involvement. Based on the number of enterprises with manual underwriting involvement and the number of deviations in risk points among different enterprises with manual underwriting involvement, the underwriting analysis and management method for enterprises other than the underwriting analysis and management objectives is determined.

[0013] Secondly, this application provides an intelligent underwriting management platform for special operations insurance supervision, employing the aforementioned intelligent underwriting management method for special operations insurance supervision, specifically including: Target screening module, intervention management module, underwriting analysis module; The target screening module is responsible for determining the underwriting analysis and management objectives of the model. The intervention management module is responsible for determining the intervention management method for the company's manual underwriting. The underwriting analysis module is responsible for determining the underwriting analysis management methods for enterprises other than the underwriting analysis management objectives.

[0014] Other features and advantages will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0015] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0016] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0017] Figure 1 This is a flowchart of an intelligent underwriting management method for special operations insurance supervision; Figure 2This is a flowchart illustrating the method for determining the underwriting analysis management objectives of the model; Figure 3 This is a framework diagram of an intelligent underwriting management platform for special operations insurance supervision. Detailed Implementation

[0018] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that the invention will be thorough and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art. The same reference numerals in the drawings denote the same or similar structures, and therefore their detailed description will be omitted.

[0019] The terms “a,” “one,” “the,” and “the” are used to indicate the existence of one or more elements, components, etc.; the terms “including” and “having” are used to indicate an open-ended meaning of inclusion and that other elements, components, etc. may exist in addition to the listed elements, components, etc.

[0020] Example 1 To solve the above problems, according to one aspect of the present invention, such as Figure 1 As shown, an intelligent underwriting management method for special operations insurance supervision is provided, specifically including: S1 uses underwriting data of special operation insurance from different enterprises to identify risk hazard points in different types of enterprises. Based on the identification data of risk hazard points in various enterprises and the degree of similarity of the identification data, the underwriting analysis management objectives of the model are determined. Furthermore, the underwriting data for the enterprise's special operations insurance is determined based on the number of special operations insurance policies underwritten during the underwriting process and the results of identifying potential risks during the underwriting process.

[0021] Specifically, the types of enterprises are classified according to their products, and enterprises with the same products are classified into the same type of enterprise.

[0022] It should be noted that the identified risk points data includes the number of risk points in different enterprises.

[0023] It is understandable that, such as Figure 2 As shown, the method for determining the underwriting analysis management objectives of the model is as follows: In this embodiment, the potential identification risks of the model in the risk points of each enterprise are determined by the identification data of the risk points of each enterprise. Based on the potential identification risks of the model in the risk points of each enterprise, the underwriting analysis management objectives of the model are determined. This lays the foundation for determining the identification deviation risk of the model in the underwriting process based on the changes in the underwriting analysis results of the model in the underwriting analysis management objectives.

[0024] It should be noted that the goal of the underwriting analysis is to perform repeated underwriting analysis on the materials submitted during the underwriting process to determine whether any omissions exist.

[0025] S11 uses the risk hazard identification data of each enterprise to identify enterprises with risk hazard points and uses them as the identification matching enterprises; The identified and matched enterprises refer to enterprises with potential risks, that is, enterprises whose risk risk data contains a number of potential risks greater than zero.

[0026] Assuming that multiple enterprises correspond to different risk hazard identification data, then determine whether the number of risk hazard points for each enterprise is greater than zero, and identify the enterprises with a number of risk hazard points greater than zero as the matching enterprises.

[0027] This step is used to identify matching companies. Its significance lies in screening out companies with potential risks. If companies with potential risks are not identified for a long time, there is a risk of omission in the identification process.

[0028] It is understandable that in the above steps, the proportion of the number of identified matching enterprises in the enterprise is used as the identification matching ratio. It is determined whether the identification matching ratio is greater than the preset matching ratio threshold. If it is, then proceed to step S12. If not, then the model has the risk of identification bias. Therefore, the underwriting analysis management target of the model is determined to be all identified matching enterprises. The identification matching ratio refers to the proportion of the number of identified and matched enterprises in the total number of enterprises; the preset matching ratio threshold is a reference value used to determine whether the number of identified and matched enterprises is too large.

[0029] Assuming multiple matching companies have been identified, the proportion of matching companies to the total number of companies is calculated and used as the matching ratio. This matching ratio is then compared with a preset matching ratio threshold. If the matching ratio is greater than the preset matching ratio threshold, the process proceeds to step S12 for further analysis. If the matching ratio is not greater than the preset matching ratio threshold, the underwriting analysis management target of the model is determined to be all matching companies.

[0030] This step determines the target by judging whether the matching ratio is greater than the preset matching ratio threshold. Its significance is that when the number of enterprises with potential risks is small, all identified and matched enterprises can be used as underwriting analysis management targets, thereby ensuring the comprehensiveness of underwriting analysis management and avoiding the problem of incomplete underwriting analysis results due to the omission of some enterprises.

[0031] S12 determines the proportion of identified matching enterprises in different risk points based on the similarity of the identification data of risk points between different identified matching enterprises, and uses it as the identification matching coefficient of the risk point. The above steps include the following: Using different risk point identification matching coefficients, determine whether there are risk points with identification matching coefficients less than a preset matching coefficient threshold. If yes, proceed to step S13. If no, determine that the probability of the model having identification bias risk is low. Therefore, determine that the underwriting analysis management target of the model is to identify matching enterprises whose number of risk points is above the preset risk point number threshold. The identification matching coefficient refers to an indicator used to reflect the coverage of risk points in the identified matching enterprises, and is determined by the proportion of the number of identified matching enterprises corresponding to the risk point; the preset matching coefficient threshold refers to a reference value used to judge whether the identification matching coefficient of the risk point is low; the preset risk point number threshold refers to a reference value used to judge whether the number of risk points of an enterprise is large.

[0032] Assuming that multiple matching enterprises correspond to different risk hazard points, for each risk hazard point, the number of matching enterprises identified for that risk hazard point is counted, and the proportion of this number to the total number of matching enterprises is calculated. This proportion is used as the matching coefficient for that risk hazard point. The matching coefficients of each risk hazard point are compared with a preset matching coefficient threshold. If the matching coefficient of any risk hazard point is less than the preset matching coefficient threshold, the process proceeds to step S13 for further analysis. If the matching coefficients of all risk hazard points are not less than the preset matching coefficient threshold, the underwriting analysis management target of the model is determined to be matching enterprises whose number of risk hazard points is above the preset threshold.

[0033] This step determines whether the model has identification bias risk by calculating the identification matching coefficient. The significance is that when the identification matching coefficient of all risk points is high, it means that the model's identification results in each identification matching enterprise are relatively consistent, and the probability of identification bias risk is low. At this time, a simpler target determination method can be adopted to improve processing efficiency.

[0034] S13 uses the identification matching coefficients of the identified matching enterprises and different risk points to determine the underwriting analysis management objectives of the model.

[0035] It should be noted that, using the identification and matching coefficients of the identified and matched enterprises and different risk points, the underwriting analysis and management objectives of the model are determined, specifically including: Risk points with a matching coefficient less than a preset matching coefficient threshold are identified as deviation risk points. The deviation risk coefficient is determined by the proportion of these deviation risk points among all risk points and the matching coefficients of different risk points. It is then determined whether the deviation risk coefficient meets the requirements. If it does, the deviation risk is low, and the underwriting analysis management target of the model is determined to be identified matching enterprises with a number of risk points above the preset threshold and identified matching enterprises with deviation risk points. If not, the underwriting analysis management target of the model is determined to be identified matching enterprises.

[0036] The deviation risk point refers to the risk point where the identification matching coefficient is less than the preset matching coefficient threshold; the identification deviation risk coefficient refers to the index used to comprehensively measure the degree of deviation risk identified by the model; the requirement to meet the requirement means that the identification deviation risk coefficient is not greater than the preset deviation risk coefficient threshold.

[0037] Assuming multiple deviation risk points have been identified, the proportion of each deviation risk point among all risk points is calculated, and the identification matching coefficient for each risk point is calculated separately. A comprehensive deviation risk coefficient is then determined, and this coefficient is compared with a preset deviation risk coefficient threshold. If the deviation risk coefficient is not greater than the preset threshold, the underwriting analysis management target of the model is determined to be identified matching enterprises with a number of risk points exceeding the preset threshold, as well as identified matching enterprises with deviation risk points. If the deviation risk coefficient is greater than the preset threshold, the underwriting analysis management target of the model is determined to be all identified matching enterprises.

[0038] This step determines the underwriting analysis management objective by judging whether the identification deviation risk coefficient meets the requirements. Its significance lies in the fact that by comprehensively considering the proportion of potential deviation risk points and the identification matching coefficient, the degree of identification deviation risk of the model can be more accurately assessed, thereby selecting a more appropriate method for determining the underwriting analysis management objective and avoiding the problem of insufficient reliability of underwriting analysis results due to improper objective determination.

[0039] Furthermore, the higher the proportion of the deviation risk points among all risk points, and the lower the identification matching coefficient of different risk points, the higher the identification deviation risk coefficient.

[0040] The formula for calculating the identification deviation risk coefficient is: Identification deviation risk coefficient = Percentage of the number of deviation risk points × (1 − Average identification matching coefficient), where the average identification matching coefficient is the average of the identification matching coefficients of all risk points.

[0041] If the proportion of potential risk points among all potential risk points is relatively high, and the identification matching coefficient of each potential risk point is relatively low, then the calculated identification deviation risk coefficient is high, indicating that the model has a large identification deviation risk.

[0042] This step clarifies the changing pattern of the identification deviation risk coefficient. Its significance lies in the fact that by quantifying the proportion of potential deviation risk points and the impact of the identification matching coefficient on the identification deviation risk coefficient, the identification deviation risk coefficient can be calculated more accurately, providing a scientific basis for determining the underwriting analysis and management objectives.

[0043] This embodiment achieves accurate determination of underwriting analysis management objectives by identifying potential risk points and determining the model's underwriting analysis management objectives. Its core value lies in three aspects: First, by identifying matching companies and determining the matching ratio, it achieves a differentiated approach to determining underwriting analysis management objectives, ensuring both efficiency and comprehensiveness. Second, by introducing identification matching coefficients and identification deviation risk coefficients to comprehensively evaluate the model's identification deviation risk level, it ensures more accurate and reliable determination of underwriting analysis management objectives. Third, by differentiating between different levels of identification deviation risk and employing different objective determination methods, it adapts to varying model identification reliability levels, improving the applicability and flexibility of the underwriting analysis management objective determination method.

[0044] The specific meanings of "Risk Hazard Point 1 to Risk Hazard Point 8" are clarified as the following eight types of potential model identification deviation spaces: whether the density of safety warning signs at the work site is sufficient; whether the actual skill level of special operation personnel matches the declared qualifications; the degree of correlation between historical similar work accident data and the current risks of the enterprise; the degree of superposition of safety risks when multiple types of work are cross-operated; the degree of impact of seasonal or temporary risk factors on the enterprise; the matching degree between the enterprise's safety investment and the effectiveness of actual risk control; the distinction and judgment of similar risk hazard points (such as the boundary between "high humidity at the work site" and "excessive humidity at the work site"); and dynamic risk prediction based on industry trends and the enterprise's operating conditions.

[0045] Complete and specific implementation of S1: Assume there are twenty companies in total, divided into four types (A, B, C, and D), with five companies in each type. Each company has different underwriting data during the special operations insurance underwriting process.

[0046] Of the twenty companies, fifteen have potential risks (the number of potential risks is greater than zero). These fifteen companies are identified as the matching companies. The total number of companies is 20, therefore the matching ratio = 15 ÷ 20 = 0.75.

[0047] The preset matching ratio threshold is 0.70. The matching ratio of 0.75 is found to be greater than the preset matching ratio threshold of 0.70, therefore proceeding to step S12.

[0048] Specific implementation of step S12: Assuming there are eight types of risk hazards identified in the fifteen matched companies (i.e., the revised eight types mentioned above), the coverage of each risk hazard in the identified matched companies is statistically analyzed: Risk Hazard Point 1 (Warning Sign Density): Identified among 12 matching enterprises, with a matching coefficient of 12 ÷ 15 = 0.80; the preset matching coefficient threshold is 0.60. Risk Hazard Points 5 (0.53) and 8 (0.47) have matching coefficients less than 0.60, therefore there are risk hazard points with matching coefficients less than the preset matching coefficient threshold, proceed to step S13.

[0049] There are two potential deviation risks: risk point 5 and risk point 8. There are a total of 8 potential risks, therefore the percentage of potential deviation risks among all potential risks is 2 ÷ 8 = 0.25.

[0050] The identification matching coefficients for each risk hazard point are 0.80, 0.67, 0.60, 0.93, 0.53, 0.87, 0.73, and 0.47, respectively. The average identification matching coefficient is (0.80 + 0.67 + 0.60 + 0.93 + 0.53 + 0.87 + 0.73 + 0.47) ÷ 8 = 5.60 ÷ 8 = 0.70.

[0051] The formula for calculating the identification deviation risk coefficient is: Identification deviation risk coefficient = Percentage of the number of potential deviation risk points × (1 − Average identification matching coefficient) = 0.25 × (1 − 0.70) = 0.25 × 0.30 = 0.075.

[0052] The preset deviation risk coefficient threshold (the threshold that meets the requirements) is 0.10. The identified deviation risk coefficient of 0.075 is not greater than the preset deviation risk coefficient threshold of 0.10, therefore the identified deviation risk is low. The underwriting analysis management objective of the model is determined to be: identifying matched enterprises with a number of risk points that are above the preset risk point number threshold (e.g., 3), and identifying matched enterprises with deviation risk points.

[0053] S2 uses the updated results of the underwriting analysis and management objectives to determine the risk hazard point data for different underwriting analysis and management objectives, and uses the risk hazard point data for different underwriting analysis and management objectives to determine the underwriting analysis and management method for the underwriting analysis and management objectives; It should be noted that, during the underwriting analysis process, all underwriting analysis targets are based on the underwriting materials at the time of underwriting and are subject to dynamic underwriting analysis.

[0054] Specifically, the method for determining the underwriting analysis management method for the underwriting analysis management objectives is as follows: In the above steps, the reliability of the identification reliability under different risk points is determined by the updated results of the underwriting analysis and management objectives and the degree of matching between the updated results and the underwriting data. Based on the reliability of the identification reliability under different risk points, the underwriting analysis and management method for the underwriting analysis and management objectives is determined. This reduces the difficulty of underwriting analysis and processing while also ensuring the reliability of the model's identification deviation under different risk points.

[0055] S21 uses risk hazard data for different underwriting analysis and management objectives to determine the number of risk hazard points in different underwriting analysis and management objectives, and determines the identification matching weight coefficient of the underwriting analysis and management objectives based on the number of risk hazard points; Specifically, the identification matching weight coefficient is determined based on the proportion of the number of risk points in the underwriting analysis and management target to the total number of risk points.

[0056] The identification and matching weight coefficient refers to the weight value used to reflect the coverage of risk points in the underwriting analysis and management objectives; the "all risk points" refers to the total set of all types of risk points that appear in all underwriting analysis and management objectives.

[0057] Assuming that multiple underwriting analysis and management objectives correspond to different risk hazard points, the number of risk hazard points in each underwriting analysis and management objective is counted, and then the proportion of the number of risk hazard points of each underwriting analysis and management objective in the total number of all risk hazard points is calculated. This proportion is used as the identification and matching weight coefficient of the underwriting analysis and management objective.

[0058] This step is used to determine the identification and matching weight coefficients for different underwriting analysis and management objectives. Its significance lies in the fact that by quantifying the coverage of risk points of each underwriting analysis and management objective, it can accurately reflect the importance of each underwriting analysis and management objective, and provide a scientific basis for the determination of subsequent underwriting analysis and management methods.

[0059] Specifically, the above steps include the following: Based on the identification and matching weight coefficients of different underwriting analysis management objectives, it is determined whether the average value of the identification and matching weight coefficients of different underwriting analysis management objectives is greater than the preset matching weight coefficient threshold. If not, the underwriting analysis management method of the underwriting analysis management objective is determined to be the underwriting analysis processing of the underwriting analysis management objective according to the basic analysis processing cycle (i.e., dynamic underwriting analysis processing of the same underwriting material to determine the degree of deviation of the underwriting analysis processing of the underwriting analysis model). When the identification result of the risk point in any underwriting analysis management objective is inconsistent with the identification result at the time of underwriting, the underwriting analysis processing of all verification analysis objectives is carried out at the target number of times in the future preset time period in the form of equal time intervals. If so, proceed to step S22. The preset matching weight coefficient threshold refers to a reference value used to determine whether the average value of the matching weight coefficient is too large; the basic analysis processing cycle refers to the time cycle for performing underwriting analysis processing at fixed time intervals; the verification analysis target refers to the underwriting analysis management target that needs to be verified; the equal time interval method refers to performing multiple underwriting analysis processes at equal time intervals within a preset future period; the target number refers to the total number of underwriting analysis processes performed within a preset future period.

[0060] Assuming that multiple underwriting analysis management objectives correspond to different identification matching weight coefficients, the average value of these identification matching weight coefficients is calculated, and the average value is compared with a preset matching weight coefficient threshold. If the average value is not greater than the preset matching weight coefficient threshold, the underwriting analysis management method is determined as follows: underwriting analysis is performed according to the basic analysis processing cycle, and if the identification result of the risk hazard point found in any underwriting analysis management objective is inconsistent with the identification result at the time of underwriting, then underwriting analysis is performed on all verification analysis objectives at equal time intervals within a preset period of time. If the average value is greater than the preset matching weight coefficient threshold, then proceed to step S22 for further processing.

[0061] This step determines the management method by judging whether the average value of the matching weight coefficient is greater than the preset matching weight coefficient threshold. Its significance is that when the average value of the matching weight coefficient is low, it indicates that the coverage of risk points of each underwriting analysis management target is low, and a simpler underwriting analysis processing method can be adopted. When inconsistencies are found in the identification results, enhanced analysis processing is triggered to ensure the reliability and timeliness of the underwriting analysis.

[0062] S22 determines the proportion of underwriting analysis and management targets in related enterprises at different risk points based on the degree of matching between the updated results and the underwriting data, and uses the proportion of underwriting analysis and management targets in related enterprises at the risk points as the underwriting matching ratio of the risk points. It should be noted that the associated companies of the aforementioned risk points are those companies that have identified the risk points.

[0063] The degree of matching between the updated results and the underwriting data refers to the degree of matching between the updated results of the underwriting analysis and management objectives (i.e., the latest determined underwriting analysis and management objectives) and the original underwriting data; the associated enterprises of the risk point refer to the enterprises that have been identified as risk points; the underwriting matching ratio refers to the proportion of enterprises belonging to the underwriting analysis and management objectives among the associated enterprises of a certain risk point.

[0064] Assuming that multiple risk points correspond to different related companies, for each risk point, the number of related companies that are the target of underwriting analysis is counted, and the proportion of this number in the total number of related companies is calculated. This proportion is used as the underwriting matching ratio for that risk point.

[0065] This step is used to determine the underwriting matching ratio for different risk points. Its significance lies in the fact that by quantifying the coverage of underwriting analysis and management objectives in the related enterprises of each risk point, it can accurately reflect the matching degree of underwriting analysis and management objectives for each risk point, and provide a data basis for determining the subsequent underwriting analysis and management methods.

[0066] It should be noted that the above steps include the following: Case 1: If there are no risk points where the underwriting matching ratio is less than the preset matching ratio threshold, then the underwriting analysis management method for the underwriting analysis management target is determined to be the underwriting analysis processing of the underwriting analysis management target according to the basic analysis processing cycle; The preset matching ratio threshold refers to a reference value used to determine whether the underwriting matching ratio of potential risk points is too low.

[0067] Assuming that multiple risk points correspond to different underwriting matching ratios, each underwriting matching ratio is compared with a preset matching ratio threshold. If the underwriting matching ratios of all risk points are not less than the preset matching ratio threshold, then the underwriting analysis management method is determined to be underwriting analysis processing according to the basic analysis processing cycle.

[0068] This situation involves using a basic analysis cycle for underwriting analysis when there are no risk points with low underwriting matching ratios. The significance of this is that when the underwriting matching ratios for all risk points are high, it indicates that the underwriting analysis management objectives have a high degree of coverage for each risk point, and a simpler underwriting analysis method can be used to improve processing efficiency.

[0069] Additionally, it should be noted that S22 also has a second scenario: if there are risk points where the underwriting matching ratio is less than the preset matching ratio threshold, it is determined whether the average underwriting matching ratio of different risk points is less than the preset matching ratio threshold. If so, the underwriting analysis management method for the underwriting analysis management target is determined to be to perform underwriting analysis processing of the underwriting analysis management target according to the basic analysis processing cycle. When the identification result of the risk point in any underwriting analysis management target is inconsistent with the identification result during underwriting, the target number of underwriting analysis processing is performed on all verification analysis targets in a way that takes equal time intervals within the future preset time period. If not, proceed to step S23. If the underwriting matching ratio for some potential risk points is less than the preset matching ratio threshold, the average underwriting matching ratio for all potential risk points is calculated and compared with the preset matching ratio threshold. If the average is less than the preset matching ratio threshold, the underwriting analysis management method is determined as follows: underwriting analysis is performed according to the basic analysis processing cycle. If the identification result of a potential risk point in any underwriting analysis management target is inconsistent with the identification result at the time of underwriting, underwriting analysis is performed on all verification analysis targets at equal time intervals within a preset period in the future. If the average is not less than the preset matching ratio threshold, the process proceeds to step S23 for further processing.

[0070] This approach employs enhanced underwriting analysis when there are risk points with low underwriting matching ratios and a low overall average. The significance of this approach lies in the fact that when the overall underwriting matching ratio is low, it indicates that the underwriting analysis management objectives have a low overall coverage of each risk point. Therefore, enhanced analysis should be triggered when inconsistencies in the identification results are found, thereby ensuring the reliability of the underwriting analysis.

[0071] S23 determines the underwriting analysis and management method for the underwriting analysis and management objectives based on the identification and matching weight coefficients of different underwriting analysis and management objectives and the underwriting matching ratios of different risk points.

[0072] It should be noted that the underwriting analysis and management method for determining the underwriting analysis and management objectives is based on the identification and matching weight coefficients of different underwriting analysis and management objectives and the underwriting matching ratios of different risk points. Specifically, this includes: The average matching ratio is determined by using the identification and matching weight coefficients of different underwriting analysis and management objectives and the average underwriting matching ratio of different risk points. It is then determined whether the average matching ratio meets the requirements. If it does, the identification and matching degree is high, and the underwriting analysis and management method for the objectives is determined to be to perform underwriting analysis and management according to the basic analysis and processing cycle. If the identification result of an underwriting analysis and management objective exceeding a preset number of objectives is inconsistent with the identification result at the time of underwriting, then all verification analysis objectives will undergo underwriting analysis and management at equal time intervals within a preset future period. Otherwise, the underwriting analysis and management method for the objectives is determined to be to perform underwriting analysis and management according to the basic analysis and processing cycle, and if the identification result of any underwriting analysis and management objective is inconsistent with the identification result at the time of underwriting, then all verification analysis objectives will undergo underwriting analysis and management at equal time intervals within a preset future period.

[0073] The average matching ratio refers to the average underwriting matching ratio of all risk points; meeting the requirement means that the average matching ratio is greater than the preset matching ratio threshold; the preset target number refers to a reference value used to judge whether the number of underwriting analysis management targets with inconsistent identification results is too large, and its value is greater than 1.

[0074] Assuming the underwriting matching ratio for each risk point has been determined, the average of these matching ratios is calculated as the average matching ratio. This average matching ratio is then compared with a preset matching ratio threshold. If the average matching ratio is greater than the preset threshold, the underwriting analysis management method is determined as follows: underwriting analysis is performed according to the basic analysis processing cycle. If the identification results for any underwriting analysis management target (e.g., 3 or more) are found to be inconsistent with the identification results at the time of underwriting, then all verification analysis targets will undergo underwriting analysis at equal time intervals within a preset future period. If the average matching ratio is not greater than the preset threshold, the underwriting analysis management method is determined as follows: underwriting analysis is performed according to the basic analysis processing cycle. If the identification results for any underwriting analysis management target are found to be inconsistent with the identification results at the time of underwriting, then all verification analysis targets will undergo underwriting analysis at equal time intervals within a preset future period.

[0075] This step determines the underwriting analysis management method by judging whether the average matching ratio meets the requirements. Its significance lies in the fact that by comprehensively considering the identification matching weight coefficient and the underwriting matching ratio, a suitable underwriting analysis processing method can be selected more accurately. This improves processing efficiency while ensuring the reliability of underwriting analysis and avoids the problem of poor processing results caused by using a single processing method.

[0076] Specifically, the variation in the identification results of the risk points of the underwriting analysis and management objectives is determined based on the degree of deviation of the identification results of the risk points of the underwriting analysis and management objectives in different identification processes.

[0077] The identification process refers to the process by which the model identifies potential risks and hazards in the underwriting analysis and management objectives; the degree of deviation refers to the degree of difference between the potential risks and hazards identified in different identification processes.

[0078] If a certain underwriting analysis and management objective is identified through multiple identification processes, the risk points identified in each identification process are compared, the number of differences in risk points between different identification processes is calculated, and the variation in the identification results is determined based on the number of differences.

[0079] This step clarifies the method for determining the changes in the identification results of risk points. Its significance lies in the fact that by quantifying the degree of deviation between different identification processes, it can accurately reflect the model's identification stability in the underwriting analysis and management objectives, providing a data foundation for subsequent manual underwriting intervention management.

[0080] This embodiment proposes an underwriting analysis and management method that determines the underwriting analysis and management objectives by comprehensively identifying matching weight coefficients and underwriting matching ratios. This method achieves scientific determination of the underwriting analysis and management objectives, with its core value manifested in three aspects: First, by determining whether the average value of the identified matching weight coefficients exceeds a preset matching weight coefficient threshold, the management method is differentiated. This allows for a simplified processing method when the average value of the identified matching weight coefficients is low, and triggers enhanced processing when inconsistencies are detected, ensuring the reliability and timeliness of the underwriting analysis. Second, by introducing the underwriting matching ratio to comprehensively assess the coverage of risk points, the determination results of the underwriting analysis and management method are more accurate and comprehensive. Third, by differentiating between different average matching ratio values ​​and applying different enhanced processing trigger conditions (inconsistencies exceeding a preset target number vs. any single inconsistency), the method can adapt to different matching levels, improving its applicability and flexibility.

[0081] S3 uses the underwriting analysis and management method to determine the changes in the identification results of risk points for different underwriting analysis and management objectives. Based on the changes, risk identification processes with consistent risk point identification results are grouped into the same group. Based on the group data in different underwriting analysis and management objectives and the newly added data of risk points, the company's manual underwriting intervention management method is determined. Specifically, the variation in the identification results of the risk points of the underwriting analysis and management objectives is determined based on the degree of deviation of the identification results of the risk points of the underwriting analysis and management objectives in different identification processes.

[0082] It should be noted that the method for determining the intervention management method for manual underwriting of the aforementioned enterprise is as follows: In this embodiment, based on the changes in the identification results of risk points under different underwriting analysis and management objectives and the changes in the identification results of risk points under the changing analysis objectives, the degree of abnormality of the model's identification results of risk points under different underwriting analysis and management objectives is determined. Based on the degree of abnormality of the model's identification results of risk points under different analysis and management objectives, the intervention management method for manual underwriting of enterprises is determined, that is, how to conduct manual underwriting for newly added underwriting enterprises, thereby laying the foundation for further improving the reliability of enterprises' underwriting processing when applying for insurance.

[0083] S31 uses the changes in the identification results of risk points for different underwriting analysis and management objectives to determine the underwriting analysis and management objectives with changes in risk points, and uses them as change analysis objectives; It is understandable that in the above steps, the proportion of the change analysis target in the underwriting analysis management target is used as the change analysis target proportion. It is determined whether the change analysis target proportion is greater than the preset change target proportion threshold. If so, the manual underwriting intervention management method for the enterprise is determined to be that as long as the number of enterprises with risk points in the underwriting analysis process in the most recent preset time period (i.e., the number of newly added underwriting analysis processes) is less than the preset enterprise number threshold, manual underwriting intervention is required in the future preset time period. That is, underwriting analysis is carried out simultaneously using manual and model methods. If not, proceed to step S32. The change analysis target refers to the underwriting analysis management target with changes in potential risk points; the preset change target ratio threshold is a reference value used to determine whether the number of change analysis targets is too large; the preset duration is a time window used to determine whether there are potential risk points in the recent underwriting analysis process; the preset enterprise number threshold is a reference value used to determine whether the number of enterprises with potential risk points is too small.

[0084] Assuming multiple underwriting analysis management objectives correspond to changes in the identification results of different risk points, the process involves determining whether each underwriting analysis management objective has changed risk points. Objectives with changes are identified as change analysis objectives. The proportion of change analysis objectives within the total underwriting analysis management objectives is calculated and used as the change analysis objective proportion. This proportion is then compared to a preset change objective proportion threshold. If the change analysis objective proportion is greater than the preset threshold, the manual underwriting intervention method is determined as follows: if the number of enterprises with risk points in the recent preset time period (i.e., the number of newly added underwriting analysis processes) is less than a preset enterprise number threshold, then manual underwriting intervention is required for the next preset time period. If the change analysis objective proportion is not greater than the preset change objective proportion threshold, the process proceeds to step S32 for further processing.

[0085] This step determines the intervention management method by judging whether the proportion of change analysis targets is greater than the preset change target proportion threshold. Its significance is that when there are a large number of underwriting analysis management targets with changes in risk points, it indicates that the model's identification results fluctuate greatly, and a more stringent manual underwriting intervention management method is required to manually intervene in the newly added underwriting companies, thereby ensuring the reliability of the underwriting process.

[0086] S32 Based on the changes in the identification results of the risk hazard points of the change analysis target, the risk identification processes with consistent identification results of the risk hazard points are grouped into the same group; The risk identification process refers to the process by which the model identifies potential risks to the underwriting analysis and management objectives; the group refers to a collection of risk identification processes with consistent results in identifying potential risks.

[0087] Assuming that multiple change analysis targets correspond to multiple risk identification processes, the risk hazard points identified in each risk identification process are compared, and risk identification processes with consistent identification results are grouped into the same group, thereby achieving classified management of risk identification processes.

[0088] This step is used to group the risk identification process according to the consistency of the identification results. Its significance lies in the fact that by dividing the results into groups, stable and abnormal patterns of the model identification results can be identified, providing a data basis for determining the subsequent manual underwriting intervention management methods and avoiding the problem of inappropriate intervention management methods due to failure to consider the stability of the identification results.

[0089] S33 determines the intervention management method for manual underwriting of the enterprise based on the change analysis target, the number of groups of different change analysis targets, and the number of newly added risk points.

[0090] Specifically, the number of newly added risk points is determined based on the difference between the identification results of the risk points and the risk points at the time of the initial underwriting process.

[0091] The newly added number of risk points refers to the difference between the number of risk points in the current identification results and the number of risk points when the underwriting process was first carried out; the risk points when the underwriting process was first carried out refer to the risk points identified when the underwriting analysis and management objectives were first processed.

[0092] If a certain underwriting analysis management objective identifies 3 risk points during the initial underwriting process, and 5 risk points are identified in the current identification results, then the number of newly identified risk points = 5 − 3 = 2.

[0093] This step clarifies the calculation method for the number of newly added risk points. Its significance lies in the fact that by quantifying the number of newly added risk points, it can accurately reflect the changing trend of the model identification results, and provide a scientific basis for determining the subsequent manual underwriting intervention management method.

[0094] It is understood that, based on the aforementioned change analysis objectives, the number of groups of different change analysis objectives, and the number of newly added risk points, the intervention management method for manual underwriting of the enterprise is determined, specifically including: S331 uses the number of newly added risk points as a basis, takes the change analysis target with a number of newly added points greater than the preset number threshold as the risk target, and judges whether the proportion of the risk target in the change analysis target meets the requirements. If not, proceed to step S332. If yes, determine that the manual underwriting intervention management method for the enterprise is as follows: as long as the number of enterprises with risk points in the recent preset time period underwriting analysis process (i.e., the number of newly added underwriting analysis processing enterprises) is less than the preset enterprise number threshold, and the proportion of enterprises with risk points (i.e., the number of newly added underwriting analysis processing enterprises) is less than the preset risk enterprise proportion threshold, then manual underwriting intervention is required in the future preset time period. That is, underwriting analysis is performed simultaneously by manual and model. The preset new quantity threshold is a reference value used to determine whether the number of new risk points is too large; the risk target is a change analysis target where the number of new risk points is greater than the preset new quantity threshold; the requirement is met when the proportion of risk targets in the change analysis targets is greater than the preset risk target proportion threshold; the preset risk enterprise proportion threshold is a reference value used to determine whether the proportion of enterprises with risk points is too low.

[0095] Assuming that multiple change analysis targets correspond to different new risk points, change analysis targets with new numbers exceeding a preset threshold are identified as risk targets. The proportion of risk targets among change analysis targets is calculated and compared with a preset risk target proportion threshold. If the proportion exceeds the preset risk target proportion threshold, the manual underwriting intervention method is determined as follows: if the number of enterprises with risk points in the recent preset time period underwriting analysis (i.e., the number of newly added enterprises underwriting analysis) is less than the preset enterprise number threshold, and the proportion of enterprises with risk points (i.e., the number of newly added enterprises underwriting analysis) is less than the preset risk enterprise proportion threshold (i.e., the reliability of risk point identification and processing among newly added enterprises used for underwriting analysis is poor), then manual underwriting intervention is required in the future preset time period. If the proportion is not greater than the preset risk target proportion threshold, proceed to step S332 for further judgment.

[0096] This step determines the intervention management method by judging whether the proportion of risk targets in the change analysis targets meets the requirements. Its significance is that when the proportion of risk targets is high, it indicates that the model's identification results show a relatively common situation of newly added risk points. It is necessary to adopt a more stringent manual underwriting intervention management method and add the condition of the proportion of enterprises with risk points to ensure the reliability of underwriting processing.

[0097] S332 determines the basic risk weight coefficient of the change analysis target based on the sum of the number of groups of different change analysis targets and the number of newly added risk points. It then determines whether the average value of the basic risk weight coefficient of different change analysis targets is greater than the preset risk weight coefficient threshold. If so, it determines that the manual underwriting intervention management method for the enterprise is as follows: if the number of enterprises with risk points in the recent preset time period underwriting analysis process (i.e., the number of newly added underwriting analysis processes) is less than the preset enterprise number threshold, then manual underwriting intervention is required in the future preset time period. That is, underwriting analysis is performed simultaneously using manual and model methods. If not, proceed to step S333. The number of groups refers to the total number of groups obtained from the risk identification process of the change analysis target; the basic risk weight coefficient refers to an indicator used to reflect the degree of fluctuation of the identification results of the change analysis target, which is determined by the sum of the number of groups of the change analysis target and the number of newly added risk points; the preset risk weight coefficient threshold refers to a reference value used to judge whether the average value of the basic risk weight coefficient is too large.

[0098] Assuming that multiple change analysis targets correspond to different numbers of groups and different numbers of new risk points, the basic risk weight coefficient (the sum of the number of groups and the number of new cases) is calculated for each change analysis target. Then, the average of these basic risk weight coefficients is calculated and compared with a preset risk weight coefficient threshold. If the average is greater than the preset risk weight coefficient threshold, the manual underwriting intervention method is determined as follows: if the number of enterprises with risk points in the recent preset time period underwriting analysis (i.e., the number of newly added enterprises underwriting analysis) is less than the preset enterprise number threshold, then manual underwriting intervention is required for all newly added enterprises underwriting analysis. If the average is not greater than the preset risk weight coefficient threshold, proceed to step S333 for further judgment.

[0099] This step determines the intervention management method by judging whether the average value of the basic risk weight coefficient is greater than the preset risk weight coefficient threshold. The significance is that when the average value of the basic risk weight coefficient is high, it indicates that the identification result of the change analysis target fluctuates greatly, and a manual underwriting intervention management method is required to ensure the reliability of the underwriting process; when the average value is low, it indicates that the identification result fluctuates less, and a more lenient intervention management method can be used.

[0100] S333 determines the underwriting risk value based on the target proportion of the change analysis and the basic risk weight coefficients of different change analysis targets. It then determines whether the underwriting risk value is greater than a preset risk threshold. If so, the manual underwriting intervention management method for the enterprise is determined to be that if the number of enterprises with potential risk points in the recent preset underwriting analysis process (i.e., the number of newly added underwriting analysis processes) is less than a preset enterprise number threshold, then manual underwriting intervention is required for the next preset period. That is, underwriting analysis is performed simultaneously using both manual and model methods. If not, the manual underwriting intervention management method for the enterprise is determined to be that as long as the number of enterprises with potential risk points in the recent preset period is less than a preset enterprise number threshold, then manual underwriting intervention is required for the next preset period. If the number of enterprises with potential risks during the recent pre-set time period underwriting analysis (i.e., the number of newly added enterprises underwriting analysis) is less than the second pre-set enterprise number threshold (less than the pre-set enterprise number threshold), or if the number of enterprises with potential risks during the recent pre-set time period underwriting analysis (i.e., the number of newly added enterprises underwriting analysis) is less than the pre-set enterprise number threshold, and the proportion of enterprises with potential risks (i.e., the number of newly added enterprises underwriting analysis) is less than the pre-set threshold for the proportion of enterprises with potential risks, then manual underwriting intervention is required for the next pre-set time period. In other words, underwriting analysis will be conducted using both manual and model-based methods.

[0101] The underwriting risk value refers to an indicator used to comprehensively measure the degree of abnormality in the identification results of the underwriting analysis and management objectives; the preset risk threshold refers to a reference value used to determine whether the underwriting risk value is too large; the second preset enterprise number threshold refers to a reference value that is less than the preset enterprise number threshold, which is a stricter standard used to determine whether the number of enterprises with potential risk points is too small.

[0102] Assuming the target proportion for change analysis and the basic risk weight coefficients for each change analysis target have been determined, the underwriting risk value is determined by combining the target proportion for change analysis and the basic risk weight coefficients. This underwriting risk value is then compared with a preset risk threshold. If the underwriting risk value is greater than the preset risk threshold, the manual underwriting intervention method is determined as follows: if the number of enterprises with potential risk points in the recent preset time period is less than a preset enterprise number threshold, then manual underwriting intervention is required for the future preset time period. If the underwriting risk value is not greater than the preset risk threshold, the manual underwriting intervention method is determined as follows: if the number of enterprises with potential risk points in the recent preset time period is less than a second preset enterprise number threshold, or if the number of enterprises with potential risk points in the recent preset time period is less than a preset enterprise number threshold and the proportion of enterprises with potential risk points is less than a preset threshold for the proportion of enterprises with potential risk points, then manual underwriting intervention is required for the future preset time period.

[0103] This step determines the intervention management method for manual underwriting by judging whether the underwriting risk value is greater than the preset risk threshold. Its significance lies in the fact that by comprehensively considering the target ratio of change analysis and the basic risk weight coefficient, it can more accurately assess the degree of abnormality in underwriting analysis, thereby selecting a more appropriate manual underwriting intervention management method and avoiding problems such as insufficient reliability of underwriting processing or excessive labor costs caused by inappropriate intervention management method selection.

[0104] This embodiment determines the manual underwriting intervention management method for enterprises by comprehensively considering the change analysis targets, the number of groups, and the number of newly added risk points. This achieves a scientific determination of manual underwriting intervention management, with core value reflected in four aspects: First, by judging whether the proportion of change analysis targets exceeds a preset change target proportion threshold, the intervention management method is determined and differentiated. This allows for the use of a more stringent manual underwriting intervention management method when the number of change analysis targets is large, ensuring the reliability of underwriting processing. Second, by introducing risk targets and risk target proportion judgments, it can identify situations where the model identification results show a general increase in risk points and adopt corresponding manual underwriting intervention management methods. Third, by introducing basic risk weight coefficients and underwriting risk values ​​to comprehensively evaluate the volatility and anomaly of the identification results, it ensures that the determination results of the intervention management method are more accurate and comprehensive. Fourth, by differentiating different risk levels and using different manual underwriting intervention trigger conditions, it can adapt to different anomaly situations in the identification results, improving the applicability and flexibility of the manual underwriting intervention management method.

[0105] S4 uses the aforementioned intervention management method to process manual underwriting intervention for different enterprises. Based on the manual underwriting intervention data of different enterprises, it determines the underwriting analysis management method for enterprises other than those with underwriting analysis management objectives.

[0106] It is understood that the method for determining the underwriting analysis management method of the aforementioned enterprise is as follows: In this embodiment, based on the consistency between the manual underwriting intervention data of different enterprises and the manual underwriting results and the underwriting processing results of the model, the degree of deviation risk of the model's underwriting processing results is determined. Based on the degree of deviation risk of the model's underwriting processing results, the underwriting analysis and management method of the enterprise is determined, that is, how to conduct underwriting analysis and processing for enterprises that have completed underwriting except for those with underwriting analysis and management objectives. This improves the efficiency of the model in accumulating and identifying deviation data, and thus improves the reliability of the model's update processing.

[0107] It should be noted that the underwriting analysis is performed using the same materials as during the underwriting process.

[0108] During underwriting analysis, the underwriting materials are input into the large model, and risk points are determined based on the identification results of the large model.

[0109] S41 determines the companies involved in manual underwriting based on the manual underwriting intervention data of different companies; It is understandable that the number of companies with manual underwriting involvement in the enterprise is obtained, and it is determined whether the proportion of companies with manual underwriting involvement in the enterprise is less than the threshold for the proportion of companies with manual underwriting involvement. If so, the underwriting analysis management method of the enterprise is determined, that is, the reliability of the underwriting process is high at this time, and the enterprises other than the underwriting analysis management target do not need to undergo repeated underwriting analysis processing. If not, proceed to step S42. The manual underwriting intervention data refers to the processing data of enterprises undergoing manual underwriting intervention during the underwriting process; the manually intervened enterprises refer to enterprises that have undergone manual underwriting intervention; the intervention enterprise ratio threshold is a reference value used to determine whether the number of manually intervened enterprises is too small.

[0110] Assuming multiple enterprises correspond to different manual underwriting intervention data, the process involves identifying whether manual underwriting intervention has been performed from each enterprise's manual underwriting intervention data. Enterprises that have undergone manual underwriting intervention are designated as manually underwritten enterprises. The proportion of manually underwritten enterprises in the total number of enterprises is calculated and compared with an intervention enterprise proportion threshold. If the proportion is less than the intervention enterprise proportion threshold, the underwriting reliability is relatively high, and the underwriting analysis management method for the enterprise is determined as follows: enterprises other than those with underwriting analysis management objectives do not need to undergo repeated underwriting analysis processing. If the proportion is not less than the intervention enterprise proportion threshold, the process proceeds to step S42 for further processing.

[0111] This step determines the management method by judging whether the proportion of enterprises requiring manual underwriting is less than the threshold for the proportion of enterprises requiring manual underwriting. Its significance is that when the number of enterprises requiring manual underwriting is small, it indicates that the underwriting processing results of the model are relatively reliable, and a simpler underwriting analysis and management method can be used (without having to repeat the analysis of enterprises that have completed underwriting), thereby improving processing efficiency. At the same time, the reliability of the model is verified by observing whether new situations requiring manual intervention arise.

[0112] S42 determines the number of deviations in risk points among different companies with manual underwriting involvement by comparing the manual underwriting results with the model's underwriting processing results. Specifically, the deviation quantity refers to the number of risk points in the manual underwriting results of enterprises with manual underwriting intervention and the number of risk points in the underwriting processing results of the model.

[0113] The degree of consistency refers to the degree of matching between the manual underwriting results and the model's underwriting processing results; the number of deviations refers to the number of differences between the risk points in the manual underwriting results and the risk points in the model's underwriting processing results.

[0114] Assuming that a certain enterprise involved in manual underwriting has both the manual underwriting result and the model underwriting result for this underwriting process, the risk points in the two results are extracted separately, the risk points in the two results are compared, and the number of risk points with differences is counted. This number is taken as the deviation of risk points in the enterprise involved in manual underwriting.

[0115] This step is used to determine the number of deviations in risk points in different enterprises with manual underwriting intervention. Its significance lies in the fact that by quantifying the difference between the manual underwriting results and the model's underwriting processing results, the degree of deviation of the model's underwriting processing results can be accurately assessed, providing a data basis for determining subsequent underwriting analysis and management methods.

[0116] S43 determines the underwriting analysis management method for enterprises other than the underwriting analysis management objectives based on the number of enterprises with manual underwriting involvement and the number of deviations in risk points among different enterprises with manual underwriting involvement.

[0117] It should be noted that the enterprises excluding those subject to underwriting analysis and management targets are those that have already completed underwriting and are excluding those subject to underwriting analysis and management targets.

[0118] Specifically, based on the number of enterprises with manual underwriting involvement and the number of deviations in risk points among different enterprises with manual underwriting involvement, the underwriting analysis and management methods for enterprises other than the underwriting analysis and management objectives are determined, including: S431 determines the underwriting risk coefficient of the manually underwriting enterprise based on the number of deviations in the risk points of the manually underwriting enterprise. Based on the underwriting risk coefficients of different manually underwriting enterprises and the proportion of manually underwriting enterprises whose underwriting risk coefficients are greater than the preset underwriting risk coefficient threshold, a risk weight value is determined. It is then determined whether the risk weight value is greater than the preset risk weight threshold. If so, the underwriting analysis management method for enterprises other than the underwriting analysis management target is determined to be underwriting analysis processing according to the second basic analysis processing cycle (greater than the basic analysis processing cycle). If not, proceed to step S432. The underwriting risk coefficient refers to an indicator used to reflect the degree of deviation in the underwriting processing results of enterprises with manual underwriting intervention, and is determined by the number of deviations in the risk points of the enterprises with manual underwriting intervention; the preset underwriting risk coefficient threshold refers to a reference value used to judge whether the underwriting risk coefficient of the enterprises with manual underwriting intervention is large; the risk weight value refers to an indicator used to comprehensively measure the degree of deviation risk in the underwriting processing results of the model; the preset risk weight threshold refers to a reference value used to judge whether the risk weight value is large; the second basic analysis processing cycle refers to an underwriting analysis processing cycle that is longer than the basic analysis processing cycle, that is, the frequency of underwriting analysis processing is high.

[0119] Assuming that multiple manually underwritten enterprises correspond to different deviations in risk points, the underwriting risk coefficient for each manually underwritten enterprise is calculated. Then, the proportion of manually underwritten enterprises whose underwriting risk coefficient is greater than the preset underwriting risk coefficient threshold is counted. The average of the underwriting risk coefficient and this proportion is used to determine the risk weight value. The risk weight value is then compared with the preset risk weight threshold. If the risk weight value is greater than the preset risk weight threshold, the underwriting analysis processing method for enterprises other than those with underwriting analysis management objectives is determined to be underwriting analysis processing according to the second basic analysis processing cycle (e.g., every 10 days, the basic analysis processing cycle is once every 5 days). If the risk weight value is not greater than the preset risk weight threshold, then proceed to step S432 for further judgment.

[0120] This step determines the management method by judging whether the risk weight value is greater than the preset risk weight threshold. The significance is that when the risk weight value is large, it indicates that there is a large risk of deviation in the underwriting process of the model. It is necessary to repeat the underwriting analysis process for other companies that have completed the underwriting at a shorter cycle (the second basic analysis and processing cycle) so as to discover the identification deviation of the model in time and improve the reliability of the underwriting process.

[0121] S432 determines whether there are any enterprises with manual underwriting intervention whose underwriting risk coefficient is greater than the preset underwriting risk coefficient threshold. If so, proceed to step S433. If not, enterprises other than those subject to underwriting analysis management do not need to undergo repeated underwriting analysis processing, thereby determining whether the model is reliable. Assuming that the underwriting risk coefficients for each enterprise requiring manual underwriting intervention have been calculated, each underwriting risk coefficient is compared with a preset underwriting risk coefficient threshold to determine whether there is any enterprise with an underwriting risk coefficient greater than the preset underwriting risk coefficient threshold. If so, proceed to step S433 for further processing. If not, determine that enterprises other than those subject to underwriting analysis management objectives do not require repeated underwriting analysis processing, and determine the reliability of the model by observing whether new situations requiring manual intervention arise.

[0122] This step determines the management method by identifying whether there are any manually underwritten companies with underwriting risk coefficients greater than the preset underwriting risk coefficient threshold. The significance of this step is that when there are no manually underwritten companies with high underwriting risk coefficients, it indicates that the deviation of the model's underwriting processing results is small, and a simpler underwriting analysis and management method can be adopted to improve processing efficiency. At the same time, the reliability of the model can be verified through continuous observation.

[0123] S433 determines whether the proportion of enterprises with manual underwriting involvement in the enterprise is greater than the preset threshold for the proportion of involved enterprises (greater than the threshold for the proportion of involved enterprises). If so, in order to provide more training data for the model on the one hand, and to determine the true recognition reliability of the model on the other hand, thereby reducing the number of enterprises with manual underwriting involvement, the underwriting analysis management method for enterprises other than the underwriting analysis management target is determined to be the underwriting analysis processing of the underwriting analysis management target according to the second basic analysis processing cycle (greater than the basic analysis processing cycle). If not, the underwriting analysis management method for enterprises other than the underwriting analysis management target is determined to be the underwriting analysis processing of the enterprise if the enterprise has not undergone manual underwriting, according to the second basic analysis processing cycle (greater than the basic analysis processing cycle).

[0124] The preset threshold for the proportion of companies involved in the underwriting process (greater than the threshold for the proportion of companies involved in the underwriting process) refers to a stricter standard used to determine whether the proportion of companies involved in the manual underwriting process is too high.

[0125] Assuming the proportion of enterprises requiring manual underwriting has been determined, this proportion is compared with a preset threshold for the proportion of enterprises requiring manual underwriting (e.g., 0.40, which is greater than the threshold of 0.20). If the proportion is greater than 0.40, the underwriting analysis management method for enterprises other than those targeted by the underwriting analysis management is determined to be underwriting analysis processing according to the second basic analysis processing cycle. This provides more training data for the model and determines the model's actual recognition reliability, reducing the number of enterprises requiring manual underwriting. If the proportion is not greater than 0.40, the underwriting analysis management method for enterprises other than those targeted by the underwriting analysis management is determined to be: if the enterprise has not undergone manual underwriting, then underwriting analysis processing is performed according to the second basic analysis processing cycle.

[0126] This step determines the management method by judging whether the proportion of companies involved in manual underwriting exceeds a stricter threshold. The significance is that when the proportion of companies involved in manual underwriting is high, it indicates that there are significant issues with the reliability of the model, and all relevant companies need to be underwritten and analyzed on a shorter cycle. This provides more training data for the model and allows for an accurate assessment of the model's recognition reliability. When the proportion is at a moderate level, underwriting and analysis can be performed only on companies that have not undergone manual underwriting, thereby improving processing efficiency while ensuring reliability.

[0127] This embodiment determines the underwriting analysis and management method for enterprises other than those targeted by the underwriting analysis and management objective by comprehensively considering data from manual underwriting intervention, consistency, and the number of deviations. This achieves a scientific determination of the underwriting analysis and management method for other enterprises. Its core value lies in four aspects: First, by determining whether the proportion of enterprises with manual underwriting intervention is less than a threshold, the management method is differentiated, allowing for simplified management methods to be used when manual intervention is minimal, while also verifying model reliability and improving processing efficiency. Second, by introducing underwriting risk coefficients and risk weight values ​​to comprehensively assess the degree of deviation risk in the model's underwriting processing results, the determination of the underwriting analysis and management method is ensured to be more accurate and comprehensive. Third, by differentiating between different risk weight values ​​and enterprise proportions and applying different management methods, differentiating between models with varying reliability levels, the applicability and flexibility of the underwriting analysis and management method are improved. Fourth, by appropriately repeating the underwriting analysis for other enterprises that have already completed underwriting, model identification deviations can be detected in a timely manner, improving the efficiency of accumulating identification deviation data and thus enhancing the reliability of model updates.

[0128] Example 2 Secondly, such as Figure 3 As shown, this application provides an intelligent underwriting management platform for special operations insurance supervision, employing the aforementioned intelligent underwriting management method for special operations insurance supervision, specifically including: Target screening module, intervention management module, underwriting analysis module; The target screening module is responsible for determining the underwriting analysis and management objectives of the model. The intervention management module is responsible for determining the intervention management method for the company's manual underwriting. The underwriting analysis module is responsible for determining the underwriting analysis management methods for enterprises other than the underwriting analysis management objectives.

[0129] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0130] The foregoing has described specific embodiments of this application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired results. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0131] The above description is merely one or more embodiments of this application and is not intended to limit this application. For those skilled in the art, various modifications and variations can be made to one or more embodiments of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this application should be included within the scope of the claims of this application.

Claims

1. An intelligent underwriting management method for special operations insurance supervision, characterized in that, Specifically, it includes: Using underwriting data of special operation insurance from different enterprises, identify risk hazard points in different types of enterprises. Based on the identification data of risk hazard points in various enterprises and the degree of similarity of the identification data, determine the underwriting analysis management objectives of the model. Based on the updated results of the underwriting analysis and management objectives, risk hazard point data for different underwriting analysis and management objectives are determined, and the underwriting analysis and management method for the underwriting analysis and management objectives is determined based on the risk hazard point data for different underwriting analysis and management objectives. The underwriting analysis and management method is used to determine the changes in the identification results of risk points for different underwriting analysis and management objectives. Based on the changes, risk identification processes with consistent risk point identification results are grouped into the same group. Based on the group data in different underwriting analysis and management objectives and the newly added data of risk points, the intervention management method for manual underwriting of the enterprise is determined. The aforementioned intervention management method is used to process manual underwriting intervention for different enterprises. Based on the manual underwriting intervention data of different enterprises, the underwriting analysis management method for enterprises other than the underwriting analysis management objectives is determined.

2. The intelligent underwriting management method for special operations insurance supervision as described in claim 1, characterized in that, The underwriting data for the company's special operations insurance is determined based on the number of special operations insurance policies underwritten during the underwriting process and the results of identifying potential risks during the underwriting process.

3. The intelligent underwriting management method for special operations insurance supervision as described in claim 1, characterized in that, The types of enterprises are classified according to their products, specifically, enterprises with the same products are classified into the same enterprise type.

4. The intelligent underwriting management method for special operations insurance supervision as described in claim 1, characterized in that, The data for identifying potential risks includes the number of potential risks in different enterprises.

5. The intelligent underwriting management method for special operations insurance supervision as described in claim 1, characterized in that, The method for determining the underwriting analysis management objectives of the model is as follows: Based on the identification data of risk points in various enterprises, enterprises with potential risks are identified and used as matching enterprises for identification. Based on the degree of similarity of the identification data of risk points between different identification matching enterprises, the proportion of the number of identification matching enterprises in different risk points is determined, and it is used as the identification matching coefficient of the risk point. By using the identification and matching coefficients of the identified and matched enterprises and different risk points, the verification analysis management objectives of the model are determined.

6. The intelligent underwriting management method for special operations insurance supervision as described in claim 5, characterized in that, Using the identification and matching coefficients of the identified and matched enterprises and different risk points, the verification analysis management objectives of the model are determined, specifically including: Risk points with a matching coefficient less than a preset matching coefficient threshold are identified as deviation risk points. The deviation risk coefficient is determined by the proportion of these deviation risk points among all risk points and the matching coefficients of different risk points. It is then determined whether the deviation risk coefficient meets the requirements. If it does, the underwriting analysis and management target of the model is identified matching enterprises with a number of risk points greater than the preset threshold and identified matching enterprises without deviation risk points. If not, the underwriting analysis and management target of the model is identified matching enterprises.

7. The intelligent underwriting management method for special operations insurance supervision as described in claim 1, characterized in that, The method for determining the underwriting analysis management method for the underwriting analysis management objectives is as follows: Using risk hazard data for different underwriting analysis and management objectives, determine the number of risk hazard points in different underwriting analysis and management objectives, and determine the identification and matching weight coefficient of the underwriting analysis and management objectives based on the number of risk hazard points. Based on the degree of matching between the updated results and the underwriting data, the proportion of underwriting analysis and management targets in related enterprises at different risk points is determined, and the proportion of underwriting analysis and management targets in related enterprises at the risk points is used as the underwriting matching proportion for the risk points. Based on the identification and matching weight coefficients of different underwriting analysis and management objectives and the underwriting matching ratios of different risk points, the underwriting analysis and management method for the stated underwriting analysis and management objectives is determined.

8. The intelligent underwriting management method for special operations insurance supervision as described in claim 7, characterized in that, The identification deviation weighting coefficient is determined based on the proportion of the number of risk points in the total number of risk points in the underwriting analysis and management objectives.

9. The intelligent underwriting management method for special operations insurance supervision as described in claim 1, characterized in that, The method for determining the underwriting analysis management method of the enterprise is as follows: Based on the manual underwriting intervention data of different enterprises, identify the enterprises that have engaged in manual underwriting among the aforementioned enterprises; The degree of consistency between the manual underwriting results of enterprises with manual underwriting involvement and the underwriting processing results of the model is used to determine the number of deviations of risk points in different enterprises with manual underwriting involvement. Based on the number of enterprises with manual underwriting involvement and the number of deviations in risk points among different enterprises with manual underwriting involvement, the underwriting analysis and management method for enterprises other than the underwriting analysis and management objectives is determined.

10. An intelligent underwriting management platform for special operations insurance supervision, employing the intelligent underwriting management method for special operations insurance supervision as described in any one of claims 1-9, characterized in that, Specifically, it includes: Target screening module, intervention management module, underwriting analysis module; The target screening module is responsible for determining the underwriting analysis and management objectives of the model. The intervention management module is responsible for determining the intervention management method for the company's manual underwriting. The underwriting analysis module is responsible for determining the underwriting analysis management methods for enterprises other than the underwriting analysis management objectives.

Images