Method, device and electronic equipment for handling insurance exclusion clause

Abstract

The application discloses a processing method and device of insurance exclusion clauses and electronic equipment. The method comprises the following steps: pre-processing an insurance contract text to obtain a candidate text segment including an exclusion clause; identifying the exclusion clause in the candidate text segment, and generating a preliminary merged text according to the exclusion clause; performing compliance verification and readability optimization on the preliminary merged text, and generating a spoken explanation script and a synchronous subtitle text; and pushing the spoken explanation script and the synchronous subtitle text to a double-recording device to realize real-time synchronization of the spoken explanation script, the synchronous subtitle text and the double-recording function. The application can avoid problems such as missing key exclusion elements and inaccurate expression, is convenient for policyholders to understand, reduces compliance risks, does not depend on the professional ability of sales personnel, and improves the standardization degree.

Description

Technical Field

[0001] This application belongs to the field of insurance technology, and in particular relates to a method, apparatus and electronic device for processing insurance exclusion clauses. Background Technology

[0002] Currently, the handling of exclusion clauses in insurance dual recording scenarios mainly relies on manual processing: before dual recording, sales personnel manually review the insurance contract text, filter out the exclusion clauses scattered in chapters such as "Exclusions of Liability", "Special Provisions", and "Term Notes", manually judge the content to remove duplicates and organize it, and then convert it into conversational expression to explain to the policyholder.

[0003] However, the above schemes rely on manual screening and merging, which is inefficient and the merging effect is not ideal. It is easy to cause problems such as human error in omitting key exclusion elements and inaccurate descriptions, which may lead to difficulties or misunderstandings for policyholders. At the same time, the compliance risk is high. It also relies on the professional ability of sales personnel and has a low degree of standardization. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, and electronic device for processing insurance exclusion clauses, in order to solve the problems of low efficiency and unsatisfactory merging effect in related technologies, easy to omit key exclusion elements, inaccurate expression, and low standardization.

[0005] To achieve the above objectives, the embodiments of this application adopt the following technical solutions: In a first aspect, embodiments of this application provide a method for processing insurance exclusion clauses, comprising: preprocessing an insurance contract text to obtain candidate text fragments including exclusion clauses; identifying the exclusion clauses in the candidate text fragments and generating a preliminary merged text based on the exclusion clauses; performing compliance verification and readability optimization on the preliminary merged text and generating a spoken explanation script and synchronized subtitle text; and pushing the spoken explanation script and the synchronized subtitle text to a dual-recording device to achieve real-time synchronization of the spoken explanation script, the synchronized subtitle text, and the dual-recording function.

[0006] Secondly, embodiments of this application provide a processing apparatus for insurance exclusion clauses, comprising: a preprocessing module for preprocessing insurance contract text to obtain candidate text fragments including exclusion clauses; an identification module for identifying the exclusion clauses in the candidate text fragments and generating preliminary merged text based on the exclusion clauses; an optimization module for performing compliance verification and readability optimization on the preliminary merged text and generating a spoken explanation script and synchronized subtitle text; and a push module for pushing the spoken explanation script and synchronized subtitle text to a dual recording device to achieve real-time synchronization of the spoken explanation script, the synchronized subtitle text, and the dual recording function.

[0007] Thirdly, embodiments of this application provide an electronic device, including: a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0008] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: In this embodiment, the exclusion clauses in the insurance contract text can be automatically identified and automatically merged to obtain a preliminary merged text. Combined with compliance verification and readability optimization, it can avoid problems such as omission of key exclusion elements and inaccurate expression, making it easier for policyholders to understand. At the same time, it reduces compliance risks and does not rely on the professional ability of sales personnel, thus improving the degree of standardization. Attached Figure Description

[0009] 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: Figure 1 A flowchart illustrating a method for processing insurance exclusion clauses, provided as an embodiment of this application; Figure 2 A schematic diagram of the structure of an insurance exclusion clause processing device provided for one embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0010] 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.

[0011] The terms "first," "second," etc., used in this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein. Furthermore, "and / or" in this application indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship. It should be noted that all data involved in this application was obtained with the user's authorization.

[0012] The technical solutions provided in the embodiments of this application are described in detail below with reference to the accompanying drawings.

[0013] Figure 1 This is a flowchart illustrating a method for handling insurance exclusion clauses, provided as an embodiment of this application. Figure 1 As shown, the method for handling insurance exclusion clauses in this application embodiment may specifically include the following steps: S101, preprocess the insurance contract text to obtain candidate text fragments including exemption clauses.

[0014] In this embodiment of the application, the executing entity of the method for processing insurance exclusion clauses is an insurance exclusion clause processing device, which can be located in an electronic device. This electronic device can be a terminal device or a server. The terminal device can be a mobile phone, tablet computer, desktop computer, laptop, in-vehicle device, etc.; the server can be a standalone server or a server cluster composed of multiple servers. For example, the insurance exclusion clause processing device can be located in an insurance business platform.

[0015] Users can input insurance contract texts in various formats such as PDF, Word, and scanned copies. For insurance contract texts that are not plain text, the plain text content can be extracted through OCR recognition and text parsing, with a recognition accuracy of ≥99%.

[0016] The plain text insurance contract text is preprocessed to obtain at least one candidate text fragment that may include an exemption clause.

[0017] As a feasible implementation method, step S101, "preprocessing the insurance contract text to obtain candidate text fragments including disclaimers," may specifically include the following steps: cleaning the insurance contract text to remove redundant information; splitting the cleaned insurance contract text into chapters to obtain multiple split text fragments; and selecting candidate text fragments from the multiple split text fragments based on preset disclaimer keywords.

[0018] Specifically, the insurance contract text undergoes format cleaning: redundant information such as headers, footers, table symbols, and advertising statements are removed, retaining only the core content of the main body of the insurance contract text. The format-cleaned insurance contract text is then split into chapters: based on the general structure of insurance contract texts (such as "Insurance Liability," "Exclusions," "Special Provisions," "Explanations," etc.), it is split using keyword matching (such as "Chapter Two," "I. Exclusions," etc.), resulting in multiple fragmented text segments. Using an insurance industry-specific keyword database (including over 200 core exclusion keywords such as "exclusion," "excluding," "non-payment," "limitation," "not liable"), potential candidate text segments containing exclusion clauses are identified, with a screening accuracy rate of ≥98%.

[0019] S102, identify the disclaimers in the candidate text fragments and generate a preliminary merged text based on the disclaimers.

[0020] In this embodiment, a large model can be used to identify all disclaimers in candidate text fragments. These disclaimers include explicit disclaimers and implicit limitations of liability. Specifically, the candidate text fragments are input into the large model, which automatically extracts all disclaimers from them with an accuracy rate of ≥97%.

[0021] The large-scale model can be a finely tuned model based on fundamental models such as LLaMA / GPT, using insurance industry corpora. For example, it can be finely tuned using 100,000 insurance contract texts (covering all types of insurance, including critical illness insurance, medical insurance, and life insurance), 500 insurance regulatory policy documents (2018-2024), and 5,000 dual-recording compliance cases to form a large-scale model specifically for the insurance industry, with a semantic understanding accuracy of ≥98.5%.

[0022] Furthermore, the "generating a preliminary merged text based on the disclaimer clauses" in step S102 above may specifically include the following steps: extracting structured disclaimer elements based on the disclaimer clauses; merging the disclaimer clauses based on the dual recording rule base and the disclaimer elements to obtain a preliminary merged text.

[0023] Specifically, the system automatically extracts the core structured exclusion elements from each exclusion clause, which may include, but are not limited to, the following elements: exclusion circumstances (e.g., "the insured committed an intentional crime"), applicable conditions (e.g., "within 2 years after the contract is established"), liability limits (e.g., "no full compensation"), legal basis (e.g., Article 45 of the Insurance Law), and related insurance liabilities (e.g., "critical illness insurance benefit"). The element extraction accuracy rate is ≥97.2%. The structured exclusion elements can also be stored in a database in the format of "clause ID-element type-element content-original text location" for easy subsequent tracing and verification.

[0024] The dual recording rule base is a rule base applicable to dual recording scenarios. The dual recording rule base can be built into three core rule subsets: deduplication rules, conflict reconciliation rules, and hierarchical reorganization rules, and supports dynamic updates (for example, new rules can be manually uploaded or automatically iterated by connecting to the policy platform of the State Financial Regulatory Commission).

[0025] Correspondingly, the above step "merging the disclaimer clauses according to the dual recording rule base and disclaimer elements to obtain a preliminary merged text" may specifically include the following steps: according to the deduplication rules, conflict reconciliation rules and hierarchical reorganization rules in the dual recording rule base, the disclaimer clauses are sequentially deduplicated, conflict reconciliation and hierarchical reorganization to obtain a preliminary merged text.

[0026] Specifically, the deduplication rule is as follows: Based on the insurance semantic similarity algorithm (which integrates text similarity and element consistency verification), disclaimers with a similarity of ≥85% and identical disclaimer elements are judged as duplicates, and only one disclaimer is retained.

[0027] Conflict resolution rules: Following the three principles of "regulatory priority (latest policies take precedence over old policies), stricter application (stricter liability limitation clauses take precedence), and legal basis priority (clauses that explicitly cite legal provisions take precedence)," conflicting exemption clauses are screened.

[0028] Hierarchical restructuring rules: The exclusion clauses are sorted according to the logic of "general exclusions (applicable to all insurance liabilities) → specific exclusions (for a single insurance liability) → exceptions (exceptions to the exclusion clauses)" and adapted to the dual recording explanation process.

[0029] The previously extracted structured disclaimer elements are input into the large model, and the dual-recording rule base is called to perform a three-step process of "deduplication → conflict reconciliation → hierarchical reorganization" to generate a preliminary merged text, ensuring that the number of disclaimer clauses after merging is reduced by 40%-60% compared to the original disclaimer clauses.

[0030] S103, perform compliance verification and readability optimization on the initially merged text, and generate spoken narration scripts and synchronized subtitle text.

[0031] In this application embodiment, compliance verification may include, but is not limited to, full-coverage verification of elements and verification of regulatory policies.

[0032] Element full coverage verification: Based on the previously extracted structured disclaimer elements, the preliminary merged text can be fully verified for element coverage, and any missing disclaimer elements can be added to the preliminary merged text. Specifically, the preliminary merged text can be automatically compared with the original structured disclaimer elements through the large model to check whether any disclaimer circumstances, applicable conditions, or other disclaimer elements are missing. If any omissions are found, the missing disclaimer elements will be automatically added back.

[0033] Regulatory policy verification: The preliminary merged text can be verified against the pre-set regulatory policy library. Specifically, the built-in regulatory policy library (which may include the "Interim Measures for Retrospective Management of Insurance Sales Activities" and the "Measures for the Management of Product Suitability of Financial Institutions") can be called to verify whether the wording of the preliminary merged text complies with the "obligation to clearly explain" requirement and avoid compliance risks.

[0034] Readability optimization can include, but is not limited to, conversational language transformation and logical coherence optimization.

[0035] Conversationalization: Transform the legal terminology in the initial merged text into colloquial expressions (e.g., transform "the insured intentionally commits a crime" into "the insured intentionally violates the law and commits a crime"), while retaining the core rigor and keeping the sentence length within the target word count range (e.g., 10-15 words / sentence) to match the rhythm of spoken explanation.

[0036] Logical connection optimization: Add transitional phrases to the initial merged text, such as "First, let's explain the general disclaimers" and "Next, we'll discuss disclaimers for specific scenarios," to improve the fluency of the explanation.

[0037] The "generating spoken language explanation script" in step S103 above may specifically include the following steps: splitting and optimizing the sentences in the preliminary merged text according to the explanation logic, and marking the pause nodes (such as ";" corresponding to a 0.5-second pause) to generate a spoken language explanation script that can be used directly.

[0038] The "generating synchronized subtitle text" in step S103 above may include the following steps: exporting the optimized preliminary merged text according to the dual-recording video frame rate (e.g., 25fps or 30fps) (supporting export in formats such as SRT and VTT) to obtain synchronized subtitle text, ensuring that the subtitles are synchronized with the spoken narration in real time.

[0039] The method for processing insurance exclusion clauses in this application embodiment can also output a compliance filing text: retain the merged written version of the exclusion clauses (i.e., the preliminary merged text) for regulatory verification and contract filing.

[0040] S104 pushes the spoken language explanation script and synchronized subtitle text to the dual recording device to achieve real-time synchronization of the spoken language explanation script, synchronized subtitle text, and dual recording function.

[0041] In this embodiment of the application, the method for processing insurance exclusion clauses also supports device linkage: pushing the spoken explanation script and synchronized subtitle text to the dual recording device to achieve real-time synchronization of the spoken explanation script, synchronized subtitle text and dual recording function.

[0042] In addition, the method for handling insurance exclusion clauses in this application embodiment also supports quality inspection linkage: it connects with the dual recording quality inspection system to automatically verify the completeness of the oral explanation script (i.e., whether it covers all merged clauses) and the compliance of the expression.

[0043] In addition, the method for processing insurance exclusion clauses in this application also supports evidence storage linkage: the preliminary merged text, oral explanation script, dual audio and video recordings, verification results and unique identifiers (device ID + product ID + timestamp) are bound and synchronized to the blockchain evidence storage platform to ensure that the data is tamper-proof and meet the regulatory requirement that "the retention period shall not be less than 10 years after the termination of the contract".

[0044] This application's embodiment achieves a unified approach to "compliance, efficiency, and adaptability" through a full-chain design that includes "industry-specific fine-tuning of the large model to ensure extraction accuracy → dedicated rule base to ensure merging compliance → scenario adaptation optimization to ensure ease of use → closed-loop linkage to ensure traceability security." Logically, it forms a complete closed loop of "input-processing-output-verification-evidence storage."

[0045] In summary, the method for processing insurance exclusion clauses in this application can automatically identify exclusion clauses in the insurance contract text and automatically merge them into a preliminary merged text. Combined with compliance verification and readability optimization, it can avoid problems such as omission of key exclusion elements and inaccurate expression, making it easier for policyholders to understand, reducing compliance risks, and improving standardization by not relying on the professional capabilities of sales personnel. Preprocessing of the insurance contract text, including format cleaning, chapter splitting, and candidate text fragment screening, improves the efficiency of subsequent identification of exclusion clauses. Structured extraction ensures information integrity. Exclusion clauses are sequentially deduplicated, conflict-reconciling, and hierarchical reorganization according to the dual-recording rule base, shortening the clause length while ensuring no omission of key compliance information. Readability optimization of the preliminary merged text improves the fluency of explanation. The merged written clauses are converted into a spoken explanation script and synchronized subtitle text adapted to dual-recording scenarios, balancing "legal rigor" and "audio / visual adaptability," making it suitable for dual-recording scenarios. The integration of the merged results with dual-recording equipment, quality inspection system, and blockchain evidence storage platform ensures traceability security.

[0046] This application also provides a device for processing insurance exclusion clauses. For example... Figure 2 As shown, the insurance exclusion clause processing device 200 of this application embodiment may specifically include: a preprocessing module 201, an identification module 202, an optimization module 203, and a push module 204. Wherein: The preprocessing module 201 is used to preprocess the insurance contract text to obtain candidate text fragments including exemption clauses.

[0047] The identification module 202 is used to identify disclaimers in candidate text fragments and generate preliminary merged text based on the disclaimers.

[0048] The optimization module 203 is used to perform compliance verification and readability optimization on the initially merged text, and generate spoken narration scripts and synchronized subtitle text.

[0049] The push module 204 is used to push the spoken language explanation script and synchronized subtitle text to the dual recording device to achieve real-time synchronization of the spoken language explanation script, synchronized subtitle text and dual recording function.

[0050] In the embodiments of this application, the specific process by which each module and unit implements its function can be found in the relevant description of the above-mentioned method for handling any of the insurance exemption clauses, and will not be repeated here.

[0051] The insurance exclusion clause processing device in this application embodiment can automatically identify exclusion clauses in the insurance contract text and automatically merge them into a preliminary merged text. Combined with compliance verification and readability optimization, it avoids problems such as missing key exclusion elements and inaccurate wording, making it easier for policyholders to understand, reducing compliance risks, and improving standardization by not relying on the professional capabilities of sales personnel. Preprocessing of the insurance contract text, including format cleaning, chapter splitting, and candidate text fragment screening, improves the efficiency of subsequent exclusion clause identification. Structured extraction ensures information integrity. Exclusion clauses are sequentially deduplicated, conflict-reconciling, and hierarchically reorganized according to the dual-recording rule base, shortening the clause length while ensuring no key compliance information is omitted. Readability optimization of the preliminary merged text improves the fluency of explanation. The merged written clauses are converted into a spoken explanation script and synchronized subtitle text adapted to dual-recording scenarios, balancing "legal rigor" and "auditory / visual adaptability," making it suitable for dual-recording scenarios. Linkage between the merged result and dual-recording equipment, quality inspection system, and blockchain evidence storage platform ensures traceability security.

[0052] This application also provides an electronic device. For example... Figure 3As shown, the electronic device 300 can vary considerably depending on its configuration or performance. It may include one or more processors 301 and memory 302, with memory 302 storing one or more programs or instructions. Memory 302 may be temporary or permanent storage. The program stored in memory 302 may include one or more modules (not shown), each module including a series of computer-executable instructions for the electronic device 300. Furthermore, processor 301 may be configured to communicate with memory 302, executing the series of programs or computer-executable instructions stored in memory 302 on the electronic device 300. The electronic device 300 may also include one or more power supplies 303, one or more wired or wireless network interfaces 304, one or more input / output interfaces 305, and one or more keyboards 306.

[0053] Specifically, in the embodiments of this application, the electronic device includes a processor, a memory, and a program or instruction stored in the memory and executable on the processor. When the program or instruction is executed by the processor, it implements the steps of the processing method embodiment for any of the above-mentioned insurance exemption clauses.

[0054] The electronic device in this application embodiment can automatically identify the exclusion clauses in insurance contract text and automatically merge them into a preliminary merged text. Combined with compliance verification and readability optimization, it avoids problems such as missing key exclusion elements and inaccurate wording, making it easier for policyholders to understand, reducing compliance risks, and improving standardization by not relying on the professional capabilities of sales personnel. Preprocessing of the insurance contract text, including format cleaning, chapter splitting, and candidate text fragment screening, improves the efficiency of subsequent exclusion clause identification. Structured extraction ensures information integrity. Exclusion clauses are sequentially deduplicated, conflict-reconciling, and hierarchically reorganized according to the dual-recording rule base, shortening the clause length while ensuring no key compliance information is omitted. Readability optimization of the preliminary merged text improves the fluency of explanation. The merged written clauses are converted into a spoken explanation script and synchronized subtitle text adapted for dual-recording scenarios, balancing "legal rigor" and "auditory / visual adaptability," making it suitable for dual-recording scenarios. Linkage between the merged result and the dual-recording device, quality inspection system, and blockchain evidence storage platform ensures traceability security.

[0055] This application also proposes a readable storage medium storing one or more computer programs or instructions that, when executed by a processor in an electronic device, enable the processor in the electronic device to perform the steps of any of the above-described embodiments of the method for processing insurance exclusion clauses.

[0056] The readable storage medium of this application embodiment can automatically identify the exclusion clauses in the insurance contract text and automatically merge them into a preliminary merged text. Combined with compliance verification and readability optimization, it avoids problems such as missing key exclusion elements and inaccurate wording, making it easier for policyholders to understand, reducing compliance risks, and improving standardization by not relying on the professional capabilities of sales personnel. Preprocessing of the insurance contract text, including format cleaning, chapter splitting, and candidate text fragment screening, improves the efficiency of subsequent exclusion clause identification. Structured extraction ensures information integrity. Exclusion clauses are sequentially deduplicated, conflict-reconciling, and hierarchically reorganized according to the dual-recording rule base, shortening the clause length while ensuring no key compliance information is omitted. Readability optimization of the preliminary merged text improves the fluency of explanation. The merged written clauses are converted into a spoken explanation script and synchronized subtitle text adapted to dual-recording scenarios, balancing "legal rigor" and "auditory / visual adaptability," making it suitable for dual-recording scenarios. Linkage between the merged result and dual-recording equipment, quality inspection system, and blockchain evidence storage platform ensures traceability security.

[0057] It should be understood that the training and prediction processes of the AI ​​models involved in the various embodiments of this specification all adhere to multiple legal and compliant principles, including legal data sources, compliant data content, compliant data governance, compliant training objectives and schemes, compliant training processes, compliant training environments and tools, and compliant ethical verification of training results, and comply with the requirements of Article 5 of the Patent Law. Among them: Data source legitimacy: All datasets used for AI model training were obtained through legal means, covering three categories: publicly authorized data, data authorized by partners, and self-collected compliant data. Publicly authorized data comes from compliant data sources following open-source licenses such as Apache 2.0, with complete copyright attribution and authorization scope clearly marked, and no unauthorized open-source code or data reuse. Data authorized by partners has been subject to formal data usage agreements, clearly defining the scope, duration, and confidentiality obligations, and possessing a complete authorization chain. For self-collected data involving personal information, strict informed consent procedures have been followed, and anonymization processes (including but not limited to field masking, feature anonymization, and differential privacy technology applications) have been implemented to remove personally identifiable information, fully complying with the requirements of relevant laws and regulations such as the "Interim Measures for the Administration of Generative Artificial Intelligence Services" and the "Personal Information Protection Law."

[0058] Data content compliance: The AI ​​model's dataset undergoes multiple screenings and cleaning processes to remove all content that may violate social morality or harm public interests. It contains no obscene, pornographic, violent, discriminatory, or information that endangers national or public safety, nor does it involve the illegal acquisition or use of genetic resources. For data in sensitive fields (such as healthcare and finance), an additional privacy-preserving computation module (including federated learning and secure multi-party computation technologies) ensures that the data is "usable but not visible," avoiding compliance risks during the original data transmission process and ensuring that the data application scenarios and uses comply with public order and good morals and industry regulatory requirements.

[0059] Data governance norms: A complete data traceability system is established during the AI ​​model training process to automatically record the source, collection time, annotation process, cleaning rules, and permission allocation of training data, generating traceable compliance reports to ensure that the data is verifiable throughout its entire lifecycle. The dataset annotation process for AI models is completed by a professional human R&D team, clearly defining the proportion of human creative contributions and avoiding reliance on AI-generated data that has not undergone substantial human modification, thus meeting the examination requirements for "human main contributions" in AI patent applications.

[0060] Training objectives and plans are compliant: The AI ​​model training objective focuses on identifying disclaimers, and the training scheme and final output results do not violate any mandatory provisions of laws or administrative regulations, do not harm the public interest or the legitimate rights and interests of others, and do not pose any potential risk of being used for illegal activities, privacy infringement, or public safety disruption. It strictly adheres to the ethical principle of "intelligent for good".

[0061] Training process compliance: A closed-loop training framework is adopted to ensure compliance and controllability of the training process. The specific process is as follows: First, training samples are obtained through compliant data sources. After the aforementioned data cleaning and desensitization, they are input into the neural network model to generate preliminary training results. Second, an expert system is introduced to verify the preliminary results. Based on preset rules and human expert experience, the feasibility of the results is evaluated, and outputs that may pose ethical risks or compliance hazards are corrected (such as removing decision-making logic that violates public order and good morals, and adjusting model parameters that do not comply with safety regulations). Finally, the loss function weights are dynamically optimized based on expert system feedback to strengthen the model's learning of compliant results, avoid overfitting errors or non-compliant labels, and form a closed-loop control of "data input - model training - expert verification - parameter optimization - result feedback" to ensure that the entire training process complies with A5 ethical review requirements.

[0062] Training environment and tool compliance: AI model training is implemented using nationally licensed chips and a compliant training platform. All open-source frameworks and components used in the training process have obtained their corresponding licenses, and copyright statements and patent citation information are fully retained, with no instances of infringement or reuse. The training environment is built using virtual devices (containers / virtual machines) with fixed random seeds and initial parameter configurations to ensure the reproducibility of the training process. Furthermore, through access control and operation log recording, risks such as data leakage and parameter tampering during training are prevented, ensuring the security and compliance of the training process.

[0063] Training results ethical verification compliance: After the model is trained, it undergoes additional third-party ethical compliance assessment and algorithm filing review to verify that the model output does not violate social morality or harm public interests. For potentially sensitive scenarios (such as public services and intelligent decision-making), a special result verification mechanism is established to ensure that the model always complies with Article 5 of the Patent Law and relevant laws and regulations in practical applications.

[0064] In summary, the data and training process used in the AI ​​model of this specification strictly comply with the relevant provisions of Article 5 of the Patent Law and the Patent Examination Guidelines (2023 Edition), and there are no violations of laws, social ethics, public interests, or illegal use of genetic resources. It fully meets the compliance requirements for patent authorization.

[0065] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0066] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0067] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, 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. Furthermore, this application can take the form of a computer program product embodied 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.

[0068] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0069] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0070] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0071] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0072] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0073] 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. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0074] This application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0075] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the apparatus embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0076] The above are merely embodiments of this application and are 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 handling insurance exclusion clauses, characterized in that, include: The insurance contract text is preprocessed to obtain candidate text fragments including exclusion clauses; Identify the disclaimers in the candidate text fragments and generate a preliminary merged text based on the disclaimers; The preliminary merged text is subjected to compliance verification and readability optimization, and a spoken narration script and synchronized subtitle text are generated; The spoken language explanation script and the synchronized subtitle text are pushed to the dual recording device to achieve real-time synchronization of the spoken language explanation script, the synchronized subtitle text, and the dual recording function.

2. The method according to claim 1, characterized in that, The preprocessing of the insurance contract text yields candidate text fragments including exclusion clauses, including: The insurance contract text is cleaned to remove redundant information. The cleaned insurance contract text is split into chapters to obtain multiple text fragments. Candidate text segments are selected from the multiple split text segments based on preset disclaimer keywords.

3. The method according to claim 1, characterized in that, The identification of the disclaimer in the candidate text fragment includes: The large model is used to identify the disclaimers in the candidate text fragments, including explicit disclaimers and implicit limitation of liability clauses.

4. The method according to claim 1, characterized in that, The generation of the preliminary merged text based on the aforementioned disclaimer includes: Extract the structured elements of the disclaimer based on the aforementioned disclaimer clauses; The disclaimer clauses are merged based on the dual recording rule base and the disclaimer elements to obtain the preliminary merged text.

5. The method according to claim 4, characterized in that, The preliminary merged text is obtained by merging the disclaimer clauses according to the dual-recording rule base and the disclaimer elements, including: Based on the deduplication rules, conflict reconciliation rules, and hierarchical reorganization rules in the dual-recording rule base, the disclaimer clauses are sequentially deduplicated, conflict reconciled, and hierarchically reorganized to obtain the preliminary merged text.

6. The method according to claim 4, characterized in that, The compliance verification of the initially merged text includes: Based on the aforementioned disclaimer elements, a full coverage verification of the elements is performed on the preliminary merged text, and any missing disclaimer elements are added to the preliminary merged text. The preliminary merged text is validated against regulatory policies based on a pre-defined regulatory policy database.

7. The method according to claim 1, characterized in that, The readability optimization of the initially merged text includes: The preliminary merged text is then transformed into colloquial language: legal terminology in the preliminary merged text is converted into plain expressions, and sentence length is controlled within the target word count range; and / or, Logical connection optimization is performed on the preliminary merged text: transitional expressions are added to the preliminary merged text.

8. The method according to claim 1, characterized in that, The generated spoken narration script and synchronized subtitle text include: Sentences in the preliminary merged text are split and optimized according to the explanation logic, and pause nodes are marked to generate the oral explanation script; The optimized preliminary merged text is exported according to the frame rate of the dual-recording video to obtain the synchronized subtitle text.

9. A device for processing insurance exclusion clauses, characterized in that, include: The preprocessing module is used to preprocess the insurance contract text to obtain candidate text fragments including exemption clauses; The identification module is used to identify the disclaimer clauses in the candidate text fragments and generate preliminary merged text based on the disclaimer clauses; The optimization module is used to perform compliance verification and readability optimization on the initially merged text, and generate spoken explanation scripts and synchronized subtitle text; The push module is used to push the spoken language explanation script and the synchronized subtitle text to the dual recording device to achieve real-time synchronization of the spoken language explanation script, the synchronized subtitle text and the dual recording function.

10. An electronic device, characterized in that, include: A processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the method as claimed in any one of claims 1-8.

Images