Process control method and device in insurance business

Abstract

The present disclosure relates to a process control method and device in the insurance business. Different large models are used for automatic claim settlement analysis for different types of insurance business, and the claim settlement analysis process is divided into two different large models for processing, thereby improving the efficiency and accuracy of claim settlement analysis. The method comprises: in response to an insurance claim operation in the interaction page, determining a target large model for claim settlement analysis of the insurance business from a plurality of pre-trained claim settlement large models according to the insurance type corresponding to the insurance business, and obtaining claim settlement materials related to the insurance business; performing claim settlement risk analysis based on the claim settlement materials through the first large model to obtain a risk assessment result, and obtaining a claim settlement suggestion for the insurance business based on the claim settlement materials and the risk assessment result through the second large model; and pushing the claim settlement materials and the claim settlement suggestion to relevant insurance business personnel.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, specifically to a process control method and apparatus in insurance business. Background Technology

[0002] Insurance business refers to the core underwriting and claims settlement activities of insurance companies, centered around insurance products. Among these, claims analysis is a crucial aspect of the insurance company's process control.

[0003] In related technologies, experienced claims adjusters or claims verifiers typically review claim application materials manually, comparing them with insurance contract terms to determine whether the claim falls under insured liability, the extent of the loss, and the amount of compensation. This manual claims analysis method is inefficient and struggles to handle insurance business scenarios with a surge in claims volume. Summary of the Invention

[0004] To overcome the problems existing in the related technologies, this disclosure provides a process control method and apparatus for insurance business, in order to solve the defects in the related technologies.

[0005] According to a first aspect of the present disclosure, a process control method in insurance business is provided, comprising: Displays an interactive page corresponding to insurance business, and the interactive page provides insurance claim payment function; In response to the insurance claim application operation on the interactive page, based on the insurance type corresponding to the insurance business, a target large model for claim analysis of the insurance business is determined from multiple pre-trained large claim models, and claim materials related to the insurance business are obtained. The target large model includes a first large model for claim risk analysis and a second large model for generating claim recommendations. The claim materials include at least one of the following: uploaded data, sensor data of the insured obtained from the policyholder, and historical claim data of the insured obtained from the insurance provider. The first major model performs a claims risk analysis based on the claims materials to obtain a risk assessment result, and the second major model obtains a claim payment recommendation for the insurance business based on the claims materials and the risk assessment result. The claim materials and compensation recommendations will be sent to the relevant insurance personnel.

[0006] In some embodiments, the insurance business is a corporate group insurance business, the uploaded data includes employee attendance data, injury photos, medical reports, and accident videos, the sensor data includes employee heart rate data and body temperature data, and the step of performing claims risk analysis based on the claims materials using the first big model to obtain risk assessment results includes: The first model constructs a risk profile based on the attendance, heart rate, and body temperature data of employees in the enterprise, and generates a risk assessment result based on the risk profile. The process of obtaining a claim settlement recommendation for the insurance business based on the claim materials and the risk assessment results using the second major model includes: The second major model is used to perform semantic analysis on the medical report, visual analysis on the injury photos, and behavioral reconstruction on the accident video to obtain accident liability information. Based on the accident liability information and the risk assessment results, a compensation recommendation corresponding to the insurance business is generated.

[0007] In some embodiments, the interactive page also provides a policy update function, and the process control method further includes: In response to the policy update operation on the interactive page, determine whether to add new insured employees based on the policy update operation, and if it is determined that new insured employees have been added, obtain the company's historical claims data and the attendance data, heart rate data and body temperature data of the employees in the company; Based on the company's historical claims data and the attendance, heart rate, and body temperature data of its employees, a premium big data model is used to generate personalized premiums for the company's employees. The personalized premium is displayed on the interactive page.

[0008] In some embodiments, generating personalized premiums for the enterprise using a premium big data model based on the enterprise's historical claims data and the attendance, heart rate, and body temperature data of the enterprise's employees includes: Based on the company's historical claims data and the attendance, heart rate and body temperature data of its employees, the premium model predicts the company's claims risk and generates a personalized premium for the company based on the predicted claims risk. The process control method also includes: Based on the compensation risk, a risk rectification suggestion is generated using a risk model, and the compensation risk and the risk rectification suggestion are displayed on the interactive page.

[0009] In some embodiments, the insurance business is a medical liability insurance business, and the uploaded data includes medical records, surgical consent forms, voice communication records of the patient's condition, and actual compensation claims. The step of performing a claims risk analysis based on the claims materials using the first large model to obtain a risk assessment result includes: Based on the medical records and surgical consent forms, the first model generates potential points of contention, and after transcribing the patient's medical communication speech into text, it performs emotion recognition to obtain the patient's informed consent information during the medical communication process. Based on the points of contention and the informed consent information, a risk assessment result is generated. The process of obtaining a claim settlement recommendation for the insurance business based on the claim materials and the risk assessment results using the second major model includes: The second major model uses the claims materials to find similar case payout ranges, and determines the liability ratio based on the claims materials and the risk assessment results. Based on the liability ratio and the similar case payout ranges, a predicted payout result is generated. Based on the predicted payout result and the actual payout request, a payout recommendation for the insurance business is generated.

[0010] In some embodiments, generating potential points of contention based on the medical records and surgical consent forms using the first large model includes: Based on the medical records and surgical consent forms, the first major model reconstructs the medical timeline and compares it with the standard medical timeline to obtain the medical node comparison results. Based on the medical node comparison results and the medical timeline, potential dispute points are generated. The process control methods include: The medical timeline is visualized, and potential points of dispute are marked on the medical timeline.

[0011] In some embodiments, the insurance business is elevator insurance business, and the process control method further includes: The AI ​​visual recognition algorithm identifies passenger behavior, number of people, and profiles in elevator monitoring videos to obtain elevator passenger information. It also fuses and calculates sensor data collected by multiple sensors installed in the elevator to obtain elevator operating status data. The elevator fault prediction model uses the elevator passenger information and elevator operation status data to monitor elevator faults. When an elevator fault is detected, an elevator fault warning is pushed to the policyholder of the elevator insurance business to prompt the policyholder whether to automatically initiate insurance claims. Upon receiving confirmation from the insured party regarding the automatic initiation of insurance claims, the insurance claims application process is automatically triggered.

[0012] According to a second aspect of the present disclosure, a process control device for insurance business is provided, comprising: The display module is used to display the interactive page corresponding to the insurance business, and the interactive page provides insurance claim payment function; The acquisition module is used to respond to the insurance claim application operation on the interactive page, determine the target big model for claim analysis of the insurance business from multiple pre-trained big model according to the insurance type corresponding to the insurance business, and acquire the claim materials related to the insurance business. The target big model includes a first big model for claim risk analysis and a second big model for generating claim suggestions. The claim materials include at least one of the following: uploaded data, sensor data of the insured obtained from the policyholder, and historical claim data of the insured obtained from the insurance provider. The analysis module is used to perform claims risk analysis based on the claims materials using the first major model to obtain risk assessment results, and to obtain compensation recommendations for the insurance business using the second major model based on the claims materials and the risk assessment results. The push module is used to push the claim materials and the compensation suggestions to relevant insurance business personnel.

[0013] According to a third aspect of the present disclosure, an electronic device is provided, the electronic device including a memory and a processor, the memory being configured to store computer instructions executable on the processor, and the processor being configured to implement the method described in any one of the first aspects when executing the computer instructions.

[0014] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the method described in any of the first aspects.

[0015] According to a fifth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, implements the method described in any of the first aspects.

[0016] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: The process control method for insurance business provided in this disclosure pre-trains multiple large-scale claims models corresponding to different insurance types. Then, based on the insurance type of the corresponding insurance business, a target large-scale model for claims analysis can be determined. Therefore, different large-scale models can be used for automatic claims analysis for different types of insurance businesses, thereby improving the efficiency and accuracy of claims analysis. Furthermore, the target large-scale model of this disclosure includes a first large-scale model for claims risk analysis and a second large-scale model for generating claims recommendations. This splits the claims analysis process into two stages, improving the parallel processing speed and accuracy of claims analysis, thereby further enhancing the efficiency and accuracy of claims analysis and better addressing insurance business scenarios with a surge in claims volume. Attached Figure Description

[0017] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0018] Figure 1 This is a flowchart illustrating a process control method in insurance business as shown in an exemplary embodiment of this disclosure; Figure 2 This is a schematic diagram of an interactive page in a process control method for insurance business, as illustrated in an exemplary embodiment of this disclosure; Figure 3 This is a schematic diagram of an interactive page in a process control method for insurance business, as illustrated in another exemplary embodiment of this disclosure; Figure 4 This is a schematic diagram of the structure of a process control device in insurance business, as shown in an exemplary embodiment of this disclosure; Figure 5 This is a structural block diagram of an electronic device illustrated in an exemplary embodiment of the present disclosure. Detailed Implementation

[0019] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0020] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0021] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0022] Firstly, at least one embodiment of this disclosure provides a process control method in insurance business; please refer to the appendix. Figure 1The diagram illustrates the process of the method, including steps S101 to S104.

[0023] In step S101, the interactive page corresponding to the insurance business is displayed, which provides insurance claim functions.

[0024] In step S102, in response to the insurance claim application operation on the interactive page, the target large model for claim analysis of the insurance business is determined from multiple pre-trained large claim models according to the insurance type corresponding to the insurance business, and the claim materials related to the insurance business are obtained.

[0025] The target large model includes a first large model for conducting claims risk analysis and a second large model for generating claims recommendations. Claim materials include at least one of the following: uploaded data, sensor data of the insured obtained from the policyholder, and historical claims data of the insured obtained from the insurance provider.

[0026] In step S103, the first major model performs a claim risk analysis based on the claim materials to obtain a risk assessment result, and the second major model obtains a claim payment recommendation for the insurance business based on the claim materials and the risk assessment result.

[0027] In step S104, the claim materials and compensation recommendations are sent to the relevant insurance business personnel.

[0028] By employing the above method, large-scale claims models corresponding to different insurance types are pre-trained. Then, based on the insurance type of the corresponding insurance business, a target large-scale model for claims analysis can be determined. Therefore, different large-scale models can be used for automated claims analysis for different types of insurance businesses, thereby improving the efficiency and accuracy of claims analysis. Furthermore, the target large-scale model disclosed herein includes a first large-scale model for claims risk analysis and a second large-scale model for generating claims recommendations. This splits the claims analysis process into two stages, improving the parallel processing speed and accuracy of claims analysis, thereby further enhancing the efficiency and accuracy of claims analysis and better addressing insurance business scenarios with surging claims volumes.

[0029] To facilitate understanding of the process control methods in insurance business provided in this disclosure, the above steps are further explained below.

[0030] For example, different interactive pages can be set for different insurance services, and the corresponding interactive page can be displayed to the policyholder according to the type of insurance service purchased. Alternatively, all policyholders can be shown the same homepage displaying different insurance service types. After a policyholder triggers a specific insurance service type, the interactive page corresponding to that triggered insurance service is displayed. Or, when a policyholder accesses the system, the type of insurance service purchased by the policyholder is determined based on their login information, and then the interactive page corresponding to that insurance service is displayed. This embodiment of the disclosure does not limit the specific display method of the interactive pages.

[0031] It should be understood that corresponding large-scale claims models can be trained in advance based on historical claims data for different types of insurance. For example, large-scale claims models A can be trained for elevator insurance, B for corporate group insurance, and C for medical malpractice insurance. Therefore, in practical applications, a target large-scale model corresponding to the actual insurance business can be selected from multiple pre-trained large-scale claims models for targeted claims analysis. Compared to using a single large-scale model for claims analysis, this approach allows for a more comprehensive integration of the claims characteristics of different insurance businesses, thereby improving the efficiency and accuracy of claims analysis.

[0032] In some embodiments, the insurance business is a group insurance business for enterprises. The uploaded data includes employee attendance data, injury photos, medical reports, and accident videos. Sensor data includes employee heart rate and body temperature data. Accordingly, the first major model performs claims risk analysis based on the claims materials to obtain risk assessment results, including: constructing a unified risk profile based on employee attendance, heart rate, and body temperature data using the first major model, and generating risk assessment results based on the unified risk profile. Accordingly, the second major model obtains insurance payment recommendations based on the claims materials and risk assessment results, including: performing semantic analysis on medical reports, visual analysis on injury photos, and behavioral reconstruction on accident videos using the second major model to obtain accident liability information, and generating corresponding insurance payment recommendations based on the accident liability information and risk assessment results.

[0033] It should be understood that corporate group insurance is a form of insurance where the company or group is the policyholder, the members of the group are the insured, and the insurer provides insurance coverage through a single insurance contract. In other words, the insured are the company's employees. Accordingly, the interactive page can prompt users to upload employee attendance data, injury photos, medical reports, and accident videos. Attendance data can include employee job titles, working hours, and work environment records. Additionally, with employee consent, smart bracelets or smart helmets can be used to monitor employees' physiological indicators such as heart rate and body temperature.

[0034] For example, employee attendance, heart rate, and body temperature data can be input into the primary model corresponding to the company's group insurance business. This model then performs cross-system semantic association to obtain risk profiles of the employees. The primary model for group insurance is trained using historical risk data from the company's group insurance business, including historical attendance, heart rate, and body temperature data. Furthermore, risk profiles for different employees are stored in a unified format. For instance, based on employee attendance, heart rate, and body temperature data, the primary model identifies a triple-overlapping risk profile of "nighttime overtime - high-temperature areas - abnormal heart rate," overcoming the limitations of traditional single-dimensional analysis.

[0035] For example, the second major model for corporate group insurance can be a multimodal large model, where accident liability information is used to characterize whether the accident was caused by work-related factors or a sudden personal illness. It should be understood that the second major model for corporate group insurance is trained using historical claims data, including historical medical reports, historical injury photos, and historical accident videos.

[0036] Therefore, large-scale models can be used to conduct targeted claims analysis on corporate group insurance business, thereby improving the efficiency and accuracy of claims analysis in corporate group insurance business.

[0037] In some embodiments, the interactive page corresponding to the corporate group insurance business also provides a policy update function. In response to the policy update operation on the interactive page, it can determine whether to add new insured employees based on the policy update operation. If it is determined that new insured employees have been added, it can obtain the company's historical claims data and the attendance data, heart rate data, and body temperature data of the employees in the company. Based on the company's historical claims data and the attendance data, heart rate data, and body temperature data of the employees in the company, it can generate personalized premiums for the employees in the company through a premium big data model. The personalized premiums are then displayed on the interactive page.

[0038] For example, such as Figure 2 As shown, personalized premiums for newly insured employees can be displayed to insured users.

[0039] It should be understood that related technologies typically rely on static occupational classification tables for premium calculation, which cannot flexibly adapt to the employment situation of different companies. In this embodiment, a large premium model can be used to generate personalized premiums, which can be flexibly adjusted according to the employment situation and historical claims history of different companies. The large premium model can be trained based on historical premium data, including historical claims data, historical premiums, and historical attendance, heart rate, and body temperature data of employees in other companies.

[0040] In some embodiments, the personalized insurance premium for a company is generated using a premium big data model based on its historical claims data and employee attendance, heart rate, and body temperature data. This includes: predicting the company's payout risk using the premium big data model based on the company's historical claims data and employee attendance, heart rate, and body temperature data, and generating a personalized insurance premium based on the predicted payout risk. Correspondingly, a risk mitigation suggestion can also be generated based on the payout risk using a risk big data model, and the payout risk and risk mitigation suggestion can be displayed on the interactive page.

[0041] For example, a premium model can determine whether a company has a history of claims and the specific details of those claims based on its historical claims data. Additionally, it can determine whether employees engage in high-risk work practices based on their attendance, heart rate, and body temperature data. If a company has a history of claims and its employees engage in high-risk practices, the premium model can predict a higher payout risk and thus generate a higher, personalized premium.

[0042] For example, the risk model can automatically generate a risk heat map for enterprises (such as a causal chain of "electrical safety defects - electric shock accidents - increased compensation costs"), and mark risk rectification suggestions (such as checking electrical safety) on the risk heat map so that enterprises can make targeted rectifications.

[0043] In some embodiments, the insurance business is medical liability insurance, and the uploaded data includes medical records, surgical consent forms, voice recordings of patient communication, and actual compensation claims. Accordingly, a first major model is used to perform a claims risk analysis based on the claims materials to obtain a risk assessment result. This includes: generating potential points of contention based on the medical records and surgical consent forms using the first major model; transcribing the voice recordings of patient communication into text and performing emotion recognition to obtain the patient's informed consent information during the communication process; and generating a risk assessment result based on the points of contention and the informed consent information. Correspondingly, a second major model is used to obtain a compensation recommendation for the insurance business based on the claims materials and the risk assessment result. This includes: finding similar case compensation ranges based on the claims materials using the second major model; determining the liability ratio based on the claims materials and the risk assessment result; generating a predicted compensation result based on the liability ratio and similar case compensation ranges; and generating a compensation recommendation for the insurance business based on the predicted compensation result and the actual compensation claim.

[0044] For example, such as Figure 3 As shown, in the medical liability insurance process, users can upload relevant documents such as the responsible party's qualifications and professional certification, medical records, a written claim application, and a description of the incident. Medical records include progress notes and surgical consent forms; the written claim application includes the actual compensation requested; and the description of the incident includes audio recordings of conversations about the patient's condition.

[0045] For example, the primary model for medical malpractice insurance is trained at least using historical medical records, surgical consent forms, and voice recordings of past medical conversations. The secondary model for medical malpractice insurance is trained at least using historical claims materials.

[0046] For example, if the difference between the predicted compensation result and the actual compensation claim exceeds a preset threshold, the insurance business can generate a compensation recommendation as: there is a risk of fraud, and the compensation should be manually verified.

[0047] In some embodiments, generating potential points of contention using a first major model based on medical records and surgical consent forms includes: reconstructing a medical timeline using the first major model based on medical records and surgical consent forms, comparing the medical timeline with a standard medical timeline to obtain medical node comparison results, and generating potential points of contention based on the medical node comparison results and the medical timeline. Correspondingly, the medical timeline can also be visualized, with potential points of contention marked and displayed within it.

[0048] Therefore, the first major model is used to reconstruct the medical timeline (such as medical order issuance - execution record - nursing log), and key node omissions are visualized, which can assist in liability determination and improve the efficiency and accuracy of claims analysis.

[0049] In some embodiments, the insurance business is elevator insurance. It can also use artificial intelligence visual recognition algorithms to identify passenger behavior, number of people, and profiles in elevator monitoring videos to obtain elevator passenger information. Furthermore, it can fuse sensor data collected by multiple sensors installed inside the elevator to obtain elevator operating status data. An artificial intelligence fault prediction model is used to monitor elevator faults based on the passenger information and operating status data. Upon detecting an elevator fault, an elevator fault warning is pushed to the policyholder of the elevator insurance business to prompt them whether to automatically initiate insurance claims. After receiving confirmation from the policyholder regarding the automatic initiation of insurance claims, the insurance claim application is automatically triggered.

[0050] In other words, elevator insurance not only provides manual claims application but also automatic claims processing, thereby reducing manual operations in the claims process and improving the user's claims experience.

[0051] According to a second aspect of the present disclosure, a process control device 400 for insurance business is provided. Please refer to the attached document. Figure 4 The process control device 400 in insurance business includes: Display module 401 is used to display the interactive page corresponding to the insurance business, and the interactive page provides insurance claim function; The acquisition module 402 is used to respond to the insurance claim application operation on the interactive page, determine the target big model for claim analysis of the insurance business from multiple pre-trained big model according to the insurance type corresponding to the insurance business, and acquire the claim materials related to the insurance business. The target big model includes a first big model for claim risk analysis and a second big model for generating claim suggestions. The claim materials include at least one of the following: uploaded data, sensor data of the insured obtained from the policyholder, and historical claim data of the insured obtained from the insurance provider. Analysis module 403 is used to perform claims risk analysis based on the claims materials using the first major model to obtain risk assessment results, and to obtain claims payment recommendations for the insurance business using the second major model based on the claims materials and the risk assessment results. The push module 404 is used to push the claim materials and the compensation suggestions to relevant insurance business personnel.

[0052] In some embodiments, the insurance business is a corporate group insurance business, the uploaded data includes employee attendance data, injury photos, medical reports, and accident videos, the sensor data includes employee heart rate data and body temperature data, and the analysis module 403 is specifically used for: The first model constructs a risk profile based on the attendance, heart rate, and body temperature data of employees in the enterprise, and generates a risk assessment result based on the risk profile. The second major model is used to perform semantic analysis on the medical report, visual analysis on the injury photos, and behavioral reconstruction on the accident video to obtain accident liability information. Based on the accident liability information and the risk assessment results, a compensation recommendation corresponding to the insurance business is generated.

[0053] In some embodiments, the interactive page also provides a policy update function, and the process control device 400 in the insurance business further includes a premium generation module, used for: In response to the policy update operation on the interactive page, determine whether to add new insured employees based on the policy update operation, and if it is determined that new insured employees have been added, obtain the company's historical claims data and the attendance data, heart rate data and body temperature data of the employees in the company; Based on the company's historical claims data and the attendance, heart rate, and body temperature data of its employees, a premium big data model is used to generate personalized premiums for the company's employees. The personalized premium is displayed on the interactive page.

[0054] In some embodiments, the premium generation module is specifically used for: Based on the company's historical claims data and the attendance, heart rate and body temperature data of its employees, the premium model predicts the company's claims risk and generates a personalized premium for the company based on the predicted claims risk. The process control device 400 in insurance business also includes a risk advice module, used for: Based on the compensation risk, a risk rectification suggestion is generated using a risk model, and the compensation risk and the risk rectification suggestion are displayed on the interactive page.

[0055] In some embodiments, the insurance business is a medical liability insurance business, and the uploaded data includes medical records, surgical consent forms, voice communication records of patient conditions, and actual compensation claims. The analysis module 403 is specifically used for: Based on the medical records and surgical consent forms, the first model generates potential points of contention, and after transcribing the patient's medical communication speech into text, it performs emotion recognition to obtain the patient's informed consent information during the medical communication process. Based on the points of contention and the informed consent information, a risk assessment result is generated. The second major model uses the claims materials to find similar case payout ranges, and determines the liability ratio based on the claims materials and the risk assessment results. Based on the liability ratio and the similar case payout ranges, a predicted payout result is generated. Based on the predicted payout result and the actual payout request, a payout recommendation for the insurance business is generated.

[0056] In some embodiments, the analysis module 403 is specifically used for: Based on the medical records and surgical consent forms, the first major model reconstructs the medical timeline and compares it with the standard medical timeline to obtain the medical node comparison results. Based on the medical node comparison results and the medical timeline, potential dispute points are generated. The process control device 400 in the insurance business also includes a dispute point display module, used for: The medical timeline is visualized, and potential points of dispute are marked on the medical timeline.

[0057] In some embodiments, the process control device 400 in the insurance business further includes a fault early warning module, used for: The AI ​​visual recognition algorithm identifies passenger behavior, number of people, and profiles in elevator monitoring videos to obtain elevator passenger information. It also fuses and calculates sensor data collected by multiple sensors installed in the elevator to obtain elevator operating status data. The elevator fault prediction model uses the elevator passenger information and elevator operation status data to monitor elevator faults. When an elevator fault is detected, an elevator fault warning is pushed to the policyholder of the elevator insurance business to prompt the policyholder whether to automatically initiate insurance claims. Upon receiving confirmation from the insured party regarding the automatic initiation of insurance claims, the insurance claims application process is automatically triggered.

[0058] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operations has been described in detail in the embodiments of the method in the first aspect, and will not be elaborated upon here.

[0059] Thirdly, please refer to the appendix. Figure 5 The diagram illustrates an exemplary block diagram of an electronic device 700, which may include a processor 701 and a memory 702. The electronic device 700 may also include one or more of a multimedia component 703, an input / output (I / O) interface 704, and a communication component 705.

[0060] The processor 701 controls the overall operation of the electronic device 700 to complete all or part of the steps in any of the methods described above. The memory 702 stores various types of data to support the operation of the electronic device 700. This data may include, for example, instructions for any application or method operating on the electronic device 700, and application-related data such as contact data, sent and received messages, pictures, audio, video, etc. The memory 702 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The multimedia component 703 may include a screen and audio components. The screen may be, for example, a touchscreen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in memory 702 or transmitted via communication component 705. The audio component also includes at least one speaker for outputting audio signals. I / O interface 704 provides an interface between processor 701 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual or physical buttons. Communication component 705 is used for wired or wireless communication between the electronic device 700 and other devices. Wireless communication may include Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof; therefore, the corresponding communication component 705 may include a Wi-Fi module, a Bluetooth module, or an NFC module.

[0061] In an exemplary embodiment, the electronic device 700 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the process control method in the insurance business described above.

[0062] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, which, when executed by a processor, implement the steps of any of the methods described above. For example, the computer-readable storage medium may be the memory 702 including the program instructions described above, which may be executed by the processor 701 of the electronic device 700 to perform any of the methods described above.

[0063] In another exemplary embodiment, a computer program product is also provided, which includes a computer program executable by a processor, wherein the computer program, when executed by the processor, implements the steps of any of the methods described above.

[0064] The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings. However, the present disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0065] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, this disclosure will not describe the various possible combinations separately.

[0066] Furthermore, various different embodiments of this disclosure can be combined in any way, as long as they do not violate the spirit of this disclosure, they should also be regarded as the content disclosed in this disclosure.

Claims

1. A process control method in insurance business, characterized in that, include: Displays an interactive page corresponding to insurance business, and the interactive page provides insurance claim payment function; In response to the insurance claim application operation on the interactive page, based on the insurance type corresponding to the insurance business, a target large model for claim analysis of the insurance business is determined from multiple pre-trained large claim models, and claim materials related to the insurance business are obtained. The target large model includes a first large model for claim risk analysis and a second large model for generating claim recommendations. The input of the second large model includes the risk assessment results output by the first large model. The claim materials include at least one of the following: uploaded data, sensor data of the insured obtained from the policyholder, and historical claim data of the insured obtained from the insurance provider. The first major model performs a claims risk analysis based on the claims materials to obtain a risk assessment result, and the second major model obtains a claim payment recommendation for the insurance business based on the claims materials and the risk assessment result. The claim materials and compensation recommendations will be sent to the relevant insurance personnel. The insurance business is medical liability insurance. The uploaded data includes medical records, surgical consent forms, voice communication records of patient conditions, and actual claim requests. The first big model performs a claim risk analysis based on the claimed materials to obtain risk assessment results, including: Based on the medical records and surgical consent forms, the first model generates potential points of contention, and after transcribing the patient's medical communication speech into text, it performs emotion recognition to obtain the patient's informed consent information during the medical communication process. Based on the points of contention and the informed consent information, a risk assessment result is generated. The process of obtaining a claim settlement recommendation for the insurance business based on the claim materials and the risk assessment results using the second major model includes: The second major model uses the claims materials to find similar case payout ranges, and determines the liability ratio based on the claims materials and the risk assessment results. Based on the liability ratio and the similar case payout ranges, a predicted payout result is generated. Based on the predicted payout result and the actual payout request, a payout recommendation for the insurance business is generated.

2. The process control method in insurance business according to claim 1, characterized in that, The insurance business is a corporate group insurance business. The uploaded data includes employee attendance data, injury photos, medical reports, and accident videos. The sensor data includes employee heart rate and body temperature data. The risk assessment results are obtained by performing a claims risk analysis based on the claims materials using the first big model, including: The first model constructs a risk profile based on the attendance, heart rate, and body temperature data of employees in the enterprise, and generates a risk assessment result based on the risk profile. The process of obtaining a claim settlement recommendation for the insurance business based on the claim materials and the risk assessment results using the second major model includes: The second major model is used to perform semantic analysis on the medical report, visual analysis on the injury photos, and behavioral reconstruction on the accident video to obtain accident liability information. Based on the accident liability information and the risk assessment results, a compensation recommendation corresponding to the insurance business is generated.

3. The process control method in insurance business according to claim 2, characterized in that, The interactive page also provides a policy update function, and the process control method further includes: In response to the policy update operation on the interactive page, determine whether to add new insured employees based on the policy update operation, and if it is determined that new insured employees have been added, obtain the company's historical claims data and the attendance data, heart rate data and body temperature data of the employees in the company; Based on the company's historical claims data and the attendance, heart rate, and body temperature data of its employees, a premium big data model is used to generate personalized premiums for the company's employees. The personalized premium is displayed on the interactive page.

4. The process control method in insurance business according to claim 3, characterized in that, The process involves using a premium model to generate personalized premiums for the company's employees based on the company's historical claims data and the attendance, heart rate, and body temperature data of its employees. This includes: Based on the company's historical claims data and the attendance, heart rate and body temperature data of its employees, the premium model predicts the company's claims risk and generates personalized premiums for the company's employees based on the predicted claims risk. The process control method also includes: Based on the compensation risk, a risk rectification suggestion is generated using a risk model, and the compensation risk and the risk rectification suggestion are displayed on the interactive page.

5. The process control method in insurance business according to claim 1, characterized in that, The process of generating potential points of contention based on the medical records and surgical consent forms using the first large model includes: Based on the medical records and surgical consent forms, the first major model reconstructs the medical timeline and compares it with the standard medical timeline to obtain the medical node comparison results. Based on the medical node comparison results and the medical timeline, potential dispute points are generated. The process control methods include: The medical timeline is visualized, and potential points of dispute are marked on the medical timeline.

6. The process control method in insurance business according to claim 1, characterized in that, The insurance business is elevator insurance, and the process control method also includes: The AI ​​visual recognition algorithm identifies passenger behavior, number of people, and profiles in elevator monitoring videos to obtain elevator passenger information. It also fuses and calculates sensor data collected by multiple sensors installed in the elevator to obtain elevator operating status data. The elevator fault prediction model uses the elevator passenger information and elevator operation status data to monitor elevator faults. When an elevator fault is detected, an elevator fault warning is pushed to the policyholder of the elevator insurance business to prompt the policyholder whether to automatically initiate insurance claims. Upon receiving confirmation from the insured party regarding the automatic initiation of insurance claims, the insurance claims application process is automatically triggered.

7. A process control device for insurance business, characterized in that, include: The display module is used to display the interactive page corresponding to the insurance business, and the interactive page provides insurance claim payment function; The acquisition module is used to respond to the insurance claim application operation on the interactive page, and determine the target big model for claim analysis of the insurance business from multiple pre-trained big model for the insurance business according to the insurance type corresponding to the insurance business, and acquire the claim materials related to the insurance business. The target big model includes a first big model for claim risk analysis and a second big model for generating claim suggestions. The input of the second big model includes the risk assessment result output by the first big model. The claim materials include at least one of the following: uploaded data, sensor data of the insured obtained from the policyholder, and historical claim data of the insured obtained from the insurance provider. The analysis module is used to perform claims risk analysis based on the claims materials using the first major model to obtain risk assessment results, and to obtain compensation recommendations for the insurance business using the second major model based on the claims materials and the risk assessment results. The push module is used to push the claim materials and the compensation suggestions to relevant insurance business personnel; The insurance business is medical liability insurance. The uploaded data includes medical records, surgical consent forms, voice communication records of the patient's condition, and actual compensation claims. The analysis module is used to generate potential dispute points based on the medical records and surgical consent forms using the first major model, and to transcribe the voice communication records of the patient's condition into text and perform emotion recognition to obtain the patient's informed consent information during the communication process. Based on the dispute points and the informed consent information, a risk assessment result is generated. Based on the claim materials using the second major model, similar cases' compensation ranges are found, and the liability ratio is determined based on the claim materials and the risk assessment result. Based on the liability ratio and the similar cases' compensation ranges, a predicted compensation result is generated. Based on the predicted compensation result and the actual compensation claim, a compensation recommendation for the insurance business is generated.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory being used to store computer instructions executable on the processor, and the processor being used to implement the steps of the method according to any one of claims 1-6 when executing the computer instructions.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps of the method according to any one of claims 1-6.

Images