An insurance application data integrated auditing method, device and system

Abstract

The application discloses an insurance application data integrated auditing method, device and system, solves the problems of redundant interactive requests and low data aggregation efficiency caused by the serial calling of the auditing process due to the scattered storage of the application data. The method comprises the following steps: obtaining application data containing at least two types of application data associated with the target application order identifier; analyzing the application data to generate user interface data for rendering a user application interface; obtaining a structured conclusion identifier determined from a pre-defined conclusion set from the user application interface; generating structured rectification data from the structured conclusion library according to the structured conclusion identifier; and updating the user interface according to the structured rectification data. The application reduces the number of interactive requests between the terminal and the server, reduces the data transmission volume and system resource occupation, and improves the aggregation response speed of multi-source heterogeneous data and the interface state update synchronization.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a method, apparatus and system for integrated review of insurance application information. Background Technology

[0002] As the digital transformation of the insurance industry accelerates, the drawbacks of traditional application document review models are becoming increasingly apparent. For example... Figure 12 As shown, current mainstream review systems design image review and dual recording (audio and video recording) quality inspection as two independent and parallel workflows. Specifically, after sales personnel complete the insurance information entry and dual video recording, they need to submit applications separately, triggering two independent review tasks. Subsequently, different reviewers conduct separate reviews and rectifications of the image data and dual video recordings in different systems. Finally, the system determines whether the insurance application materials are approved as a whole based on the two independent review conclusions.

[0003] This decentralized, highly manual-dependent architecture presents several serious challenges. First, in terms of efficiency and cost, the two processes repeatedly review overlapping content within the same insurance application, resulting in a waste of human resources. The parallel operation of two teams and systems also leads to fragmented resource allocation and wasted system resources. More fundamentally, due to the lack of comprehensive automated quality inspection capabilities for multimodal insurance application materials such as text, images, and videos, almost all verification work relies on manual labor, resulting in high manual operation costs. Second, in terms of timeliness and risk, the different progress of the two task lines for the same business extends the overall review cycle. For businesses that require manual review, the risk warnings provided by the system are not clear enough, and there are no structured standardized conclusions. Reviewers must manually input free-text comments, resulting in low efficiency and difficulty in standardizing processes. Operational risks due to staff fatigue or judgment bias are also difficult to avoid. Finally, regarding customer experience, the inconsistent timing and content of the conclusions from the two review lines can easily lead to customers being subjected to invalid or repeated requests for rectification. Furthermore, the existing rectification suggestions are often presented as a series of paragraphs with little connection between different rectification points, resulting in poor rectification effectiveness and severely impacting the underwriting experience.

[0004] In summary, existing technologies suffer from a series of interconnected drawbacks, including fragmented review processes, lack of automation capabilities, high operating costs, low review efficiency, significant operational risks, and poor customer experience. Therefore, there is an urgent need in this field for an integrated insurance application document review system and method that can integrate the review process, introduce intelligent analysis, and structure review conclusions and rectification suggestions to systematically solve the aforementioned problems and achieve the goals of cost reduction, efficiency improvement, and compliance risk control. Summary of the Invention

[0005] This application proposes an integrated method, apparatus, and system for reviewing insurance application materials, which solves the problems of redundant interactive requests and low data aggregation efficiency caused by the scattered storage of application materials and the serial calling of the review process in computer systems.

[0006] Firstly, this application provides an integrated review method for insurance application materials, including the following steps:

[0007] Obtain insurance application data; the insurance application data includes at least two types of insurance information associated with the target insurance policy identifier;

[0008] The insurance data is parsed to generate user interface data for rendering the user application interface;

[0009] Obtain the structured conclusion identifier determined from the predefined conclusion set from the user application interface;

[0010] Based on the structured conclusion identifier, structured rectification data is generated by matching from the structured conclusion database;

[0011] The user interface is updated based on the structured rectification data.

[0012] In one embodiment, at least two types of the insurance information include image files and audio / video files.

[0013] In one embodiment, the insurance data also includes risk warning information generated by intelligent analysis of at least two types of insurance information.

[0014] In one embodiment, the insurance data further includes risk warning information; the risk warning information includes at least one of the following:

[0015] Text verification prompts generated based on OCR recognition, semantic compliance prompts generated based on speech recognition, identity verification prompts generated based on face recognition, or operation compliance prompts generated based on white screen detection.

[0016] In one embodiment, at least two types of insurance information included in the insurance data are subject to automated compliance verification;

[0017] The automated compliance verification includes at least one of the following: OCR recognition verification, speech recognition verification, face recognition verification, or white screen detection.

[0018] In one embodiment, the structured rectification data is generated by matching from a structured conclusion database based on the structured conclusion identifier.

[0019] In one embodiment, the structured rectification data includes location information for locating a specific insurance information element in the user interface, as well as rectification operation instructions associated with that element.

[0020] Secondly, embodiments of this application also provide an integrated insurance application data verification device, used to implement the integrated insurance application data verification method described in any embodiment of the first aspect, comprising: an acquisition module, used to acquire application data; the application data includes at least two types of application data associated with a target application identifier; and is further used to acquire a structured conclusion identifier determined from a predefined conclusion set from a user application interface. A processing module, used to parse the application data and generate user interface data for rendering the user application interface. A determining module, used to match and generate structured rectification data from a structured conclusion library based on the structured conclusion identifier. An updating module, used to update the user interface based on the structured rectification data.

[0021] Thirdly, this application also proposes an integrated insurance application data verification system, comprising: a terminal device, deployed with the integrated insurance application data verification device as described in the second aspect embodiment, or configured to execute the method described in any one of the first aspect embodiments; and a server, communicatively connected to the terminal device, for providing the application data to the terminal device and generating the structured rectification data by matching the structured conclusion identifier from a structured conclusion database.

[0022] Fourthly, this application also proposes an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the method described in any embodiment of the first aspect of this application.

[0023] Fifthly, embodiments of this application also provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method described in any one of the embodiments of the first aspect.

[0024] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects:

[0025] This application reduces the number of interaction requests between the terminal and the server by aggregating at least two types of insurance application data stored separately in different systems to the same insurance application identifier; it achieves parallel loading and unified rendering of multi-source heterogeneous data by parsing the aggregated insurance application data to generate user interface data, thereby reducing the front-end interface rendering latency; and it transforms unstructured audit opinions into machine-parseable structured instructions by obtaining structured conclusion identifiers from a predefined conclusion set and matching them to generate structured rectification data, thereby improving data exchange efficiency and interface status update synchronization. Attached Figure Description

[0026] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0027] Figure 1 This is a schematic diagram of the insurance application document review process in the existing technology.

[0028] Figure 2 This is a flowchart illustrating the integrated review method for insurance application materials as described in this application.

[0029] Figure 3 This application provides an embodiment of an integrated insurance application document verification method for terminal devices. (Flowchart shown)

[0030] Figure 4 This application provides an embodiment of an integrated insurance application data verification method, which is used in a server-side device flowchart.

[0031] Figure 5 This is a flowchart illustrating the interaction between the insurance application data review system and an embodiment of this application.

[0032] Figure 6 This is a schematic diagram illustrating the types of insurance application materials used in an embodiment of this application.

[0033] Figure 7 This is a basic business process diagram illustrating the integrated review of insurance application materials in an embodiment of this application;

[0034] Figure 8 This is a structural diagram of the integrated insurance application data verification device according to an embodiment of this application;

[0035] Figure 9 This is a schematic diagram of an application scenario system according to an embodiment of this application;

[0036] Figure 10 This is a structural diagram of the server-side device according to an embodiment of this application;

[0037] Figure 11 This is a structural diagram of a terminal device according to an embodiment of this application;

[0038] Figure 12 This is a schematic diagram of the structure of a server-side device according to another embodiment of this application;

[0039] Figure 13 This is a block diagram of a terminal device according to another embodiment of this application. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0041] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0042] Figure 1 This is a schematic diagram of the existing insurance application document review process. For example... Figure 1 As shown, the existing review process begins with the sales staff's information collection phase. Then, for the same insurance application, two basic review tasks are initiated in parallel: one is assigned to one person for image review, and the other to another person for dual-recording quality inspection. These two review tasks are independent of each other. After the review conclusions are reached, rectification suggestions are issued through two basic group push channels (e.g., pushed to E-store and the dual-recording App respectively). Finally, the system needs to combine the review conclusions of both to determine whether the insurance application materials have passed the overall review. This process has inherent flaws such as task fragmentation, repetitive work, scattered conclusions, and low efficiency.

[0043] Figure 2 This is a flowchart illustrating the integrated review method for insurance application materials in an embodiment of this application.

[0044] In the first aspect, the embodiments of this application provide an integrated review method for insurance application materials, including steps 110-150.

[0045] Step 110: Obtain insurance data; the insurance data includes at least two types of insurance information associated with the target insurance policy identifier.

[0046] For example, this method uses the terminal device as the main user interaction subject and works in collaboration with the server to jointly realize the review process.

[0047] Send first request data to the server; the first request data includes the target insurance policy identifier.

[0048] In response to a user's audit operation on a specific insurance application on a terminal (such as an auditor's workstation or mobile device), the terminal device generates and sends the first request data to the server. The target insurance application identifier (such as the policy number) is the core business index uniquely associated with all insurance application materials. This request aims to retrieve all materials bound to this identifier, thereby aggregating the traditionally separate image audit and dual recording (audio and video recording) quality inspection tasks into a single audit task flow at the start of the process.

[0049] Receive first response data from the server, namely insurance application data; the first response data contains at least two types of insurance information associated with the target insurance policy identifier.

[0050] The terminal device receives the aggregated response returned by the server. Based on the received target insurance policy identifier, the server retrieves and integrates all relevant insurance application information from multiple data sources, such as the image library and the dual-recording video library, and encapsulates it into insurance application data for return. This data includes at least two types, such as image files (scanned copies of ID cards, bank cards, etc.) and audio and video files (videos of the dual-recording process).

[0051] Step 120: Parse the insurance data to generate user interface data for rendering the user application interface.

[0052] The first response data is parsed and displayed.

[0053] For example, the terminal device parses the received first response data, extracts the file data, metadata, and any additional information (such as the risk warnings mentioned later), and generates corresponding user interface data. The interface will use an integrated view, such as displaying image thumbnails and an embedded video player side by side, thus providing reviewers with a unified review environment that intuitively presents all related materials, avoiding the cumbersome switching between different systems.

[0054] Step 130: Obtain the structured conclusion identifier determined from the predefined conclusion set from the user application interface.

[0055] Retrieve the structured conclusion identifier determined by the user from a predefined set of conclusions.

[0056] For example, after reviewing the documents on the review interface, the auditor inputs their review conclusion. The key innovation of this application is that the auditor does not need to manually input free-text comments, but instead selects a conclusion from a standardized, predefined set of conclusions. This set corresponds to a structured audit conclusion library, with each conclusion corresponding to a unique structured conclusion identifier (such as a specific code). The auditor's selection is captured by the terminal and used as audit input data.

[0057] Step 140: Based on the structured conclusion identifier, generate structured rectification data by matching from the structured conclusion database.

[0058] For example, a second request data is sent to the server; the second request data includes the structured conclusion identifier, which is used by the server to generate structured rectification data based on the structured conclusion identifier.

[0059] The terminal device encapsulates the user-selected structured conclusion identifier into second request data and sends it to the server. This request triggers the server to perform intelligent matching based on the identifier. The server uses the identifier to search and match the corresponding, predefined structured rectification opinions from its maintained structured conclusion database, thereby generating machine-readable structured rectification data.

[0060] Receive second response data from the server; the second response data contains the structured rectification data.

[0061] The terminal device receives the second response data returned by the server, which includes structured rectification data generated by the server. This data is not a simple text description, but a structured instruction set rich in semantics that can be directly parsed by the terminal interface and used to drive interaction. For example, it contains problem location information and specific rectification operation instructions.

[0062] Step 150: Update the user interface based on the structured rectification data.

[0063] Update the user interface based on the second response data.

[0064] For example, the terminal device parses the received structured rectification data and dynamically updates the user interface accordingly. For instance, the interface can pop up a guide window to accurately locate the specific data element with a problem (such as a blurry ID card image) and clearly display the targeted rectification steps.

[0065] Understandable, Figure 2 The illustrated process outlines the core interactive steps of the integrated review method of this application. To more clearly demonstrate the respective logic and collaborative relationship between the terminal and server sides, this application will subsequently... Figure 3 and Figure 4 In this paper, the above process will be broken down and explained in detail from the perspectives of terminal devices and server devices. Figure 1 and Figure 3 , Figure 4 They complement each other and together constitute a complete description of the technical solution of this application.

[0066] Figure 3 This application provides an embodiment of an integrated review method for insurance application information, which is used in a terminal device flowchart.

[0067] In a first aspect, embodiments of this application provide an integrated verification method for insurance application materials, used in a terminal device, comprising:

[0068] Step 210: Send first request data; the first request data includes the target insurance policy identifier.

[0069] In this embodiment, the terminal device (such as an auditor's workstation or a salesperson's mobile device) responds to a user's audit trigger operation for a specific insurance application by sending first request data to the server. The target insurance application identifier is a key index used to uniquely identify an insurance application in the insurance business system, such as a policy number or application number. The system uses a unified task scheduling module to map image auditing and dual-recording quality inspection—two auditing tasks that are traditionally separate—into a single auditing task flow, using the target insurance application identifier as the unique task identifier for association and scheduling. The purpose is to obtain all insurance application materials associated with this identifier, thereby directly overcoming the drawbacks of the traditional decentralized manual auditing model, which easily leads to low policy issuance efficiency, duplicate audits, and scattered resource allocation. It combines the process that originally required two separate requests in the image system and the dual-recording system into one.

[0070] Step 220: Receive first response data; the first response data contains at least two types of insurance information associated with the target insurance policy identifier.

[0071] The terminal device receives the first response data returned by the server. The first response data is encapsulated by the server after retrieving and integrating multiple data sources such as the image library and the dual-recording video library according to the target insurance policy identifier. It contains at least two types of insurance information associated with the target insurance policy identifier.

[0072] The terminal device receives the first response data returned by the server. This first response data contains at least two types of insurance application materials associated with the target insurance policy identifier. The server, through a multi-source data aggregation interface, aggregates image and audio / video files associated with the same target insurance policy identifier into a unified service response data packet. This first response data transforms the traditional model where sales personnel submit insurance information and record videos separately, triggering two independent review processes (image review and dual-recording quality inspection, respectively), into an aggregation model that retrieves all materials at once. This allows subsequent reviews to be based on a complete set of materials, avoiding duplicate reviews of content in image review and dual-recording quality inspection. It also avoids the waste of manpower and resource fragmentation caused by two teams reviewing image review and dual-recording quality inspection on different systems.

[0073] In one embodiment, at least two types of the insurance information include image files and audio / video files.

[0074] The image files typically refer to electronic photos or scans of paper documents such as ID cards, bank cards, and health questionnaires required for insurance application.

[0075] The audio and video files mentioned specifically refer to video files generated during the dual recording (audio and video recording) process required by regulations, which record the key process of sales personnel explaining the terms to customers and customers confirming their intention to purchase insurance.

[0076] For example, a critical illness insurance application might include a photo of the customer's ID card, a photo of their bank card, a scanned copy of their medical examination report (image file), and a 15-minute dual-recorded video (audio and video file).

[0077] In one embodiment, the first response data further includes risk warning information generated by intelligent analysis of at least two types of insurance information.

[0078] In one embodiment, the first response data further includes risk warning information; the risk warning information includes at least one of the following:

[0079] Text verification prompts generated based on OCR recognition, semantic compliance prompts generated based on speech recognition, identity verification prompts generated based on face recognition, or operation compliance prompts generated based on white screen detection.

[0080] This application embodies the core innovation of constructing a multimodal data intelligent analysis model to accurately identify key operational risks.

[0081] Before returning the original data, the server has pre-analyzed the data using a background intelligent model and added any identified risk points to the data.

[0082] In other words, this application's embodiment introduces a pre-processing intelligent analysis step. By deploying a multimodal analysis model on the server side, it automatically and in parallel verifies the compliance of the aggregated insurance application data. The structured risk semantic information obtained from the verification is then associated with the original data and distributed together, thereby transforming the manual review mode from full screening to focused verification.

[0083] This application further defines the specific technical sources of the risk warning information. The multimodal analysis module adopts a parallel processing architecture, integrating an OCR recognition submodule, an ASR speech recognition submodule, a face recognition submodule, and a behavior analysis (such as white screen detection) submodule. Each submodule processes the data of its corresponding modality in parallel and outputs structured risk labels, which are finally aggregated into a unified risk warning dataset for distribution.

[0084] The text verification prompts are derived from OCR (Optical Character Recognition) technology, for example, recognizing that the ID number does not match the manually entered one.

[0085] The semantic compliance prompts are derived from ASR (Automatic Speech Recognition) and NLP technologies, for example, detecting that key disclaimers are not mentioned in dual recordings.

[0086] The identity verification prompt is based on facial recognition technology, for example, prompting that the person in the video has a low similarity to the ID card photo.

[0087] The compliance prompts are derived from video analytics technologies, such as white screen detection, which indicate that there has been an unauthorized interruption during the recording process.

[0088] These technologies together constitute a multimodal data intelligent analysis model, which realizes comprehensive automated quality inspection capabilities for text, images, and audio. Relying on technologies such as OCR recognition, rule verification, speech recognition, face recognition, on-the-fly interpretation, signature control, screen projection interaction, and blank screen detection, it performs intelligent analysis on three types of insurance application materials: text, images, and audio.

[0089] It should be noted that OCR recognition, speech recognition, face recognition, and white screen detection are only some embodiments of the multimodal data intelligent analysis model constructed in this application, used to illustrate how to automatically analyze insurance application materials of different modalities such as text, images, and audio. The technical solutions protected by this application are not limited to the specific technical means mentioned above. Any other model or combination of technologies that can automatically verify the compliance of at least two types of insurance application materials to achieve intelligent risk prompts and assist or replace some manual operations, such as image integrity detection based on deep learning, compliance analysis of text content by semantic understanding models, voiceprint recognition, and other intelligent analysis technologies that may emerge in the future, as long as they are applied to the integrated review process and system proposed in this application, are all within the protection scope of the technical solutions of this application.

[0090] Step 230: Parse the first response data, determine the interface rendering instructions, and generate user interface data.

[0091] After receiving the initial response data, which aggregates information and risk warnings, the terminal device needs to transform it into an interface that users can intuitively understand and operate. This embodiment of the application parses data packets, extracts file content, risk tags, and their relationships, and then generates the user interface data. This user interface data drives a unified rendering engine to construct an integrated review task interface on the screen. This interface employs a multi-threaded processing framework, loading and rendering different modalities of information in parallel, and achieving standardized display through an interface component library.

[0092] This process determines how to build an integrated review task interface on the screen. For example, the interface data will instruct the display of a list of image thumbnails in the left panel, with a corresponding risk label (such as a blurred icon) next to each image. A video player will be embedded in the right panel, and segments with risk warnings will be marked on the timeline (e.g., silent segments will be displayed as red bars). The front-end interface uses a Canvas drawing component and a timeline control to dynamically bind risk warnings to specific data elements (such as image areas and video segments), achieving risk visualization and contextualized display.

[0093] This presentation method directly addresses the problems of insufficiently clear risk warnings and the presentation of opinions in the form of entire paragraphs in the traditional model, which lacks correlation between the rectification points of different rectification items, resulting in poor rectification effects.

[0094] Step 240: Obtain the audit input data associated with the first response data from the user application interface; the audit input data is a structured conclusion identifier determined from a predefined conclusion set.

[0095] After reviewing the documents on the integrated review interface, the auditor inputs their decision.

[0096] In this embodiment, the auditor does not freely write an audit opinion, but instead selects conclusions from a standardized, structured list. This list is a predefined set of conclusions, derived from the structured quality inspection conclusion + rectification opinion standard system established in this application, and the standardized, structured insurance application document audit conclusion database embedded in the system. The system establishes a standardized conclusion coding system (such as DEFECT_001) and associates it with front-end interface component IDs and rectification action instructions to form mapping rules. By establishing a structured quality inspection conclusion + rectification opinion standard system and embedding a standardized, structured insurance application document audit conclusion database into the system, data structure management is facilitated.

[0097] For example, when faced with a blurry image of an ID card, the auditor does not need to manually enter "Customer's ID card photo is unclear, please re-upload," but can directly select the corresponding option 01 - "ID document is unclear" from the drop-down menu.

[0098] The only code behind this option is the structured conclusion identifier.

[0099] This design allows operators to directly select the corresponding question instead of manually entering it, which reduces the amount of manual operation and significantly shortens the time for manual review.

[0100] Step 250: Send second request data; the second request data includes the audit input data.

[0101] The terminal device takes the structured conclusion identifier selected by the user as the core review conclusion data, encapsulates it into second request data, and sends it back to the server, thereby submitting the final judgment of this manual review.

[0102] Step 260: Receive the second response data and update the user interface; the second response data contains structured data corresponding to the structured conclusion identifier.

[0103] Based on the received structured conclusion identifier, the server performs rule matching and returns second response data to the terminal. The structured data in this second response data is a detailed, machine-readable instruction set precisely bound to the conclusion identifier, not just text. The structured conclusion library uses a key-value pair storage structure, where the key is the structured conclusion identifier and the value is the corresponding structured rectification opinion object. This object contains interface element location information (such as component ID, XPath), rectification action instructions (such as highlighting or displaying specific text), and associated file identifiers. The matching results are encapsulated into a standard structured data format using technologies such as JSON serialization.

[0104] The terminal updates the user interface based on this data.

[0105] For example, a structured rectification window will pop up on the interface, clearly listing: Problem 1: Unclear ID card image (related file: ID card front.jpg). Rectification requirement: Please retake a clear photo of the front of your ID card and upload it. The system encapsulates the review conclusions and rectification instructions into a standard data format (such as JSON) through a defined structured data synchronization protocol, supporting adaptation and rendering on both the e-store and dual-recording app platforms, achieving a consistent structured and layered display across terminals. This approach enables a structured and layered display of problems and rectification measures on both the e-store and dual-recording app platforms, allowing customers and sales personnel to clearly understand the points requiring rectification and how to rectify them, thereby improving the underwriting success rate.

[0106] Figure 4 This application provides an embodiment of an integrated insurance application data verification method using a server-side device flowchart.

[0107] Secondly, this application also provides an integrated verification method for insurance application materials, for use in server-side equipment, including:

[0108] Step 310: Receive first request data; the first request data includes the target insurance policy identifier.

[0109] The server-side startup process receives the first request from the terminal and extracts the core business index: the target insurance policy identifier.

[0110] In one embodiment, before step 320 sends the first response data, the method further includes:

[0111] Perform automated compliance verification on at least two types of insurance application materials;

[0112] The automated compliance verification includes at least one of the following: OCR recognition verification, speech recognition verification, face recognition verification, or white screen detection.

[0113] Before distributing the data, the server uses a built-in multimodal data intelligent analysis model to automatically pre-screen the aggregated insurance application data. The automated compliance verification is performed by a rule executor and a threshold determination module working together. The automatic verification module makes a preliminary compliance judgment on the data, replacing some manual operations.

[0114] The automated compliance verification refers to a series of machine checks that do not require human intervention.

[0115] For example,

[0116] The OCR recognition verification automatically reads the text on the ID card image and compares it with the information filled in on the insurance application form to verify their consistency.

[0117] The voice recognition verification process converts the dual-recorded audio into text and analyzes whether the prescribed terms and conditions have been read completely and accurately.

[0118] The facial recognition verification compares the policyholder's face in the dual-recorded video with their ID card photo to verify whether the policyholder is the actual person who applied for the policy.

[0119] The white screen detection analyzes whether there is long-term invalid content or interruption in the video.

[0120] These verifications replace some basic, repetitive manual tasks, substitute for some review scenarios, and assist in manual review.

[0121] In one embodiment, before step 320 sends the first response data, the method further includes:

[0122] The system performs automated compliance verification on at least two types of insurance application materials and generates risk warning information based on the automated compliance verification; the first response data also includes the risk warning information.

[0123] In this embodiment, the server not only performs verification but also transforms the verification results into specific, readable risk warning information, which is then appended to the first response data to be issued. The system uses a risk tag generator to convert the outputs of each analysis submodule into structured risk semantic tags and dynamically binds these risk tags to elements in the original data interface. This allows the reviewer to be aware of potential problems in advance.

[0124] For example, if the verification finds the ID card to be blurry, a message will be generated: "ID card image is unclear, please verify." If the verification finds that key sentences are missing in the dual-recorded video, a message will be generated: "Term X was not read aloud, please supplement."

[0125] Step 320: Send first response data; the first response data contains at least two types of insurance information associated with the target insurance policy identifier.

[0126] After completing data aggregation and optional intelligent analysis and risk information addition, the server sends the integrated first response data back to the terminal device.

[0127] Step 330: Receive second request data; the second request data includes audit input data, which is a structured conclusion identifier determined from a predefined set of conclusions.

[0128] The server receives the manual review conclusion from the terminal, which is represented by a structured code (structured conclusion identifier).

[0129] Step 340: Based on the structured conclusion identifier, match the corresponding structured rectification opinions from the structured conclusion database, and then generate the structured rectification data.

[0130] In this embodiment, the server maintains a structured conclusion database. Upon receiving a structured conclusion identifier, the server performs precise matching according to preset rules (i.e., a structured conclusion rule engine) to retrieve the corresponding, pre-configured structured rectification opinions. The structured conclusion database is a key-value mapping library that integrates a template engine, supporting precise matching and generation of conclusion identifiers to structured rectification opinion objects. This process ensures accurate matching between problems, conclusions, and rectification measures.

[0131] For example, if the identifier DEFECT_ID_01 is received, a rectification suggestion containing specific operation instructions will be matched: Please re-upload a clear photo of the front of your ID card, ensuring that all four corners are intact and the text is clear.

[0132] In one embodiment, the structured rectification suggestions include location information for locating a specific insurance information element in the user interface, and rectification operation guidance associated with that element.

[0133] The structured rectification opinions described in this application embodiment are not simple text, but contain machine-readable instruction information.

[0134] For example, the location information could be a unique ID of a specific image component in the front-end interface, and the rectification operation guide could be a specific command for that component, such as highlighting and displaying the prompt text: 'Please replace'.

[0135] Step 350: Send second response data; the second response data contains structured data corresponding to the structured conclusion identifier.

[0136] The server encapsulates the generated structured data (i.e., the aforementioned structured rectification opinions or their transformed form) into second response data and sends it to the terminal device, thus completing the closed loop of the entire integrated review process. By changing the originally parallel system interaction into a serialized task chain controlled by a state machine engine and managing the dependencies between tasks, the sequential connection of the process is achieved.

[0137] Figure 5 This is a flowchart illustrating the interaction between the insurance application data review system and an embodiment of this application. Figure 5 As shown, the integrated audit method in Figure 2 The system architecture shown is specifically implemented through the collaboration of four core business systems: e-store, dual recording system, contract platform, and review platform. The following section will combine these with... Figure 5 The following is a detailed explanation of the interaction details of the integrated review process for this application within specific business systems:

[0138] The process begins at the e-store. Sales staff enter the insurance information here. The e-store then determines, based on its business rules, whether dual recording (audio and video recording) is required for this insurance application. This step corresponds to... Figure 3 The preparation stage before the terminal device sends the first request data.

[0139] If dual recording is required, the dual recording system is invoked to complete video recording and upload. The contract platform then receives image data uploaded from e-store (such as photos of ID cards and bank cards) and performs automated compliance verification (e.g., ...). Figure 4 (Intelligent analysis before step 420). This corresponds to... Figure 4 The core component of the process is the automated compliance verification of at least two types of insurance application materials on the server side.

[0140] The review platform receives preliminary results from the contract platform's intelligent review. If the intelligent review identifies clear risks or rule requirements, it can directly generate an automated review conclusion or determine that the review needs to be transferred to a human reviewer. Here, the review platform integrates... Figure 3 The review interface function of the terminal device, and Figure 4 The server-side rule matching and scheduling function allows reviewers to conduct initial / secondary reviews and select structured conclusion identifiers (such as...) from a predefined set of conclusions. Figure 3 (as described in step 340).

[0141] The review results (whether approved, rejected, or requiring rectification) will be simultaneously returned to the e-store and dual-recording system for display. Crucially, when the conclusion is "requiring rectification," the system does not issue two separate text comments, but rather... Figure 4Step 440 matches the generated structured rectification opinions, generates a clear image rectification task on the e-store terminal, and generates a dual-recording rectification task on the dual-recording App terminal.

[0142] Sales staff rectified video recordings on the e-store platform and videos on the dual-recording app. Once rectification was complete, the system notified both the contract platform and the review platform.

[0143] The platform has completed the rectification process, and the entire review process for insurance application materials is now finished.

[0144] pass Figure 5 The interactive flow shown demonstrates that the integrated review method proposed in this application intuitively showcases... Figure 3 and Figure 4 The complete implementation path of the method in a specific business system.

[0145] Figure 6 This is a schematic diagram illustrating the types of insurance application materials used in an embodiment of this application. For example... Figure 6 As shown, to achieve integrated intelligent review, the system performs the following specific processing on the main types of aggregated insurance application materials:

[0146] Insurance information: This involves verification using external data sources. For example, information such as occupation and employer filled in on the insurance application is compared and verified with trusted external databases (such as credit reporting systems and business registration databases) to confirm its authenticity and consistency.

[0147] Document Images: OCR (Optical Character Recognition) recognition and processing are performed. The system automatically extracts text information (such as name, document number, and expiration date) from images of documents such as ID cards and bank cards, and converts it into structured text for subsequent automatic filling, comparison, and compliance verification.

[0148] Dual recording (audio and video): Performing multimodal recognition and analysis. This includes, but is not limited to: Automatic Speech Recognition (ASR) to translate and analyze whether the content being explained is compliant; facial recognition to verify the identity of the person handling the matter; and technologies such as behavioral analysis and blank screen detection to ensure that the recording process complies with regulatory requirements.

[0149] Figure 7 This is a basic business process diagram illustrating the integrated review of insurance application materials in an embodiment of this application. Figure 7 As shown, the basic business process, in sequence as indicated by the arrows, includes the following main steps:

[0150] Information collection: The starting point of the process, completing the collection of insurance application materials and information.

[0151] Unified Upload: Submit all collected data in one go, in a unified manner.

[0152] One person reviews all insurance application materials: This is the core part of the process, where one person reviews all insurance application materials.

[0153] Below the step where one person reviews all insurance application materials, two key sub-processing modules are divided:

[0154] Preliminary intelligent review: Before or simultaneously with human review, the system automatically performs preliminary verification and analysis of the data.

[0155] Conclusion: Structured management involves standardizing and structurally encapsulating and managing the outputs generated during the review process (including intelligent analysis results and human judgment conclusions).

[0156] Following the review process, the workflow triggers two rectification branches in parallel based on the conclusions:

[0157] E-store rectification

[0158] Dual Recording App Rectification

[0159] Finally, the results of the two rectification branches mentioned above are merged into the synchronous rectification stage, marking the completion of the closed loop of the entire review and rectification process.

[0160] Figure 8 This is a structural diagram of the integrated insurance application data verification device according to an embodiment of this application.

[0161] This application also provides an integrated insurance application document verification device for implementing the above method. The device includes:

[0162] The acquisition module 801 is used to acquire insurance data, as described in step 110 of the embodiment; it is also used to acquire a structured conclusion identifier determined from a predefined conclusion set from the user application interface, as described in step 130 of the embodiment; the insurance data includes at least two types of insurance information associated with the target insurance policy identifier.

[0163] The processing module 802 is used to parse the insurance data and generate user interface data for rendering the user application interface, as described in step 120 of the embodiment.

[0164] The determination module 803 is used to match and generate structured rectification data from the structured conclusion library according to the structured conclusion identifier, as described in step 140 of the embodiment.

[0165] The update module 804 is used to update the user interface according to the structured rectification data, as described in step 150 of the embodiment.

[0166] Furthermore, the acquisition module includes:

[0167] The first acquisition unit is used to receive at least two types of insurance application information associated with the target insurance application identifier from the server, wherein the at least two types of insurance application information include image files and audio and video files.

[0168] The second acquisition unit is used to acquire the structured conclusion identifier determined by the user from a predefined set of conclusions from the user application interface.

[0169] The processing module includes:

[0170] The first processing unit is used to parse the insurance data and extract the image files, audio and video files and possible risk warning information.

[0171] The second processing unit is used to generate user interface data for displaying integrated review tasks through a unified rendering engine. The user interface data drives the front-end interface to load different modal data in parallel, realizing the integrated display of image thumbnails and audio and video players.

[0172] The determining module includes:

[0173] The first determining unit is used to match the corresponding structured rectification data object from the structured conclusion library according to the structured conclusion identifier; the structured conclusion library adopts a key-value pair storage structure, with the structured conclusion identifier as the key, and the structured rectification data object includes interface element positioning information, rectification operation instructions and associated file identifiers.

[0174] The update module includes:

[0175] The first update unit is used to parse the structured rectification data and extract the positioning information and rectification operation instructions.

[0176] The second update unit is used to drive the front-end interface to highlight the corresponding data elements and display structured rectification guidelines.

[0177] In one embodiment, the acquisition module further includes:

[0178] The third acquisition unit is used to acquire insurance data that also includes risk warning information generated by intelligent analysis of at least two types of insurance data, as described in step 220 of the embodiment.

[0179] In one embodiment, the acquisition module further includes:

[0180] The fourth acquisition unit is used to acquire risk warning information including at least one of the following: text verification prompts generated based on OCR recognition, semantic compliance prompts generated based on speech recognition, identity verification prompts generated based on face recognition, or operation compliance prompts generated based on white screen detection, as described in step 220 of the embodiment.

[0181] In one embodiment, at least two types of insurance information included in the insurance application data acquired by the acquisition module have undergone automated compliance verification; the automated compliance verification includes at least one of the following: OCR recognition verification, voice recognition verification, face recognition verification, or white screen detection, as described in step 320 of the embodiment.

[0182] In one embodiment, the structured rectification data generated by the determining module is generated by matching from the structured conclusion database based on the structured conclusion identifier, as described in step 340 of the embodiment.

[0183] In one embodiment, the structured rectification data generated by the determining module includes location information for locating a specific insurance information element in the user interface, and rectification operation guidance associated with that element, as described in step 340 of the embodiment.

[0184] It should be noted that, Figure 8 The core logic module structure of the device in this application is shown. Figures 1-7 The embodiments describe the specific implementation and interaction process of the device, and its functional modules can be connected with... Figure 8 The corresponding logical modules.

[0185] Understandable Figure 8 The device shown realizes integrated acquisition of insurance application data, structured review and decision-making, and intelligent rectification feedback from the perspective of functional modules. The collaborative work of the acquisition module 801, processing module 802, determination module 803, and update module 804 corresponds to all the steps of the method embodiment described in the first aspect above. For specific implementation details, please refer to... Figures 2-4 The method flow and corresponding steps are described.

[0186] Figure 9 This is a schematic diagram of an application scenario system according to an embodiment of this application.

[0187] This application proposes a system for integrated review of insurance application materials, including terminal equipment 900 and server equipment 800.

[0188] The terminal device 900 is equipped with an integrated insurance application document verification device as described in the second aspect embodiment, or is configured to perform the method described in any one of the first aspect embodiments. The terminal device 900 can be operated by insurance sales personnel or document verification personnel, and includes a user interface and a client program execution module.

[0189] The server-side device 800 is communicatively connected to the terminal device 900, and is used to provide the insurance data to the terminal device 900, and generate the structured rectification data by matching the structured conclusion identifier from the structured conclusion database. The server-side device 800 includes an automatic response server-side program execution module device.

[0190] The terminal device 900 described in this application may refer to a user equipment (UE) for mobile communication, a personal mobile terminal, a smart terminal, a mobile phone, a computer with communication functions, or an electronic system or computer system that provides services to the above-mentioned devices. It may also refer to any system, subsystem, module, circuit, chip, or software operating device that provides information reception, transmission, identification, and processing for the above-mentioned devices.

[0191] The server-side device 800 described in this application may refer to network facilities, network-side devices connected to a wireless communication system, or application servers connected to a communication system. It may also refer to an electronic system or computer system that provides services to the aforementioned devices, or any system, subsystem, module, circuit, chip, or software operating device that provides information reception, transmission, identification, and processing for the aforementioned devices.

[0192] The server-side device 800 can be a single device or multiple devices connected via a communication or information network. For example, in one embodiment, the server-side device 800 includes a first server 801 and a second server 802.

[0193] The first server 801 is used to support the interaction of integrated review tasks, and schedules a multimodal data analysis model to intelligently verify the insurance application materials and generate risk warning information. Specifically, the first server 801 receives first request data from the terminal device 900, retrieves and integrates at least two types of insurance application materials from multiple data sources such as an image library and a dual-recording video library according to the target insurance application identifier, calls the multimodal data analysis model to automatically verify the compliance of the insurance application materials, generates risk warning information, and sends first response data containing the insurance application materials and risk warning information to the terminal device 900.

[0194] The second server 802 is used to run the matching rules of the structured conclusion library and generate structured rectification data based on the structured conclusion identifier. Specifically, the second server 802 receives second request data from the terminal device 900, the second request data containing a structured conclusion identifier determined from a predefined conclusion set, matches and generates structured rectification data from the structured conclusion library based on the structured conclusion identifier, and sends second response data containing the structured rectification data to the terminal device 900.

[0195] The key technical modules for implementing the above-mentioned integrated review system are described below with reference to embodiments of this application:

[0196] Task aggregation and scheduling mechanism: The system maps image review tasks and dual recording quality inspection tasks into the same review task flow through a unified task scheduling module, and associates multiple types of materials based on the insurance application identifier to realize the aggregated execution of the task flow.

[0197] Multimodal analysis pipeline design: The multimodal analysis module adopts a parallel processing architecture, integrating OCR recognition submodule, speech recognition submodule, face recognition submodule and white screen detection submodule. Each submodule outputs structured risk labels and aggregates them into a unified risk warning dataset.

[0198] Structured Conclusion Library and Mapping Engine: The structured conclusion library adopts a key-value pair storage structure, where the key is the structured conclusion identifier and the value is the corresponding structured rectification opinion object, which includes interface element location information, rectification action instructions and associated file identifiers.

[0199] Cross-terminal structured data synchronization: The system encapsulates audit conclusions and rectification instructions into standard JSON format by defining a structured data synchronization protocol, which supports adaptation and rendering on the e-store terminal and the dual-recording App terminal, and achieves consistent display across terminals.

[0200] Risk visualization and interaction mechanism: The front-end interface uses the Canvas drawing component and timeline control to dynamically bind risk warnings with specific data elements (such as image areas and video segments), supporting user clicks for location and interactive rectification guidance.

[0201] Figure 10 This is a structural diagram of a server-side device according to an embodiment of this application. This application also proposes a server-side device for implementing the integrated review method for insurance application materials in any embodiment of this application. The server-side device is used to perform, for example... Figure 4 The functions described in steps 310-350 are a server or server cluster 800, which includes a server sending module 801, a server determining module 802, a server receiving module 803, a first database 804, a second database 805, and a third database 806 that are interconnected.

[0202] The server receiving module 803 is used to receive data from the terminal device, including: first request data (containing the target insurance policy identifier) ​​and second request data (containing the structured conclusion identifier determined from the predefined conclusion set), as described in steps 310 and 330 of the embodiment.

[0203] The server determination module 802 is used to perform logical judgments and information determination based on the received data. Specifically, it is used to: determine at least two types of insurance application materials associated with the target insurance application identifier; determine the risk warning information generated after automated compliance verification of the at least two types of insurance application materials; and, based on the received structured conclusion identifier, match and determine the corresponding structured rectification opinions from the second database 805, as described in steps 320 and 340 of the embodiment.

[0204] The server sending module 801 is used to send data to the terminal device, including: first response data (containing at least two types of insurance information associated with the target insurance policy identifier and optional risk warning information), and second response data (containing structured rectification data corresponding to the structured conclusion identifier), as described in steps 320 and 350 of the embodiment.

[0205] The first database 804 (insurance application data database) is used to store the original insurance application data associated with each insurance application identifier, including but not limited to structured insurance application information, image files (such as ID photos), and audio and video files (such as dual-recorded videos). The data stored therein forms the basis for the first response data.

[0206] The second database 805 (structured conclusion database) is used to store preset, standardized review conclusion rules and mapping relationships. Specifically, it includes: a predefined set of conclusions (i.e., each structured conclusion identifier), and a mapping relationship between each identifier and one or more structured rectification opinions. The structured rectification opinions include location information for locating specific insurance application data elements in the user interface, and rectification operation guidance associated with that element.

[0207] The third database 806 (intelligent analysis model library) is used to store and support the operation of multimodal data analysis models and rules. The models and rules in this library are used to perform automated compliance verification on the insurance application information retrieved from the first database 804. The verification includes at least one of OCR recognition verification, voice recognition verification, face recognition verification, and white screen detection, and generates the risk warning information accordingly.

[0208] The specific methods for implementing the server sending module 801, server determining module 802, server receiving module 803, and the functions of each database are as described in the various method embodiments of this application, and will not be repeated here.

[0209] Figure 11 This is a structural diagram of a terminal device according to an embodiment of this application. This application also proposes a terminal device for implementing the integrated insurance application data verification method described in any embodiment of this application. The terminal device is used to perform, for example... Figure 3 And the functions described in steps 210-260.

[0210] To implement the above technical solution, this application proposes a terminal device 900, which includes a terminal transmitting module 901, a terminal determining module 902, and a terminal receiving module 903 connected to each other.

[0211] The terminal sending module 901 is used to send data to the server device, including: first request data (containing the target insurance policy identifier) ​​and second request data (containing the structured conclusion identifier determined from the predefined conclusion set), as described in steps 210 and 250 of the embodiment.

[0212] The terminal receiving module 902 is used to receive a response dataset from the server-side device, including: first response data (containing at least two types of insurance information associated with the target insurance policy identifier and optional risk warning information), and second response data (containing structured rectification data corresponding to the structured conclusion identifier), as described in steps 220 and 260 of the embodiment.

[0213] The terminal determination module 902 is used to construct the interface and determine the conclusion based on the received data and user input, and performs the following functions:

[0214] The first response data is parsed to determine the interface rendering instructions in order to generate an integrated audit task user interface, as described in step 230 of the embodiment.

[0215] Based on the user interface, the user's review input is obtained, and the structured conclusion identifier corresponding to the review input data is determined (from a predefined conclusion set), as described in step 240 of the embodiment.

[0216] The specific methods for implementing the functions of the terminal sending module 901, the terminal receiving module 902, and the terminal determining module 903 are as described in the various method embodiments of this application, and will not be repeated here.

[0217] Based on the same inventive concept, this application also provides a graphical user interface for integrated review of insurance application materials. The interface runs on a terminal device and achieves integrated and closed-loop guidance of the review process through interaction with a server.

[0218] The graphical user interface includes at least a first display area, a second display area, and an interactive feedback area.

[0219] Upon receiving the first response data from the server, the interface performs the following operations:

[0220] In the first display area, at least one image file (such as an ID card or bank card image) is displayed in the form of a thumbnail list, and a risk warning label generated based on intelligent analysis is displayed next to each thumbnail. The label includes, but is not limited to, blurry text, inconsistent information, or mismatched face.

[0221] In the second display area, an audio and video player is embedded to play dual-recorded videos, and video segments with semantic compliance risks or operational compliance risks are marked on the player's timeline control with visual markers (such as color blocks and icons).

[0222] The first and second display areas are arranged side by side, allowing auditors to review multimodal materials and their associated risks simultaneously.

[0223] After receiving a structured conclusion identifier selected by the user from a predefined set of conclusions, the interface sends the identifier to the server.

[0224] Upon receiving the second response data from the server, the interface performs the following operations:

[0225] The structured rectification data contained in the second response data is parsed, and the data includes at least target element location information and rectification operation instructions;

[0226] Based on the target element location information, in the first display area or the second display area, the specific insurance information element to be rectified (such as a blurry ID card image or a risk video clip) is highlighted or focused.

[0227] In the interactive feedback area, an interactive rectification guidance window is dynamically generated and overlaid, which clearly lists the specific rectification operation steps associated with the positioning element.

[0228] Through the above graphical user interface design, the system achieves the associated and aggregated display of multimodal insurance application materials, the visual and accurate prompts of review risks, and the structured interactive guidance of rectification opinions. It transforms the traditional scattered text notifications into an integrated visual interaction process, which greatly improves the review efficiency, accuracy and user experience.

[0229] Based on the same inventive concept, this application also relates to key data structures transmitted and used in the integrated review process of insurance application materials, and data signals carrying such structures.

[0230] In the review process, the second response data generated by the server and sent to the terminal adopts a predefined structured data format. This data structure includes at least:

[0231] The conclusion identifier field contains the structured conclusion identifier uploaded by the terminal.

[0232] The data body field is rectified, and its value is a structured object, which further contains:

[0233] The element location subfield is used to store the unique identifier or coordinate information of the target insurance information element in the user interface;

[0234] The operation guidance subfield is used to store the description text or instruction code of the executable rectification action associated with the target element;

[0235] The associated file identifier subfield (optional) is used to specify the specific insurance application document file corresponding to the target element.

[0236] A data signal transmitted in an integrated insurance application document verification system is generated by a server and sent to a terminal. This signal carries the aforementioned data structure. After being transmitted via a communication network, the signal is received and parsed by the terminal and used to drive user interface updates to execute precise rectification guidance.

[0237] The structured data format is preferably JSON, XML, or Protocol Buffers. By using this semantically rich structured data and signals to replace unstructured natural language text, the audit conclusions and rectification instructions are made machine-readable, parsable, and automatically executable.

[0238] Figure 12 This is a schematic diagram of the structure of a server-side device according to another embodiment of this application. As shown in the figure, the server-side device 1000 includes a processor 1001, a communication interface 1002, and a memory 1003. The communication interface may consist of multiple components, including a transmitter and a receiver, providing a unit for communicating with various other devices over a transmission medium. The communication interface is used to implement communication functions with terminal devices and external systems, processing wireless signals through receiving and transmitting devices. The data carried by the signals is communicated with the memory or processor via an internal bus structure. The memory 1003 stores a computer program, which, when executed by the processor 1001, is used to implement the steps performed by the server in the integrated insurance application data verification method as described in any method embodiment of this application (such as...). Figure 4 (as described in steps 310-350). The processor 1001, communication interface 1002, and memory 1003 are interconnected through a bus system, which includes a data bus, power bus, control bus, and status signal bus, which will not be described in detail here.

[0239] Figure 13 This is a block diagram of a terminal device according to another embodiment of this application. The terminal device 1100 includes at least one processor 1101, a memory 1102, a user interface 1103, and at least one network interface 1104. The various components in the terminal device 1100 are coupled together through a bus system, which includes a data bus, a power bus, a control bus, and a status signal bus.

[0240] The user interface 1103 may include a display, keyboard, or clicking device, such as a mouse, trackball, touchpad, or touch screen, for running an integrated insurance application data review interface.

[0241] The memory 1102 stores executable modules or data structures. The memory can store an operating system and application programs. The operating system includes various system programs, such as a framework layer, core library layer, and driver layer, used to implement various basic business functions and handle hardware-based tasks. Application programs include various applications, such as media players and browsers, used to implement various application functions.

[0242] In this embodiment of the application, the memory 1102 stores a computer program, which, when executed by the processor 1101, is used to implement the steps performed by the terminal device in the integrated review method for insurance application information as described in any method embodiment of the application (such as...). Figure 3 (as described in steps 210-260). The network interface 1104 is used to enable communication with the server.

[0243] The memory 1102 includes a computer-readable storage medium. The processor 1101 reads the information in the memory 1102 and, in conjunction with its hardware, completes the steps of the above-described method. Specifically, the computer-readable storage medium stores a computer program, which, when executed by the processor 1101, implements the steps of the method embodiments described in any of the above embodiments.

[0244] The processor 1101 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the method described in this application can be completed by the integrated logic circuitry in the hardware of the processor 1101 or by instructions in software form. The processor 1101 may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, or discrete hardware component. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the embodiments of this application can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor.

[0245] Those skilled in the art will understand that embodiments of this application can be provided as methods or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. In a typical configuration, the device of this application includes one or more processors (CPUs), an input / output user interface, a network interface, and memory.

[0246] Furthermore, this application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0247] Therefore, this application also proposes a computer-readable medium storing a computer program that, when executed by a processor, implements the steps of the method described in any embodiment of this application. For example, in the foregoing embodiments, the memory 1003, 1102 may include non-permanent memory in the form of computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM.

[0248] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0249] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this application means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when an element is “connected” or “coupled” to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein may include wireless connections or wireless coupling. The term “and / or” as used herein includes all or any units and all combinations of one or more associated listed items.

[0250] Those skilled in the art will understand that, unless otherwise defined, all terms used herein (including technical, terminological, and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0251] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for integrated review of insurance application materials, characterized in that, Including the following steps: Obtain insurance application data; the insurance application data includes at least two types of insurance information associated with the target insurance policy identifier; The insurance data is parsed to generate user interface data for rendering the user application interface; Obtain the structured conclusion identifier determined from the predefined conclusion set from the user application interface; Based on the structured conclusion identifier, structured rectification data is generated by matching from the structured conclusion database; The user interface is updated based on the structured rectification data.

2. The integrated review method for insurance application materials according to claim 1, characterized in that, At least two types of the insurance application materials include image files and audio / video files.

3. The integrated review method for insurance application materials according to claim 1, characterized in that, The insurance data also includes risk warning information generated by intelligent analysis of at least two types of insurance information.

4. The integrated review method for insurance application materials according to claim 1, characterized in that, The insurance data also includes risk warning information; the risk warning information includes at least one of the following: Text verification prompts generated based on OCR recognition, semantic compliance prompts generated based on speech recognition, identity verification prompts generated based on face recognition, or operation compliance prompts generated based on white screen detection.

5. The integrated review method for insurance application materials according to claim 1, characterized in that, At least two types of insurance application materials included in the insurance application data have undergone automated compliance verification; The automated compliance verification includes at least one of the following: OCR recognition verification, speech recognition verification, face recognition verification, or white screen detection.

6. The integrated review method for insurance application materials according to claim 1, characterized in that, The structured rectification data is generated by matching the structured conclusions from the structured conclusions database based on the structured conclusions identifier.

7. The integrated review method for insurance application materials according to claim 1, characterized in that, The structured rectification data includes location information for locating specific insurance information elements in the user interface, as well as rectification operation guidance associated with that element.

8. An integrated insurance application document verification device, used to implement the integrated insurance application document verification method according to any one of claims 1-7, characterized in that, Include: The acquisition module is used to acquire insurance data; the insurance data includes at least two types of insurance information associated with the target insurance policy identifier; it is also used to acquire a structured conclusion identifier determined from a predefined conclusion set from the user application interface; The processing module is used to parse the insurance data and generate user interface data for rendering the user application interface; The determination module is used to match and generate structured rectification data from the structured conclusion database based on the structured conclusion identifier. An update module is used to update the user interface based on the structured rectification data.

9. An integrated verification system for insurance application materials, characterized in that, include: The terminal device is equipped with the integrated insurance application data verification device as described in claim 8, or is configured to execute the method described in any one of claims 1 to 7; The server is communicatively connected to the terminal device and is used to provide the insurance data to the terminal device and generate the structured rectification data by matching the structured conclusion database according to the structured conclusion identifier.

10. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method as described in any one of claims 1-7.

Images