Insurance business review methods, devices, electronic equipment and storage media
By combining risk prediction models and custom annotations with aspect components, the problem of low efficiency in traditional insurance business review is solved, achieving automated and accurate insurance business review.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINA PING AN LIFE INSURANCE CO LTD
- Filing Date
- 2026-05-09
- Publication Date
- 2026-07-31
AI Technical Summary
Traditional insurance business review relies on human experience, resulting in slow review speed, difficulty in handling large-scale insurance business data, and low efficiency.
Risk prediction models are used to evaluate insurance business data. Combined with custom annotations and aspect components, automated review is achieved through different verification interfaces, improving review efficiency and accuracy.
It has enabled automated review of insurance business data, improved review efficiency, and ensured the accuracy and reliability of review results.
Smart Images
Figure CN122492369A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of artificial intelligence technology, and is applied to the field of financial technology, particularly to a method, apparatus, electronic device, and storage medium for reviewing insurance business. Background Technology
[0002] Insurance business review refers to a secondary risk assessment of relevant data (such as policy information, claim amounts, and insurance amounts after policy changes) after the completion of insurance business processes such as underwriting, claims settlement, and policy maintenance, in order to identify potential policy risks. Traditional insurance business reviews mainly rely on human experience; however, manual reviews are time-consuming, slow, and unable to handle large-scale insurance business data, resulting in low review efficiency. Summary of the Invention
[0003] The main objective of this application is to provide an insurance business review method, apparatus, electronic device, and storage medium, which aims to improve the review efficiency of insurance business.
[0004] To achieve the above objectives, a first aspect of this application proposes an insurance business review method, the method comprising: Obtain insurance business operations, execute the insurance business operations, and obtain insurance business data; The insurance business data is assessed using a pre-defined risk prediction model to obtain a target risk score. The insurance review procedure is determined based on the target risk score. According to the insurance review operation, a preset annotation is obtained; wherein, the preset annotation includes a first interface name, a second interface name, and an interface comparison parameter, the first interface name is the name of the first verification interface, the second interface name is the name of the second verification interface, and the interface comparison parameter is used to compare the first verification interface and the second verification interface; The aspect component is invoked based on the insurance review operation and the interface comparison parameters. The insurance business data is reviewed by calling the first verification interface based on the facet component and the first interface name to obtain a first review result; wherein, the first review result includes review passed or review failed; The insurance business data is reviewed by calling the second verification interface based on the facet component and the second interface name to obtain a second review result; wherein, the second review result includes review passed or review failed; Based on the first review result and the second review result, the insurance review result of the insurance review operation is determined; wherein, the insurance review result includes review passed or review failed.
[0005] In some embodiments, the risk prediction model includes a first prediction sub-model and a second prediction sub-model, the first prediction sub-model and the second prediction sub-model being different. The step of performing a risk assessment on the insurance business data using a preset risk prediction model to obtain a target risk score includes: The insurance business data is subjected to a first risk assessment using the first prediction sub-model to obtain a first risk score. A second risk assessment is performed on the insurance business data using the second prediction sub-model to obtain a second risk score. Calculate the score difference between the first risk score and the second risk score; The target risk score is calculated based on the score difference value, the first risk score, and the second risk score.
[0006] In some embodiments, calculating the target risk score based on the score difference value, the first risk score, and the second risk score includes: Calculate the first score confidence level of the first risk score based on the score difference value; Calculate the second confidence level of the second risk score based on the score difference value; Calculate the confidence difference between the first rating confidence level and the second rating confidence level; The first risk score and the second risk score are filtered based on the confidence difference value to obtain the target risk score.
[0007] In some embodiments, calculating the first confidence level of the first risk score based on the score difference value includes: A random deactivation layer is added to the first prediction sub-model to obtain a reference prediction sub-model; A third risk assessment is performed on the insurance business data using the reference prediction sub-model to obtain a predicted risk score. Calculate the variance of the predicted risk score; The confidence level of the first rating is calculated based on the rating variance.
[0008] In some embodiments, the step of filtering the first risk score and the second risk score based on the confidence difference value to obtain the target risk score includes: Based on the confidence difference value, obtain the first version of the first prediction sub-model and the second version of the second prediction sub-model; Compare the first version and the second version to obtain the version comparison results; The first risk score and the second risk score are filtered based on the version comparison results to obtain the target risk score.
[0009] In some embodiments, the step of filtering the first risk score and the second risk score based on the version comparison result to obtain the target risk score includes: If the version comparison result indicates that the first version and the second version are the same, then compare the confidence levels of the first rating and the second rating. If the confidence level of the first score is greater than or equal to the confidence level of the second score, then the first risk score is taken as the target risk score; If the confidence level of the first score is less than that of the second score, then the second risk score is taken as the target risk score.
[0010] In some embodiments, determining the insurance review operation based on the target risk score includes: The target score range is selected from the preset score range based on the target risk score; The risk level is determined based on the target scoring range; The insurance review operation is determined based on the risk level.
[0011] To achieve the above objectives, a second aspect of this application provides an insurance business review device, the device comprising: The operation acquisition module is used to acquire insurance business operations, execute the insurance business operations, and obtain insurance business data. The risk assessment module is used to assess the risk of the insurance business data through a preset risk prediction model and obtain a target risk score. The judgment module is used to determine the insurance review operation based on the target risk score; The annotation acquisition module is used to acquire preset annotations according to the insurance review operation; wherein, the preset annotations include a first interface name, a second interface name and interface comparison parameters, the first interface name is the name of the first verification interface, the second interface name is the name of the second verification interface, and the interface comparison parameters are used to compare the first verification interface and the second verification interface; The component invocation module is used to invoke the aspect component based on the insurance review operation and the interface comparison parameters. The first interface calling module is used to call the first verification interface to review the insurance business data according to the aspect component and the first interface name, and obtain a first review result; wherein, the first review result includes review passed or review failed; The second interface calling module is used to call the second verification interface to review the insurance business data according to the aspect component and the second interface name, and obtain a second review result; wherein, the second review result includes review passed or review failed; The review module is used to determine the insurance review result of the insurance review operation based on the first review result and the second review result; wherein the insurance review result includes review passed or review failed.
[0012] To achieve the above objectives, a third aspect of the present application provides an electronic device, the electronic device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method described in the first aspect.
[0013] To achieve the above objectives, a fourth aspect of the present application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.
[0014] The insurance business review method, device, electronic device, and computer-readable storage medium proposed in this application obtain insurance business data related to the insurance business operations by acquiring and executing these operations. To identify whether the insurance business data poses a risk, a risk prediction model is used to assess the data and obtain a target risk score. This target risk score is then used to determine whether to review the insurance business data. The insurance review operation is determined based on the target risk score. To avoid repeatedly writing review logic code, custom annotations combined with aspect components are used to centrally manage the review logic. The aspect components are called based on the insurance review operation and preset annotations to automate the insurance business review, thereby improving review efficiency. To improve review accuracy, a first verification interface and a second verification interface are introduced. These different verification interfaces are used to review the insurance business data, ensuring the accuracy and reliability of the review results. Attached Figure Description
[0015] Figure 1 This is a flowchart of the insurance business review method provided in the embodiments of this application; Figure 2 yes Figure 1 The flowchart of step S120 in the middle; Figure 3 yes Figure 2 The flowchart of step S240 in the text; Figure 4 yes Figure 3 The flowchart of step S310 in the process; Figure 5 yes Figure 3 The flowchart of step S340 in the text; Figure 6 yes Figure 5 The flowchart of step S530 in the text; Figure 7 yes Figure 1 The flowchart of step S130 in the process; Figure 8 This is a schematic diagram of the structure of the insurance business review device provided in the embodiments of this application; Figure 9 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation
[0016] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.
[0017] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.
[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.
[0019] Insurance business review refers to a secondary risk assessment of relevant data (such as policy information, claim amounts, and insurance amounts after policy changes) after the completion of insurance business processes such as underwriting, claims settlement, and policy maintenance, in order to identify potential policy risks. Traditional insurance business reviews mainly rely on human experience; however, manual reviews are time-consuming, slow, and unable to handle large-scale insurance business data, resulting in low review efficiency.
[0020] Based on this, embodiments of this application provide an insurance business review method, an insurance business review device, an electronic device, and a computer-readable storage medium, aiming to improve the review efficiency of insurance business.
[0021] The insurance business review method, insurance business review device, electronic device, and computer-readable storage medium provided in this application are specifically described through the following embodiments. First, the insurance business review method in this application embodiment is described.
[0022] The insurance business review method provided in this application relates to the field of artificial intelligence technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application implementing the insurance business review method, but is not limited to the above forms.
[0023] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.
[0024] Figure 1 This is an optional flowchart of the insurance business review method provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S110 to S180.
[0025] Step S110: Obtain insurance business operations, execute insurance business operations, and obtain insurance business data; Step S120: Perform a risk assessment on the insurance business data using a preset risk prediction model to obtain a target risk score; Step S130: Determine the insurance review operation based on the target risk score; Step S140: Obtain preset annotations according to the insurance review operation; wherein, the preset annotations include a first interface name, a second interface name, and interface comparison parameters. The first interface name is the name of the first verification interface, the second interface name is the name of the second verification interface, and the interface comparison parameters are used to compare the first verification interface and the second verification interface. Step S150: Call the aspect component according to the insurance review operation and interface comparison parameters; Step S160: Based on the aspect component and the first interface name, call the first verification interface to review the insurance business data and obtain the first review result; wherein, the first review result includes review passed or review failed; Step S170: Call the second verification interface to review the insurance business data according to the aspect component and the second interface name, and obtain the second review result; wherein, the second review result includes review passed or review failed; Step S180: Determine the insurance review result of the insurance review operation based on the first review result and the second review result; wherein, the insurance review result includes whether the review is passed or not.
[0026] Steps S110 to S180, as illustrated in this embodiment, automate the review of insurance business by centrally managing the review logic through custom annotations and aspect components. Compared with manual review, this improves review efficiency. Furthermore, by auditing insurance business data through different verification interfaces, the accuracy and reliability of the review results are ensured.
[0027] In step S110 of some embodiments, insurance business operations are obtained. Insurance business operations are specific business operation types related to insurance business, such as underwriting operations, claims operations, and policy maintenance operations. Executing insurance business operations completes the business processes of policy underwriting, claims, and policy maintenance, resulting in insurance business data. Insurance business data is the business data generated after executing insurance business operations. For example, if the insurance business operation is an underwriting operation, the insurance business data is policy data, including policy number, insured amount, premium, insurance period, and insured information. If the insurance business operation is a claims operation, the insurance business data is claims data, including claims application number, claims amount, cause of accident, and payment date. If the insurance business operation is a policy maintenance operation, the insurance business data is policy maintenance data, including policy maintenance application number, policy maintenance type, change details, and operation date.
[0028] To identify potential risks in insurance business data, the data is input into a risk prediction model for risk assessment, resulting in a target risk score. A higher target risk score indicates a higher probability of risk in the insurance business data. Risk prediction models can employ techniques such as random forests, long short-term memory networks, and convolutional neural networks. These models can be used to assess the risks of underwritten policies, paid-out insurance cases, policy changes related to policy maintenance, and policy surrenders, thereby identifying potential risks such as misinformation, insurance fraud, and abnormal operations.
[0029] Please see Figure 2 In some embodiments, step S120 may include, but is not limited to, steps S210 to S240: Step S210: Perform a first risk assessment on the insurance business data using the first prediction sub-model to obtain a first risk score; Step S220: Perform a second risk assessment on the insurance business data using the second prediction sub-model to obtain a second risk score; Step S230: Calculate the score difference between the first risk score and the second risk score; Step S240: Calculate the target risk score based on the score difference value, the first risk score, and the second risk score.
[0030] In some embodiments, step S210 uses a single risk prediction model for risk assessment, which can lead to low prediction accuracy. To improve the accuracy of risk prediction, the risk prediction model includes a first prediction sub-model and a second prediction sub-model, which are different models. For example, the first prediction sub-model is a long short-term memory network, and the second prediction sub-model is a convolutional neural network. Insurance business data is input into the first prediction sub-model for risk assessment to obtain a first risk score.
[0031] In step S220 of some embodiments, insurance business data is input into a second prediction sub-model for risk assessment to obtain a second risk score.
[0032] In step S230 of some embodiments, a subtraction operation is performed on the first risk score and the second risk score to obtain a score difference. The absolute value of the score difference is used as the score difference value. The score difference value is used to measure the risk assessment deviation between the first prediction sub-model and the second prediction sub-model for the same insurance business data.
[0033] In step S240 of some embodiments, a target risk score is obtained by calculating a score based on the score difference value, the first risk score, and the second risk score.
[0034] Through the above steps S210 to S240, the target risk score of the insurance business data can be obtained, and the insurance review operation can be determined based on the target risk score.
[0035] Please see Figure 3 In some embodiments, step S240 may include, but is not limited to, steps S310 to S340: Step S310: Calculate the first score confidence level of the first risk score based on the score difference value; Step S320: Calculate the second score confidence level of the second risk score based on the score difference value; Step S330: Calculate the confidence difference between the first rating confidence level and the second rating confidence level; Step S340: Filter the first risk score and the second risk score based on the confidence difference value to obtain the target risk score.
[0036] In steps S310 to S320 of some embodiments, if the score difference value is greater than a preset score difference threshold, it indicates that the risk assessment deviation between the first prediction sub-model and the second prediction sub-model is large. In order to determine the credibility of the first risk score predicted by the first prediction sub-model, the confidence level of the first score is obtained. In order to determine the credibility of the second risk score predicted by the second prediction sub-model, the confidence level of the second score is obtained.
[0037] In step S330 of some embodiments, the confidence levels of the first and second scores are subtracted to obtain a confidence difference. The absolute value of this confidence difference is used as the confidence difference value. The confidence difference value is used to measure the prediction reliability deviation between the first and second prediction sub-models.
[0038] In step S340 of some embodiments, if the confidence difference value is greater than the preset confidence difference threshold, it indicates that the prediction confidence of the first prediction sub-model and the second prediction sub-model has a large deviation. Then, the first risk score and the second risk score are screened to obtain the target risk score.
[0039] If the confidence difference value is less than or equal to the preset confidence difference threshold, it indicates that the prediction confidence deviation between the first prediction sub-model and the second prediction sub-model is small. Considering that the score difference between the first risk score and the second risk score is large, the average of the first risk score and the second risk score is calculated to obtain the target risk score.
[0040] If the score difference is less than or equal to the preset score difference threshold, it indicates that the risk assessment deviation between the first prediction sub-model and the second prediction sub-model is small. Either the first risk score or the second risk score can be randomly selected as the target risk score, or the maximum value of the first risk score and the second risk score can be selected to obtain the target risk score.
[0041] Through steps S310 to S340 above, the risk score is screened based on the model prediction credibility to obtain a more accurate risk score, thereby improving the accuracy of risk prediction.
[0042] Please see Figure 4 In some embodiments, step S310 may include, but is not limited to, steps S410 to S440: Step S410: Add a random deactivation layer to the first prediction sub-model to obtain the reference prediction sub-model; Step S420: A third risk assessment is performed on the insurance business data by referring to the prediction sub-model to obtain the predicted risk score; Step S430: Calculate the score variance of the predicted risk score; Step S440: Calculate the confidence level of the first rating based on the rating variance.
[0043] In step S410 of some embodiments, the first prediction sub-model includes a fully connected layer, and a dropout layer is added after the fully connected layer to obtain a reference prediction sub-model. The dropout layer has a dropout rate between 0 and 1, representing the proportion of neurons dropped. The dropout layer randomly selects neurons from the fully connected layer according to the dropout rate and sets the output of that neuron to 0.
[0044] In step S420 of some embodiments, insurance business data is input into a reference prediction sub-model for risk assessment to obtain a predicted risk score.
[0045] In step S430 of some embodiments, the random deactivation layer randomly discards a portion of the neuron outputs, causing the predicted risk score output by the reference prediction sub-model based on insurance business data to differ each time. Multiple predicted risk scores from the reference prediction sub-model based on insurance business data are obtained; the number of predictions can be set according to actual conditions, such as 10. The mean of each predicted risk score is calculated to obtain the average score. The variance is calculated based on each predicted risk score and the average score to obtain the score variance. The score variance reflects the degree of deviation between the predicted risk score and the average score.
[0046] In step S440 of some embodiments, a first score confidence level is calculated based on the score variance. The smaller the score variance, the higher the stability of the predicted risk score. The first prediction sub-model can more accurately assess risk when faced with new, unseen insurance business data, resulting in higher prediction reliability and a greater first score confidence level. The formula for calculating the first score confidence level is as follows: , Where var represents the rating variance.
[0047] The confidence level of the second rating can be calculated by referring to steps S410 to S440.
[0048] Through steps S410 to S440 above, the credibility of the model prediction results can be determined, so as to screen risk scores based on credibility.
[0049] Please see Figure 5 In some embodiments, step S340 may also include, but is not limited to, steps S510 to S530: Step S510: Obtain the first version of the first prediction sub-model and the second version of the second prediction sub-model based on the confidence difference value; Step S520: Compare the first version and the second version to obtain the version comparison result; Step S530: Based on the version comparison results, the first risk score and the second risk score are filtered to obtain the target risk score.
[0050] In step S510 of some embodiments, if the confidence difference value is greater than a preset confidence difference threshold, considering the impact of version differences on the accuracy and reliability of the prediction results, a first version of the first prediction sub-model and a second version of the second prediction sub-model are obtained. The first version has a first version number, and the second version has a second version number. As the model is continuously improved and optimized, the version number will gradually increase.
[0051] In step S520 of some embodiments, a first version number and a second version number are compared to obtain a version comparison result. The version comparison result includes the first version number being less than the second version number, the first version number being equal to the second version number, or the first version number being greater than the second version number.
[0052] In step S530 of some embodiments, in the three cases of the first version number being less than the second version number, the first version number being equal to the second version number, and the first version number being greater than the second version number, the first risk score and the second risk score are filtered by different filtering methods to obtain the target risk score.
[0053] By using steps S510 to S530 above, a more accurate risk score can be obtained.
[0054] Please see Figure 6 In some embodiments, step S530 may include, but is not limited to, steps S610 to S630: Step S610: If the version comparison result indicates that the first version and the second version are the same, then compare the confidence levels of the first rating and the second rating. Step S620: If the confidence level of the first score is greater than or equal to the confidence level of the second score, then the first risk score is taken as the target risk score. Step S630: If the confidence level of the first score is less than that of the second score, then the second risk score is taken as the target risk score.
[0055] In step S610 of some embodiments, if the version comparison result indicates that the first version and the second version are the same, that is, the first version number is equal to the second version number, it means that the performance of the first prediction sub-model and the second prediction sub-model are similar, and then the first score confidence and the second score confidence are compared.
[0056] In step S620 of some embodiments, if the confidence level of the first score is greater than or equal to the confidence level of the second score, it means that the prediction confidence level of the first prediction sub-model is greater than or equal to the prediction confidence level of the second prediction sub-model, and then the first risk score output by the first prediction sub-model is used as the target risk score.
[0057] In step S630 of some embodiments, if the confidence level of the first score is less than the confidence level of the second score, it means that the prediction confidence level of the first prediction sub-model is less than the prediction confidence level of the second prediction sub-model. Then, the second risk score output by the second prediction sub-model is used as the target risk score.
[0058] If the version comparison result shows that the first version number is less than the second version number, it indicates that the risk prediction performance of the second prediction sub-model is better than that of the first prediction sub-model. In this case, the second risk score output by the second prediction sub-model will be used as the target risk score. If the version comparison result shows that the first version number is greater than the second version number, it indicates that the risk prediction performance of the first prediction sub-model is better than that of the first prediction sub-model. In this case, the first risk score output by the first prediction sub-model will be used as the target risk score.
[0059] Steps S610 to S630 above improve the accuracy of risk prediction by selecting risk prediction results with higher credibility when the model performance is similar.
[0060] Please see Figure 7 In some embodiments, step S130 may include, but is not limited to, steps S710 to S730: Step S710: Select the target score range from the preset score range based on the target risk score; Step S720: Determine the risk level based on the target scoring range; Step S730: Determine the insurance review operation based on the risk level.
[0061] In step S710 of some embodiments, the preset scoring interval is a predefined range of risk scores, which may include a low-risk interval, a medium-risk interval, or a high-risk interval. If the target risk score falls within the preset scoring interval, then the preset scoring interval is used as the target scoring interval.
[0062] In step S720 of some embodiments, the risk level includes low risk, medium risk, or high risk. If the target scoring interval is a low-risk interval, the risk level is determined to be low risk; if the target scoring interval is a medium-risk interval, the risk level is determined to be medium risk; if the target scoring interval is a high-risk interval, the risk level is determined to be high risk.
[0063] In step S730 of some embodiments, if the risk level is medium risk, an insurance review operation is performed to review the insurance business data. If the risk level is low risk or high risk, no review of the insurance business data is required.
[0064] Through steps S710 to S730 above, insurance business data can be reviewed based on the target risk score to identify potential risks.
[0065] In step S140 of some embodiments, a preset annotation is obtained according to the insurance review operation. This preset annotation is a custom annotation based on Spring Boot, used to mark the methods requiring comparison and provide the metadata needed for the comparison. The preset annotation includes a first interface name, a second interface name, and interface comparison parameters. The first interface name is the name of the first verification interface, and the second interface name is the name of the second verification interface. The verification logic of the first and second verification interfaces is different; the first verification interface is the old system interface, and the second verification interface is the new system interface. The interface comparison parameters are used to indicate whether to perform a comparison between the first and second verification interfaces or not.
[0066] In step S150 of some embodiments, if the interface comparison parameters indicate that the first verification interface and the second verification interface should be compared, then the aspect component is invoked according to the insurance review operation. The aspect component is a class containing aspect logic, typically defined and managed using an AOP framework. If the interface comparison parameters indicate that the first verification interface and the second verification interface should not be compared, then the second verification interface is invoked to review the insurance business data, obtaining the insurance review result. The insurance review result includes whether the review passed or failed.
[0067] In step S160 of some embodiments, the insurance business data is reviewed by calling the first verification interface according to the aspect component and the first interface name, and a first review result is obtained. The first review result includes whether the review is passed or not.
[0068] In step S170 of some embodiments, the second verification interface is called to review the insurance business data according to the aspect component and the second interface name, and a second review result is obtained. The second review result includes whether the review is passed or not.
[0069] In step S180 of some embodiments, the first review result and the second review result are compared. If the first review result and the second review result are the same, then either the first review result or the second review result is used as the insurance review result of the insurance review operation. If the first review result and the second review result are different, then the second review result is used as the insurance review result. The insurance review result includes whether the review is passed or failed.
[0070] Please see Figure 8 This application also provides an insurance business review device that can implement the above-mentioned insurance business review method. The insurance business review device includes: The operation acquisition module 810 is used to acquire insurance business operations, execute insurance business operations, and obtain insurance business data. The risk assessment module 820 is used to assess the risk of insurance business data through a preset risk prediction model and obtain a target risk score. Module 830 is used to determine the insurance review operation based on the target risk score; The annotation acquisition module 840 is used to acquire preset annotations based on the insurance review operation. The preset annotations include a first interface name, a second interface name, and interface comparison parameters. The first interface name is the name of the first verification interface, the second interface name is the name of the second verification interface, and the interface comparison parameters are used to compare the first verification interface and the second verification interface. Component calling module 850 is used to call aspect components based on insurance review operations and interface comparison parameters; The first interface call module 860 is used to call the first verification interface to review insurance business data based on the aspect component and the first interface name, and obtain the first review result; wherein, the first review result includes review passed or review failed; The second interface calling module 870 is used to call the second verification interface to review insurance business data based on the aspect component and the second interface name, and obtain the second review result; wherein, the second review result includes review passed or review failed; The review module 880 is used to determine the insurance review result of the insurance review operation based on the first review result and the second review result; wherein the insurance review result includes whether the review is passed or not.
[0071] The specific implementation method of the insurance business review device is basically the same as the specific implementation method of the above-mentioned insurance business review method, and will not be described again here.
[0072] This application also provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the aforementioned insurance business verification method. This electronic device can be any smart terminal, including tablet computers, in-vehicle computers, etc.
[0073] Please see Figure 9 , Figure 9 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes: The processor 910 can be implemented using a general-purpose central processing unit (CPU), microprocessor, application specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application. The memory 920 can be implemented as a read-only memory (ROM), static storage device, dynamic storage device, or random access memory (RAM). The memory 920 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 920 and is called and executed by the processor 910 using the insurance business review method of the embodiments of this application. The input / output interface 930 is used to implement information input and output; The communication interface 940 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.). Bus 950 transmits information between various components of the device (e.g., processor 910, memory 920, input / output interface 930, and communication interface 940); The processor 910, memory 920, input / output interface 930 and communication interface 940 are connected to each other within the device via bus 950.
[0074] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described insurance business review method.
[0075] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.
[0076] The insurance business review method, device, electronic equipment, and computer storage medium provided in this application, through the use of custom annotations combined with aspect components to centrally manage review logic, achieve automated insurance business review, improving review efficiency compared to manual review methods. Furthermore, by auditing insurance business data through different verification interfaces, the accuracy and reliability of the review results are ensured.
[0077] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.
[0078] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.
[0079] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.
[0080] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.
[0081] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0082] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.
[0083] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0084] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0085] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0086] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0087] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.
Claims
1. An insurance business review method characterized by comprising: The method includes: Obtain insurance business operations, execute the insurance business operations, and obtain insurance business data; The insurance business data is assessed using a pre-defined risk prediction model to obtain a target risk score. The insurance review procedure is determined based on the target risk score. According to the insurance review operation, a preset annotation is obtained; wherein, the preset annotation includes a first interface name, a second interface name, and an interface comparison parameter, the first interface name is the name of the first verification interface, the second interface name is the name of the second verification interface, and the interface comparison parameter is used to compare the first verification interface and the second verification interface; The aspect component is invoked based on the insurance review operation and the interface comparison parameters. The insurance business data is reviewed by calling the first verification interface based on the facet component and the first interface name to obtain a first review result; wherein, the first review result includes review passed or review failed; The insurance business data is reviewed by calling the second verification interface based on the facet component and the second interface name to obtain a second review result; wherein, the second review result includes review passed or review failed; Based on the first review result and the second review result, the insurance review result of the insurance review operation is determined; wherein, the insurance review result includes review passed or review failed.
2. The method of claim 1, wherein, The risk prediction model includes a first prediction sub-model and a second prediction sub-model, which are different from each other. The step of performing a risk assessment on the insurance business data using the preset risk prediction model to obtain a target risk score includes: The insurance business data is subjected to a first risk assessment using the first prediction sub-model to obtain a first risk score. A second risk assessment is performed on the insurance business data using the second prediction sub-model to obtain a second risk score. Calculate the score difference between the first risk score and the second risk score; The target risk score is calculated based on the score difference value, the first risk score, and the second risk score.
3. The method of claim 2, wherein, The step of calculating the target risk score based on the score difference value, the first risk score, and the second risk score includes: Calculate the first score confidence level of the first risk score based on the score difference value; Calculate the second confidence level of the second risk score based on the score difference value; Calculate the confidence difference between the first rating confidence level and the second rating confidence level; The first risk score and the second risk score are filtered based on the confidence difference value to obtain the target risk score.
4. The method according to claim 3, characterized in that, The step of calculating the first confidence level of the first risk score based on the score difference value includes: A random deactivation layer is added to the first prediction sub-model to obtain a reference prediction sub-model; A third risk assessment is performed on the insurance business data using the reference prediction sub-model to obtain a predicted risk score. Calculate the variance of the predicted risk score; The confidence level of the first rating is calculated based on the rating variance.
5. The method according to claim 3, characterized in that, The step of filtering the first risk score and the second risk score based on the confidence difference value to obtain the target risk score includes: Based on the confidence difference value, obtain the first version of the first prediction sub-model and the second version of the second prediction sub-model; Compare the first version and the second version to obtain the version comparison results; The first risk score and the second risk score are filtered based on the version comparison results to obtain the target risk score.
6. The method according to claim 5, characterized in that, The step of filtering the first risk score and the second risk score based on the version comparison results to obtain the target risk score includes: If the version comparison result indicates that the first version and the second version are the same, then compare the confidence levels of the first rating and the second rating. If the confidence level of the first score is greater than or equal to the confidence level of the second score, then the first risk score is taken as the target risk score; If the confidence level of the first score is less than that of the second score, then the second risk score is taken as the target risk score.
7. The method according to any one of claims 1 to 6, characterized in that, The process of determining the insurance review operation based on the target risk score includes: The target score range is selected from the preset score range based on the target risk score; The risk level is determined based on the target scoring range; The insurance review operation is determined based on the risk level.
8. An insurance business verification device, characterized in that, The device includes: The operation acquisition module is used to acquire insurance business operations, execute the insurance business operations, and obtain insurance business data. The risk assessment module is used to assess the risk of the insurance business data through a preset risk prediction model and obtain a target risk score. The judgment module is used to determine the insurance review operation based on the target risk score; The annotation acquisition module is used to acquire preset annotations according to the insurance review operation; wherein, the preset annotations include a first interface name, a second interface name and interface comparison parameters, the first interface name is the name of the first verification interface, the second interface name is the name of the second verification interface, and the interface comparison parameters are used to compare the first verification interface and the second verification interface; The component invocation module is used to invoke the aspect component based on the insurance review operation and the interface comparison parameters. The first interface calling module is used to call the first verification interface to review the insurance business data according to the aspect component and the first interface name, and obtain a first review result; wherein, the first review result includes review passed or review failed; The second interface calling module is used to call the second verification interface to review the insurance business data according to the aspect component and the second interface name, and obtain a second review result; wherein, the second review result includes review passed or review failed; The review module is used to determine the insurance review result of the insurance review operation based on the first review result and the second review result; wherein the insurance review result includes review passed or review failed.
9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method according to any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the method described in any one of claims 1 to 7.