Insurance business knowledge base intelligent optimization method, system, medium and electronic equipment

Abstract

The application provides an insurance business knowledge base intelligent optimization method, system, medium and electronic equipment. The insurance business knowledge base intelligent optimization method comprises: acquiring user question data; using a business model to answer the user question data to obtain a feedback signal, the feedback signal comprising user feedback and business expert feedback; based on the feedback signal, diagnosing the answer information to generate an optimization strategy; using the optimization strategy to dynamically optimize the business model to obtain an optimization execution result; verifying and feeding back the optimization execution result to obtain an insurance business knowledge base intelligent optimization result. The insurance business knowledge base intelligent optimization method of the application can realize intelligent and autonomous evolution of the insurance knowledge base.

Description

Technical Field

[0001] This application belongs to the field of artificial intelligence technology and relates to the optimization of insurance knowledge base, particularly to an intelligent tuning method, system, medium, and electronic device for insurance business knowledge base. Background Technology

[0002] Intelligent optimization of insurance knowledge bases refers to the automated and intelligent continuous optimization and updating of professional knowledge bases in the insurance field using artificial intelligence technologies, particularly advanced methods such as natural language processing, machine learning, and knowledge graphs. This optimization not only focuses on the accuracy of knowledge content but also emphasizes the rationality of the knowledge structure, its adaptability to application scenarios, and its alignment with business needs, thereby achieving dynamic evolution and maximizing the value of the knowledge base. However, current intelligent optimization of insurance knowledge bases still relies on human intervention, and there are transformation barriers between insurance business and technology. Summary of the Invention

[0003] The purpose of this application is to provide a method, system, medium, and electronic device for intelligent optimization of an insurance business knowledge base, which enables the intelligent and autonomous evolution of the insurance knowledge base.

[0004] Firstly, this application provides an intelligent optimization method for an insurance business knowledge base, comprising: acquiring user question data; responding to the user question data using a business model to obtain feedback signals, the feedback signals including user feedback and business expert feedback; performing problem diagnosis on the response information based on the feedback signals to generate an optimization strategy; dynamically optimizing the business model using the optimization strategy to obtain optimization execution results; and verifying and providing feedback on the optimization execution results to obtain intelligent optimization results for the insurance business.

[0005] In one implementation of the first aspect, the process of responding to the user question data using a business model to obtain a feedback signal includes: performing semantic analysis on the user question data to obtain a semantic analysis result; processing the semantic analysis result based on a five-dimensional dynamic data stream to obtain the response information; and obtaining the feedback signal based on the response information.

[0006] In one implementation of the first aspect, the process of diagnosing problems in response information based on the feedback signal to generate an optimization strategy includes: identifying problems in the response information based on the feedback signal to obtain business problems; and automatically generating the optimization strategy based on the business problems.

[0007] In one implementation of the first aspect, the process of identifying problems in the response information based on the feedback signal to obtain business problems includes: performing in-depth analysis of the feedback signal using natural language processing based on a five-dimensional dynamic data stream to obtain reflection results; and identifying problems in the response information based on the reflection results to obtain the business problems.

[0008] In one implementation of the first aspect, the intelligent optimization method for the insurance business knowledge base further includes: merging a historical case library with an implementation regulatory version tree to obtain a dynamic business perception matrix; matching the business problem with the dynamic business perception matrix; and using a liability boundary topology algorithm to decompose the complex logic in the dynamic business perception matrix to obtain the optimization strategy.

[0009] In one implementation of the first aspect, the process of diagnosing problems with the response information based on the feedback signal to generate an optimization strategy further includes: obtaining a business problem description from the business expert based on the feedback; extracting key semantic information from the business problem description to obtain business language; establishing a correspondence between the business language and the business model using a terminology mapping library; and performing deep semantic parsing on the business description to obtain technical optimization instructions.

[0010] In one implementation of the first aspect, the intelligent optimization method for the insurance business knowledge base further includes: obtaining a dual-track collaborative business processing flow; constructing model suggestions based on the dual-track collaborative business processing model to obtain an enhanced collaborative iterative business processing model; and constructing a model self-reflection based on the enhanced collaborative iterative business processing model to obtain an autonomously evolving intelligent business processing model.

[0011] Secondly, this application provides an intelligent optimization system for an insurance business knowledge base, comprising: a user question data acquisition module for acquiring user question data; a feedback signal acquisition module for responding to the user question data using a business model to obtain feedback signals, the feedback signals including user feedback and business expert feedback; an optimization strategy generation module for performing problem diagnosis on the response information based on the feedback signals to generate an optimization strategy; an optimization execution result acquisition module for dynamically optimizing the business model using the optimization strategy to obtain optimization execution results; and a verification feedback module for verifying and providing feedback on the optimization execution results to obtain intelligent optimization results for the insurance business knowledge base.

[0012] Thirdly, this application provides an electronic device, which includes: a memory storing a computer program thereon; and a processor communicatively connected to the memory for executing the computer program to implement the above-mentioned intelligent optimization method for the insurance business knowledge base.

[0013] Fourthly, this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by an electronic device, implements the above-mentioned intelligent optimization method for the insurance business knowledge base.

[0014] As described above, the intelligent optimization method, system, medium, and electronic equipment for insurance business knowledge bases described in this application have the following beneficial effects:

[0015] Based on feedback signals, problem diagnosis is performed on the response information to generate optimization strategies. The optimization strategies are then used to dynamically optimize the business model, and the optimization results are verified and fed back. Through autonomous diagnosis and feedback updates based on feedback signals, an automated closed-loop optimization mechanism is formed. By establishing a continuous learning framework, the timeliness of business rule iteration in the insurance business knowledge base is ensured, and expert resources are effectively freed up, resulting in an exponential improvement in question-and-answer accuracy. Attached Figure Description

[0016] Figure 1A The diagram shows an application scenario of the intelligent optimization method for the insurance business knowledge base described in this application embodiment.

[0017] Figure 1B This diagram illustrates the structure of the mid-cloud interaction scenario in these implementation methods.

[0018] Figure 2 The diagram shows the transition process of the intelligent optimization method for the insurance business knowledge base described in this application embodiment.

[0019] Figure 3 The diagram shown is a schematic representation of the business processing process of the autonomously evolving intelligent business processing mode described in the embodiments of this application.

[0020] Figure 4 The diagram shown is a schematic representation of the process of obtaining feedback signals as described in an embodiment of this application.

[0021] Figure 5 This diagram illustrates the process of diagnosing problems with response information based on feedback signals, as described in an embodiment of this application.

[0022] Figure 6 The diagram shown is a structural schematic of the intelligent optimization system for the insurance business knowledge base described in this application embodiment.

[0023] Figure 7 The diagram shown is a structural schematic of the electronic device described in an embodiment of this application.

[0024] Component designation explanation

[0025] 1. Intelligent optimization device for insurance business knowledge base

[0026] 11 Storage devices

[0027] 12 Local Processors

[0028] 13 Display Terminals

[0029] 2-Terminal-Cloud Interactive System

[0030] 20 terminals

[0031] 21 Cloud Servers

[0032] 100 Insurance Business Knowledge Base Intelligent Optimization System

[0033] 101 User Problem Data Acquisition Module

[0034] 102 Feedback Signal Acquisition Module

[0035] 103 Tuning Strategy Generation Module

[0036] 104 Module for Obtaining Execution Results

[0037] 105 Verification Feedback Module

[0038] 200 electronic devices

[0039] 201 Memory

[0040] 202 processor

[0041] 203 Monitor

[0042] Steps S11 to S15

[0043] Steps S21 to S23

[0044] Steps S31 to S34 Detailed Implementation

[0045] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0046] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0047] Intelligent optimization of insurance knowledge bases refers to the automated and intelligent continuous optimization and updating of professional knowledge bases in the insurance field using artificial intelligence technologies, particularly advanced methods such as natural language processing, machine learning, and knowledge graphs. This optimization not only focuses on the accuracy of knowledge content but also emphasizes the rationality of the knowledge structure, its adaptability to application scenarios, and its alignment with business needs, thereby achieving dynamic evolution and maximizing the value of the knowledge base. However, current intelligent optimization of insurance knowledge bases still relies on manual intervention, and there are transformation barriers between business and technology.

[0048] The current optimization of insurance business models faces a triple structural dilemma: While insurance experts possess deep expertise in underwriting and claims rules, as well as the logic of exclusion clauses, they generally lack the ability to translate this expertise into technical specifications. They cannot accurately translate business descriptions such as "cross-invalidation of primary and supplementary insurance liabilities" into technical instructions for document supplementation and parameter adjustment, and they also face a vacuum in underlying model operation capabilities. While IT teams are proficient in algorithm architecture, they have a cognitive gap regarding the nested actuarial rules in policy cash value calculation, the correlation between waiting periods and exclusion clauses, and other business logic. Furthermore, they struggle to predict the profound impact of regulatory policy changes on the underwriting rule tree. This two-way capability gap leads to an inefficient cycle in manual optimization mechanisms—business issues require a multi-level transmission chain of "domain expert → structured report → technical interpretation → execution," with an average interaction frequency of up to 6 times per scenario, causing delays of over 24 hours in critical business operations such as updating auto insurance claims rules. More seriously, a certain percentage of business descriptions suffer semantic distortion during technical translation (e.g., "missing basis for determining liability for sudden death" is misinterpreted as a missing general clause), resulting in a misalignment between optimization actions and business needs. The bottleneck in adapting to dynamic business scenarios further exacerbates the system's vulnerability: there is a synchronization time difference of more than 72 hours between the update of the terms and conditions and the effective date of the regulatory policies, and there are problems such as the high error parsing rate in the nested scenarios of main insurance / supplementary insurance liabilities.

[0049] At least in response to the above problems, the following embodiments of this application provide an intelligent optimization method for the insurance business knowledge base.

[0050] The technical solutions in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0051] Figure 1A This diagram illustrates an application scenario of the intelligent optimization method for the insurance business knowledge base described in this application. The intelligent optimization device 1 for the insurance business knowledge base can be used to implement the intelligent optimization method for the insurance business provided in this application embodiment, but the application scenarios of the intelligent optimization method for the insurance business knowledge base provided in this application embodiment are not limited to... Figure 1A The insurance business knowledge base intelligent optimization device 1 shown is as follows. Figure 1AAs shown, the intelligent optimization device 1 for the insurance business knowledge base includes a storage device 11, a local processor 12, and a display terminal 13. The intelligent optimization method for the insurance business knowledge base provided in this embodiment can be applied to the local processor 12.

[0052] in, Figure 1A The local processor 12 can be a single local processor, a cluster of multiple local processors, or a cloud computing center, etc., and is not specifically limited here. Although Figure 1A Only one storage device 11, one local processor 12, and one display terminal 13 are shown, but it should be understood that... Figure 1A The examples in this paper are only for understanding this solution. The specific number of local processors 12 and display terminals 13 should be flexibly determined based on the actual situation.

[0053] In some other implementations, the intelligent optimization device 1 for the insurance business knowledge base may not include a display terminal 13, but only a local processor 12 with display function and a storage device 11. The intelligent optimization method for the insurance business knowledge base provided in this application embodiment can be applied to the local processor 12. The local processor 12 with display function may include tablet computers, laptops, handheld computers, mobile phones, personal computers, and voice interaction devices, or it may be a monitoring device, etc., which is not limited here.

[0054] In some other implementations, the intelligent optimization method for the insurance business knowledge base described in this application can be applied to end-to-cloud interaction scenarios. Figure 1B This diagram illustrates the structure of the endpoint-cloud interaction scenario in these implementation methods. For example... Figure 1B As shown, the terminal-cloud interaction system 2 includes a terminal 20 and a cloud server 21. The terminal 20 and the cloud server 21 can communicate with each other, and the communication method is not limited to wired or wireless.

[0055] The terminal 20 can be mobile or fixed. For example, it can be a wireless terminal or a wired terminal. A wireless terminal can refer to a device with wireless transceiver capabilities, which can be deployed indoors, outdoors, and in industrial workshops. The terminal 20 can be a mobile phone, tablet computer, laptop computer, etc., and is not limited thereto. The cloud server 21 can include one or more servers, or one or more processing nodes, or one or more virtual machines running on the server. The cloud server 21 can also be referred to as a server cluster, management platform, data processing center, etc., and is not limited thereto in this embodiment.

[0056] Figure 2 The diagram shows the transition process of the intelligent optimization method for the insurance business knowledge base in one embodiment of this application. Figure 2As shown, the intelligent optimization method for the insurance business knowledge base includes: obtaining a dual-track collaborative business processing mode; constructing model suggestions based on the dual-track collaborative business processing mode to obtain an enhanced collaborative iterative business processing mode; and constructing a model self-reflection based on the enhanced collaborative iterative business processing mode to obtain an autonomously evolving intelligent business processing mode.

[0057] Specifically, the dual-track collaborative business processing model adopts a linear process of "problem identification → model response → expert feedback → technical optimization → expert retesting". This model faces two major knowledge transformation barriers: First, while insurance experts possess professional knowledge such as underwriting and claims rules and the logic of exclusion clauses, their lack of technical transformation capabilities makes it difficult to accurately map business semantics such as "cross-invalidation of primary and supplementary insurance liabilities" into technical instructions such as knowledge graph supplementation and parameter configuration updates. Second, although the IT team has the ability to optimize algorithm architecture, they have cognitive biases regarding the nested actuarial rules for policy cash value and the correlation between waiting periods and exclusion clauses, making it even more difficult to predict the cascading impact of regulatory policy changes on the underwriting rule tree.

[0058] Based on a dual-track collaborative business processing model, an enhanced collaborative iterative business processing model is proposed. This model employs a closed-loop process: "problem identification → model response → expert feedback → model suggestion → expert optimization → expert retesting." The model examines user questions, model answers, document slices, model instructions, and user feedback on the question's answer, providing optimization suggestions such as supplementing knowledge base documents, tagging knowledge slices, and adjusting model divergence. The model then deeply considers these suggestions and optimizes in several aspects, including supplementing business knowledge by clearly labeling the applicable scenarios for documents A and B, adding product types and related Q&As, adding tags for product types and rule types, and strengthening delivery conditions; adding semantic information to documents; and adjusting parameter values ​​on the model side to adjust divergence, making the model's answers more stable and accurate. Business experts optimize the model by adding tags, supplementing documents, and providing supplementary materials. This model establishes a three-tiered expert capability advancement system: Level 1 provides basic model operation capabilities; Level 2 masters advanced skills such as parameter optimization; and Level 3 enables in-depth applications such as knowledge distillation, thereby achieving a leap in the capabilities of business experts. While this model enhances expert decision support through model suggestion, it still has limitations: the model can only output optimization suggestions, and ultimately, manual knowledge injection and parameter tuning are still required.

[0059] Based on an enhanced collaborative iterative business processing model, a self-evolving intelligent business processing model is constructed through model self-reflection. This model employs a fully automated closed-loop process: "problem posing → model response → expert feedback → self-reflection → model optimization → result verification → feedback feedback." The model receives user questions, model answers, document slices, model instructions, and user feedback on the questions and answers. It then identifies and reflects on existing business problems, understands the reasons behind the answers, and outputs optimization strategies, performing optimization operations accordingly. During the self-reflection phase, user feedback and questions are tagged. When a user poses the same question, the model retrieves both feedback and the question, and then uses inference to provide a response. This self-evolving intelligent business processing model establishes self-diagnosis based on user feedback, automatically updating the knowledge base, optimizing parameters, and correcting logic, forming an automated closed-loop optimization mechanism. A continuous learning framework ensures the timeliness of business rule iteration, frees up expert resources, and achieves an exponential improvement in question-answering accuracy. Figure 3 This diagram illustrates the business processing flow of an autonomously evolving intelligent business processing mode as shown in one embodiment of this application. Figure 3 As shown, the business processing process of the autonomous evolution intelligent business processing mode includes the following steps S11 to S15.

[0060] Step S11: Obtain user question data. User question data includes questions entered by the user regarding insurance business.

[0061] Step S12: The business model responds to the user question data to obtain feedback signals. These feedback signals, collected by the system, drive the business model to perform self-reflection and optimization. Feedback signals include user feedback and business expert feedback. Business expert feedback is the evaluation of the correctness or error of the model's response results by business experts. This includes the accuracy of the business model's response information (e.g., whether the correct clauses are cited), the logical rationality of the response information, the existence of semantic misunderstandings or useless clauses, and whether the business model's suggestions have been adopted or corrected. User feedback includes user feedback tags (e.g., user satisfaction ratings for the response information).

[0062] Figure 4 This is a schematic diagram illustrating the process of obtaining a feedback signal in one embodiment of this application. For example... Figure 4 As shown, the process of responding to the user's question data using the business model to obtain feedback signals includes the following steps S21 to S23.

[0063] Step S21: Perform semantic analysis on the user question data to obtain semantic analysis results.

[0064] Step S22: Process the semantic analysis results based on the five-dimensional dynamic data stream to obtain the response information. The five-dimensional dynamic data stream includes user question insurance type code, business model answer clause index, document slice version number, current business model instruction set, and user feedback tags, etc.

[0065] Step S23: Obtain the feedback signal based on the response information.

[0066] Specifically, users input questions into the business model. The business model performs semantic analysis on different types of insurance questions input by users and combines this with a five-dimensional dynamic data stream to determine the response information for the insurance questions. Users receive the response information output by the model and provide corresponding user feedback. At the same time, business experts evaluate the response information output by the model and provide professional judgment.

[0067] Step S13: Based on the feedback signal, perform problem diagnosis on the response information to generate an optimization strategy.

[0068] In one embodiment of this application, the process of diagnosing problems in response information based on the feedback signal to generate an optimization strategy includes: identifying problems in the response information based on the feedback signal to obtain business problems; and automatically generating the optimization strategy based on the business problems.

[0069] In one embodiment of this application, the process of identifying problems in response information based on the feedback signal to obtain business problems includes: performing in-depth analysis of the feedback signal using natural language processing based on a five-dimensional dynamic data stream to obtain reflection results; and identifying problems in the response information based on the reflection results to obtain the business problems.

[0070] Specifically, the business model undergoes self-reflection and model optimization through a large deep-reasoning model, r1. When the business model receives user feedback and feedback from business experts, it automatically diagnoses problems, uses a natural language processing reflection engine to perform error attribution analysis, identifies existing business issues (such as incorrect clause citations or logical deviations), and triggers logical corrections.

[0071] Figure 5 This diagram illustrates a process for diagnosing problems based on feedback signals in an embodiment of this application. Figure 5 As shown, the process of diagnosing problems with the response information based on the feedback signal to generate an optimization strategy further includes the following steps S31 to S34.

[0072] Step S31: Obtain the business problem description from the business expert based on the feedback from the business expert.

[0073] Step S32: Extract key semantic information from the business problem description to obtain business language.

[0074] Step S33: Establish the correspondence between the business language and the business model using a terminology mapping library.

[0075] Step S34: Perform deep semantic parsing on the business description to obtain technical optimization instructions.

[0076] Specifically, business experts describe existing problems (such as nested conflicts between primary and supplementary insurance liabilities) using pre-set structured report templates. These templates embed an insurance entity anchoring mechanism, automatically linking to the terms of the Insurance Law and company regulations, thus ensuring that the descriptions of business problems contain identifiable key semantic information. A terminology mapping library establishes a one-to-one correspondence between business language (e.g., missing determination of liability for sudden death) and corresponding technical operations (e.g., knowledge graph supplementation and parameter adjustment), thereby constructing a built-in mapping relationship between business terminology and technical operations. This ensures that business expressions such as "missing determination of liability for sudden death" can be transformed into technical optimization instructions such as "knowledge graph supplementation" and "parameter adjustment." When business experts use structured templates to describe problems (such as the failure of cross-liability between primary and supplementary insurance), the business model can use natural language processing and awareness recognition technologies to perform deep semantic analysis on the business description based on technical tuning instructions. This automatically identifies key business entities (such as primary insurance, supplementary insurance, and cross-liability), extracts key entities and logical relationships, and maps them to predefined technical actions (such as supplementing relationships in the knowledge graph, adjusting the weight parameters of specific clauses, and / or modifying inference logic rules). This generates precise tuning strategies (such as adjusting the accident insurance liability boundary weight coefficient to 0.85), automatically matching executable technical processes. This achieves a lossless transformation from business requirements to technical tuning instructions, effectively solving the comprehension gap between business experts and technical personnel.

[0077] A mapping system and semantic anchoring technology were used to establish a correspondence between business terms and technical operations, ensuring that business descriptions such as "missing liability determination for sudden death" could be accurately translated into technical optimization instructions such as "knowledge graph supplementation." Natural language processing and intent recognition enabled the business model to perform deep semantic analysis of business problems, extracting key entities and logical relationships, thereby compensating for technical personnel's insufficient understanding of business logic. Simultaneously, business experts described problems using preset templates, automatically linking them to insurance law clauses and company rules, reducing semantic translation errors. By establishing a mapping bridge between business semantics and technical operations, the problem of optimization instruction conversion failures caused by the lack of technical capabilities among insurance back-office experts was solved, as well as the dilemma of logical analysis failures caused by blind spots in business understanding within the IT team.

[0078] Step S14: Use the optimization strategy to dynamically optimize the business model to obtain the optimization execution result.

[0079] Step S15: Verify the optimization execution results and provide feedback to obtain the intelligent optimization results of the insurance business knowledge base.

[0080] In one embodiment of this application, the intelligent optimization method for the insurance business knowledge base further includes: fusing a historical case library with an implemented regulatory version tree to obtain a dynamic business perception matrix; matching the business problem with the dynamic business perception matrix; and using a liability boundary topology algorithm to decompose the complex logic in the dynamic business perception matrix to obtain the optimization strategy. The dynamic business perception matrix is ​​a dynamic knowledge base, a model representing entities (insurance types, clauses, and liabilities), relationships (nesting, exclusion, and dependence), rules (calculation and judgment logic), and dynamic weights in insurance business. It can perceive business changes and understand complex liability nesting and execution logic reasoning. The real-time regulatory version tree is a regulatory policy knowledge graph, capable of storing the latest regulatory documents and recording the entire lifecycle of policy clauses—from promulgation, effectiveness, revision, and repeal—and the relationships between policy clauses in a machine-readable and reasonable manner. Deeply fusing past experience patterns in the historical case library with the latest legally binding rules of the real-time regulatory version tree outputs a structured dynamic business matrix with weights and probabilistic relationships, enabling the perception and prediction of business scenarios.

[0081] The liability boundary topology algorithm is used to handle complex liability relationships in insurance business. It abstracts liability management in insurance clauses (such as exclusions, dependencies, conflicts, and inclusions) into a structured liability topology graph, where nodes represent insurance entities and edges represent logical relationships between entities. By quickly traversing all relevant paths in the liability topology graph, it automatically identifies business conflicts between different paths and quantifies conflicting liability paths using weights, thereby automatically outputting corresponding optimization strategies.

[0082] For example, when a user asks, "A customer who purchased accident insurance and supplementary high-risk sports insurance fractured a bone while skiing. Is it reasonable for the insurance company to refuse compensation because skiing is an exclusion clause?", the business model matches the user's question with a dynamic business perception matrix. The liability boundary topology algorithm decomposes the logical paths in the dynamic business perception matrix. Based on the key entities of the question, such as "accident insurance," "fracture," "skiing," "high-risk sports exclusion," and "supplementary high-risk sports insurance," it extracts all relevant nodes and their corresponding entity logical relationships from the dynamic business perception matrix, resulting in a liability topology graph. After traversing the liability topology graph, two paths are obtained: Path 1: Fracture is covered by accident insurance, and compensation is paid; Path 2: Fracture is a high-risk sport—triggers the exclusion clause—excludes accident insurance liability. Since there is a business conflict between the two paths, the dynamic business perception matrix is ​​used to determine the appropriate path. The historical weights and regulatory rules integrated into the business awareness matrix are further comprehensively calculated. It is found that the "Additional High-Risk Sports Insurance" node has a heavier weight when there is a "coverage" edge pointing to the "Disclaimer Clause". Therefore, path two is deemed invalid, and an optimization strategy corresponding to path one is generated. When natural language processing identifies errors in clause citation or logical deviations in the response information, the responsibility boundary topology algorithm is used to traverse the dynamic business awareness matrix based on the latest regulatory version tree and historical case library. Historical cases, new policy versions, and response information are associated and analyzed. Based on the analysis results, technical optimization instructions are generated, and optimization strategies are obtained according to the technical optimization instructions. This automatically adjusts the knowledge base content and model parameters to obtain optimization execution results. The optimization execution results are then simulated and verified in a business sandbox and fed back to the business model, thereby completing closed-loop adaptive adjustment optimization and continuous evolution.

[0083] In one embodiment of this application, after the business model completes autonomous optimization and verification, the feedback optimization results are fed back into the system to update model parameters, optimize the knowledge base, and correct the dynamic business awareness matrix. The feedback optimization results include user feedback tags, response information and clause indexes generated by the business model, optimization execution results, and business sandbox verification results. The feedback optimization results are fed back into the knowledge base to drive dynamic updates, such as insurance clauses, correcting liability boundary definitions, synchronizing regulatory policy changes, and supplementing typical cases. Valid feedback data from the optimization results is stored in the model training dataset for continuous learning by the business model. The feedback optimization results can also be used to optimize dynamic strategies and compliance judgment logic in the dynamic business awareness matrix. The feedback optimization results enhance the model's self-reflection and simulation verification capabilities. Feedback feedback achieves a complete closed loop from problem discovery to continuous optimization, improving the model's autonomous evolution capability.

[0084] In one embodiment of this application, when business rules change (e.g., auto insurance claims rules are adjusted due to regulatory policies), the business model automatically identifies the change signals by collecting five-dimensional dynamic data streams and automatically adjusts and optimizes the response logic in conjunction with a dynamic business perception matrix. A natural language processing reflection engine is used to analyze the root causes of differences, adjusting the calculation weights of relevant rules, correcting logical judgment conditions, or adding new rules in the dynamic business perception matrix. The adjusted rules are then tested and verified through a business sandbox. Once verification is successful, the results are fed back into the model for deployment and online application. The new rules are then applied in similar situations, forming a self-refreshing knowledge base mechanism. For example, the original model used the old rules for the question of "whether compensation is paid for accidents occurring during the waiting period." After optimization and updates, the business model can automatically identify regulatory updates and apply the new rules, achieving dynamic adaptation of underwriting and claims rules and avoiding delays caused by manual intervention.

[0085] In one embodiment of this application, a clause version tracking mechanism is used to capture notifications of clause updates, additions, or repeals based on the regulatory version tree and document slice version numbers, triggering an update of the regulatory version tree. When a clause is updated, the knowledge base is automatically refreshed, and the business model synchronously updates its referencing logic to ensure that the clauses referenced by the model are always the latest valid versions. For example, if the original business model referenced an expired clause when answering the question "Is sudden death covered by accidental death?", after optimization and updates, the business model can automatically identify clause changes and update its response content, achieving real-time synchronization of clause versions.

[0086] In one embodiment of this application, the business model utilizes a liability boundary topology algorithm to transform complex insurance clauses into a liability relationship graph. For example, nodes represent insurance liabilities, and edges represent inclusion, exclusion, or conditional triggering relationships. When a user queries, the business model maps the user's question to a starting point in the graph. The algorithm automatically traverses all relevant nodes and edges to obtain one or more liability determination paths and identifies conflict points (e.g., liability exclusion) within these paths. It automatically decomposes the complex liability relationships between the main insurance and supplementary insurance, identifying issues such as cross-invalidation and overlapping liabilities. The model is optimized by adjusting the association weights between nodes in the graph or correcting the logical relationships of conflicting edges. For example, the original business model might have logical inconsistencies when handling issues such as "exclusion of the main insurance leading to the invalidation of the supplementary insurance." After model optimization, the liability boundary topology algorithm can clearly parse the liability chain and output accurate judgments.

[0087] In summary, the intelligent optimization method for insurance business knowledge base provided in this application enables the business model to determine response information by comprehensively analyzing the semantic content of user input and combining it with five-dimensional dynamic data flow. Upon receiving user feedback, it no longer simply outputs optimization suggestions for manual execution. Instead, it uses a natural language processing reflection engine to deeply analyze the feedback signal. Through a built-in insurance knowledge graph and clause association mechanism, the model automatically matches relevant legal provisions, company rules, and historical cases, identifies the business scenario to which the problem belongs, and performs semantic parsing. The parsed business semantics are automatically matched to executable technical operation processes, thereby autonomously generating executable optimization strategies. This establishes a mapping bridge between business semantics and technical operations, reducing manual translation steps and solving the problem of optimization instruction conversion failures caused by the lack of technical capabilities among insurance back-office experts. Simultaneously, it resolves the dilemma of logical parsing failures caused by blind spots in business understanding within the IT team. It achieves automatic identification and execution of business requirements into technical actions, as well as lossless conversion from business requirements to technical optimization instructions, effectively solving the problem of comprehension gaps between business experts and technical personnel. Building upon this foundation, the model also employs a natural language processing reflection engine for self-evaluation, assessing the accuracy and compliance of its responses. When sensitive operations are involved, a business sandbox verification mechanism is activated to ensure the security and reliability of the output. Simultaneously, leveraging a dynamic strategy matrix that integrates historical cases and real-time regulatory rules, a responsibility boundary topology algorithm is used to decompose complex business logic. The optimization effects are then simulated and verified within the business sandbox to ensure operational compliance and security. Finally, the optimization results are fed back into the knowledge base and model parameters through a feedback mechanism, driving continuous model evolution and forming a complete adaptive optimization process of "problem identification—autonomous diagnosis—automatic optimization—closed-loop verification."

[0088] The scope of protection of the intelligent optimization method for insurance business knowledge base described in this application is not limited to the execution order of the steps listed in this embodiment. Any solution implemented by adding, subtracting, or replacing steps in the prior art based on the principles of this application is included within the scope of protection of this application.

[0089] This application also provides an intelligent optimization system for an insurance business knowledge base. The intelligent optimization system for an insurance business knowledge base can implement the intelligent optimization method for an insurance business knowledge base described in this application. However, the implementation device for the intelligent optimization method for an insurance business knowledge base described in this application includes, but is not limited to, the structure of the intelligent optimization system for an insurance business knowledge base listed in this embodiment. All structural modifications and substitutions of the prior art made based on the principles of this application are included within the protection scope of this application.

[0090] Figure 6 The diagram shown is a structural schematic of an intelligent optimization system for an insurance business knowledge base according to one embodiment of this application. Figure 6As shown, the intelligent optimization system 100 for the insurance business knowledge base includes: a user problem data acquisition module 101, a feedback signal acquisition module 102, an optimization strategy generation module 103, an optimization execution result acquisition module 104, and a verification feedback module 105. The user problem data acquisition module 101 is used to acquire user problem data. The feedback signal acquisition module 102 is used to respond to the user problem data using a business model to obtain feedback signals, including user feedback and business expert feedback. The optimization strategy generation module 103 is used to perform problem diagnosis on the response information based on the feedback signals to generate optimization strategies. The optimization execution result acquisition module 104 is used to dynamically optimize the business model using the optimization strategies to obtain optimization execution results. The verification feedback module 105 is used to verify the optimization execution results and provide feedback to obtain intelligent optimization results for the insurance business knowledge base.

[0091] It should be noted that, Figure 6 The modules in the insurance business knowledge base intelligent optimization system 100 shown are related to... Figure 2 The steps in the intelligent optimization method for the insurance business knowledge base correspond one-to-one, and will not be elaborated here.

[0092] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, or methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of apparatuses or modules or units may be electrical, mechanical, or other forms.

[0093] The modules / units described as separate components may or may not be physically separate. The components shown as modules / units may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules / units can be selected to achieve the objectives of the embodiments of this application, depending on actual needs. For example, the functional modules / units in the various embodiments of this application may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0094] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0095] This application also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the intelligent optimization method for the insurance business knowledge base provided in this application. Those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing a processor. The program can be stored in a computer-readable storage medium, which is a non-transitory medium, such as random access memory, read-only memory, flash memory, hard disk, solid-state hard disk, magnetic tape, floppy disk, optical disk, and any combination thereof. The storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. This available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0096] This application embodiment may also provide an electronic device. Figure 7 The diagram shown is a structural schematic of an electronic device 200 according to an embodiment of this application. Figure 7 As shown, in this embodiment, the electronic device 200 includes a memory 201 and a processor 202.

[0097] The memory 201 is used to store computer programs. In some possible implementations, the memory 201 may include various media capable of storing program code, such as ROM, RAM, magnetic disk, USB flash drive, memory card, or optical disk.

[0098] In this embodiment, memory 201 may include a computer system readable medium in the form of volatile memory, such as RAM and / or cache memory. Electronic device 200 may further include other removable / non-removable, volatile / non-volatile computer system storage media. Memory 201 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application.

[0099] The processor 202 is connected to the memory 201 and is used to execute the computer program stored in the memory 201 so that the electronic device 200 executes the intelligent tuning method of the insurance business knowledge base.

[0100] For example, processor 202 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc. In other embodiments, processor 202 may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0101] In some implementations, the electronic device 200 provided in this application embodiment may further include a display 203. The display 203 is communicatively connected to the memory 201 and the processor 202, and is used to display the relevant graphical user interface (GUI) of the intelligent tuning method for the insurance business knowledge base.

[0102] In this embodiment, the display 203 may include a display screen (display panel). In some implementations, the display panel may be configured using a liquid crystal display (LCD), an organic light-emitting diode (OLED), or other similar forms. Furthermore, the display 203 may also be a touch panel (touchscreen, touch screen), which may include a display screen and a touch-sensitive surface. When the touch-sensitive surface detects a touch operation on or near it, it transmits the information to the processor 202 to determine the type of touch event. Subsequently, the processor 202 provides corresponding visual output on the display device based on the type of touch event.

[0103] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0104] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A method for intelligent optimization of an insurance business knowledge base, characterized in that, The intelligent optimization method for the insurance business knowledge base includes: Obtain user issue data; The business model is used to respond to the user question data to obtain feedback signals, including user feedback and business expert feedback; Based on the feedback signal, the response information is used to diagnose problems and generate an optimization strategy. The business model is dynamically optimized using the optimization strategy to obtain the optimization results. The optimization results are verified and feedback is provided to obtain intelligent optimization results for the insurance business knowledge base.

2. The intelligent optimization method for the insurance business knowledge base according to claim 1, characterized in that, The process of responding to the user question data using a business model to obtain feedback signals includes: Perform semantic analysis on the user question data to obtain semantic analysis results; The semantic analysis results are processed based on a five-dimensional dynamic data stream to obtain the response information; The feedback signal is obtained based on the response information.

3. The intelligent optimization method for the insurance business knowledge base according to claim 1, characterized in that, The process of diagnosing problems based on the feedback signal to generate an optimization strategy includes: Based on the feedback signal, the response information is used to identify problems and obtain business issues. The optimization strategy is automatically generated based on the aforementioned business problem.

4. The intelligent optimization method for the insurance business knowledge base according to claim 3, characterized in that, The process of identifying business issues based on the feedback signal in the response information includes: Based on five-dimensional dynamic data stream, natural language processing is used to perform in-depth analysis of the feedback signal to obtain reflection results; Based on the reflection results, the response information is used to identify problems and obtain the business issues.

5. The intelligent optimization method for the insurance business knowledge base according to claim 3, characterized in that, The intelligent optimization method for the insurance business knowledge base also includes: Integrate the historical case library with the implementation regulatory version tree to obtain a dynamic business awareness matrix; Match the business problem with the dynamic business awareness matrix; The complex logic in the dynamic business awareness matrix is ​​decomposed using the responsibility boundary topology algorithm to obtain the optimization strategy.

6. The intelligent optimization method for the insurance business knowledge base according to claim 1, characterized in that, The process of diagnosing problems with the response information based on the feedback signal to generate an optimization strategy also includes: Obtain the business problem description from the business expert based on the feedback from the business expert; Key semantic information is extracted from the business problem description to obtain the business language; Establish the correspondence between the business language and the business model using a terminology mapping library; Perform deep semantic analysis on the business description to obtain technical optimization instructions.

7. The intelligent optimization method for the insurance business knowledge base according to claim 1, characterized in that, The intelligent optimization methods for the insurance business knowledge base also include: Obtain a dual-track collaborative business processing model; Based on the aforementioned dual-track collaborative business processing mode, a model suggestion is constructed to obtain an enhanced collaborative iterative business processing mode. Based on the enhanced collaborative iterative business processing model, a model is constructed to autonomously reflect and obtain an autonomously evolving intelligent business processing model.

8. An intelligent optimization system for an insurance business knowledge base, characterized in that, The intelligent optimization system for the insurance business knowledge base includes: The user issue data acquisition module is used to acquire user issue data; The feedback signal acquisition module is used to respond to the user question data using the business model to obtain feedback signals, including user feedback and business expert feedback. The optimization strategy generation module is used to perform problem diagnosis on the response information based on the feedback signal in order to generate an optimization strategy. The optimization execution result acquisition module is used to dynamically optimize the business model using the optimization strategy to obtain the optimization execution result. The verification and feedback module is used to verify the optimization execution results and provide feedback to obtain the intelligent optimization results of the insurance business knowledge base.

9. An electronic device, characterized in that, The electronic device includes: A memory on which computer programs are stored; A processor, communicatively connected to the memory, is used to execute the computer program to implement the intelligent optimization method for the insurance business knowledge base as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by an electronic device, the program implements the intelligent optimization method for the insurance business knowledge base as described in any one of claims 1 to 7.

Images