An agent-based product prototyping method, apparatus, device, and medium

Abstract

The application relates to the technical field of product research and development, and discloses a product prototype generation method, device, equipment and medium based on an agent, which is applied to an insurance product prototype generation platform and comprises the following steps: receiving a product prototype generation request based on a first agent, using a second agent to perform intent recognition on the product prototype generation request, obtaining a function requirement, using a third agent to generate a flowchart according to the function requirement in combination with a preset knowledge graph, using a fourth agent to check the generation quality of the flowchart based on a preset rule library and obtaining a checking result, and if the first agent judges that the checking result meets preset conditions, using a fifth agent to generate a target product prototype according to the flowchart in combination with a preset specification library. The application can be applied to product prototype design scenes in the fields of financial technology and medical health, and through the collaborative work of multiple agents, the design and generation of a product prototype are automatically realized, and the generation efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of product development technology and can be applied to the fields of financial technology and healthcare. In particular, it relates to a method, apparatus, equipment and medium for generating product prototypes based on intelligent agents. Background Technology

[0002] A product prototype is an initial model used in the product development phase to showcase product functionality, interaction logic, and interface layout. It serves as a crucial bridge connecting requirements analysis and technical implementation. Currently, product prototype design primarily relies on designers manually drawing prototypes using tools such as Figma, Sketch, and Axure. For example, in the design of insurance underwriting or wealth management account opening product prototypes in the financial sector, designers manually draw pages, configure navigation logic, and annotate interaction descriptions based on requirements documents. However, this method of product prototype generation has limitations in terms of automation. The entire process heavily depends on manual operation by designers. From understanding requirements to page drawing, from logical connection to interaction configuration, a large amount of repetitive work is required, resulting in low efficiency. Specifically, designers need to understand each function in the requirements document, translate it into page layout, and then manually configure the navigation relationships and interaction logic between pages. Each step consumes a significant amount of time and effort. When requirements change, the above process needs to be repeated, further lengthening the prototype design cycle. In complex business scenarios such as insurance and healthcare, involving dozens or even hundreds of pages and processes, relying entirely on manual drawing makes it difficult to quickly respond to business needs. Therefore, how to automatically generate product prototypes to improve generation efficiency has become an urgent problem to be solved. Summary of the Invention

[0003] This invention provides a product prototype generation method, apparatus, computer equipment, and medium based on intelligent agents, to solve the problem of how to automatically generate product prototypes and improve generation efficiency.

[0004] Firstly, a product prototype generation method based on intelligent agents is provided. This method is applied to an insurance product prototype generation platform, which includes a first intelligent agent, a second intelligent agent, a third intelligent agent, a fourth intelligent agent, and a fifth intelligent agent, comprising: Based on the first intelligent agent receiving the user's product prototype generation request for the target insurance application, the product prototype generation request is sent to the second intelligent agent. The second intelligent agent is used to perform intent recognition on the product prototype generation request, obtain the functional requirements corresponding to the product prototype generation request, send the functional requirements to the first intelligent agent, and send the functional requirements to the third intelligent agent based on the first intelligent agent. Using the third intelligent agent, based on the functional requirements and combined with a preset knowledge graph, a flowchart for implementing the functional requirements is generated, and the flowchart is sent to the first intelligent agent. Based on the first intelligent agent, the flowchart is then sent to the fourth intelligent agent. The fourth intelligent agent checks the generation quality of the flowchart based on a preset rule base, obtains the check result, and sends the check result to the first intelligent agent. Based on the first agent's determination of whether the inspection result meets the preset conditions, if the inspection result meets the preset conditions, the flowchart is sent to the fifth agent. Using the fifth intelligent agent, according to the flowchart and in conjunction with the preset specification library, a target product prototype corresponding to the target insurance application is generated, and the target product prototype is sent to the first intelligent agent. Based on the first intelligent agent, the target product prototype is sent to the user.

[0005] Secondly, a product prototype generation device based on intelligent agents is provided. This device is applied to an insurance product prototype generation platform, which includes a first intelligent agent, a second intelligent agent, a third intelligent agent, a fourth intelligent agent, and a fifth intelligent agent. The receiving module is used to receive a product prototype generation request from a user for a target insurance application from the first intelligent agent and send the product prototype generation request to the second intelligent agent. The intent recognition module is used to use the second intelligent agent to perform intent recognition on the product prototype generation request, obtain the functional requirements corresponding to the product prototype generation request, send the functional requirements to the first intelligent agent, and send the functional requirements to the third intelligent agent based on the first intelligent agent. The process generation module is used to use the third intelligent agent to generate a flowchart to implement the functional requirements based on the functional requirements and combined with a preset knowledge graph, send the flowchart to the first intelligent agent, and send the flowchart to the fourth intelligent agent based on the first intelligent agent. The rule checking module is used to use the fourth intelligent agent to check the generation quality of the flowchart based on a preset rule base, obtain the check result, and send the check result to the first intelligent agent. The inspection and judgment module is used to determine whether the inspection result meets the preset conditions based on the first intelligent agent. If the inspection result meets the preset conditions, the flowchart is sent to the fifth intelligent agent. The prototype generation module is used by the fifth intelligent agent to generate a target product prototype corresponding to the target insurance application according to the flowchart and in conjunction with a preset specification library, and sends the target product prototype to the first intelligent agent, and sends the target product prototype to the user based on the first intelligent agent.

[0006] Thirdly, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described agent-based product prototype generation method.

[0007] Fourthly, a computer-readable storage medium is provided, which stores a computer program that, when executed by a processor, implements the steps of the above-described agent-based product prototype generation method.

[0008] In the aforementioned solution implemented by the intelligent agent-based product prototype generation method, apparatus, computer equipment, and storage medium, the product prototype generation method is applied to an insurance product prototype generation platform. The insurance product prototype generation platform includes a first intelligent agent, a second intelligent agent, a third intelligent agent, a fourth intelligent agent, and a fifth intelligent agent. The first intelligent agent receives a user's product prototype generation request for a target insurance application and sends the request to the second intelligent agent. The second intelligent agent performs intent recognition on the product prototype generation request to obtain the corresponding functional requirements, which are then sent to the first intelligent agent. Based on the first intelligent agent, the functional requirements are sent to the third intelligent agent. The third intelligent agent then... Based on the functional requirements and a pre-defined knowledge graph, a flowchart for implementing the functional requirements is generated and sent to the first intelligent agent. The first intelligent agent then sends the flowchart to the fourth intelligent agent. The fourth intelligent agent, based on a pre-defined rule base, checks the generation quality of the flowchart and obtains the check result, which is sent back to the first intelligent agent. The first intelligent agent then determines whether the check result meets pre-defined conditions. If the check result meets the pre-defined conditions, the flowchart is sent to the fifth intelligent agent. The fifth intelligent agent, based on the flowchart and a pre-defined specification library, generates a target product prototype corresponding to the target insurance application and sends the target product prototype to the first intelligent agent. Based on the first intelligent agent, the target product prototype is sent to the user.

[0009] Specifically, by leveraging the collaborative efforts of the first, second, third, fourth, and fifth intelligent agents within the insurance product prototype generation platform, the design and generation of insurance product prototypes are automated, improving generation efficiency. The first intelligent agent acts as a central coordinator, uniformly receiving and distributing user requests, and managing task flow and state synchronization among the agents, thereby enhancing overall coordination and task execution efficiency among multiple agents. The second intelligent agent performs intent recognition to extract functional requirements, enabling subsequent agents to accurately execute downstream tasks based on these requirements without deviating from user needs. The third intelligent agent generates flowcharts using a pre-set knowledge graph, ensuring the professionalism of the generation process. The fourth intelligent agent performs checks using a pre-set rule base, ensuring the compliance and completeness of the process. Finally, the fifth intelligent agent designs the product prototype from the flowchart based on a pre-set specification library, ensuring the standardization and consistency of the design. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of an application environment for a product prototype generation method based on an intelligent agent provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart illustrating a product prototype generation method based on an intelligent agent, provided in Embodiment 2 of the present invention. Figure 3 This is a flowchart illustrating a product prototype generation method based on an intelligent agent, provided in Embodiment 3 of the present invention. Figure 4 This is a schematic diagram of a product prototype generation device based on an intelligent agent provided in Embodiment 4 of the present invention; Figure 5 This is a schematic diagram of the structure of a computer device provided in Embodiment 5 of the present invention. Detailed Implementation

[0012] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0013] The artificial intelligence-based agent-based product prototype generation method provided in Embodiment 1 of this invention can be applied to, for example... Figure 1 The invention pertains to a server-side application environment. The server connects to a database and a client. The client communicates with the server via a network. The server receives product prototype generation requests from the client for the target insurance application. The client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster. Specific embodiments are described in detail below.

[0014] For example, suppose an insurance product manager needs to design a prototype for a cancer insurance product targeting 50-year-old users. He generates a natural language request for a cancer insurance application process for 50-year-old users through platform input, including health declaration, premium calculation, and payment confirmation functions. The first agent receives this request and forwards it to the second agent. The second agent performs intent recognition, parsing the functional requirements, including the target user's age, insurance type, and core process steps. This functional requirement is then passed to the third agent through the first agent. The third agent, based on the functional requirements and a pre-set knowledge graph (which stores the standard application steps and execution order for cancer insurance, such as underwriting judgment after health declaration, premium calculation and payment confirmation only after underwriting approval, and rejection notification if approval is not granted), generates a flowchart containing nodes and branches such as health declaration, underwriting judgment, premium calculation, payment confirmation, and rejection notification. This flowchart is then sent to the fourth agent through the first agent. The fourth agent checks the flowchart based on a pre-set rule base to verify its validity. The system checks whether the product complies with insurance industry compliance requirements (such as whether the health declaration includes necessary inquiries and whether the underwriting decision branches are complete) and whether the topology is complete (whether there are isolated nodes or missing termination nodes). Once the first agent detects that the check results meet preset conditions, it sends the flowchart to the fifth agent. The fifth agent, based on the flowchart and page templates and component specifications in a preset specification library, converts the health declaration node into an interactive page containing health declaration questions and yes / no options, the premium calculation node into a page displaying the premium amount and calculation details, and the payment confirmation node into a page containing a payment button and a compliance statement. Finally, it generates a complete product prototype that is clickable, navigable, and simulates the entire insurance application process, and sends it to the user through the first agent. Throughout this process, the design and generation of the insurance product prototype are automated, improving efficiency.

[0015] Please see Figure 2 As shown, Figure 2This is a flowchart illustrating a product prototype generation method based on intelligent agents according to Embodiment 2 of the present invention. The product prototype generation method is applied to an insurance product prototype generation platform, which includes a first intelligent agent, a second intelligent agent, a third intelligent agent, a fourth intelligent agent, and a fifth intelligent agent, and includes the following steps: S21: The first intelligent agent receives the user's product prototype generation request for the target insurance application and sends the product prototype generation request to the second intelligent agent.

[0016] In this embodiment, the insurance product prototype generation platform can refer to an intelligent system used to automatically generate prototypes of insurance application products. The first intelligent agent can refer to a coordinating intelligent agent responsible for receiving user requests, distributing tasks, and summarizing results. The second intelligent agent can refer to an intelligent agent responsible for interpreting the intent and extracting the requirements of user requests. The target insurance application can refer to an insurance business software program to be prototyped, such as a mini-program that implements insurance application functions, a claims application application, etc. The product prototype generation request can refer to natural language description information submitted by the user that contains the functional requirements of the target insurance application.

[0017] For example, a first intelligent agent can receive a user's request to generate a product prototype for a health insurance application mini-program. This request could be to generate a cancer insurance application process for users aged 50, including health declaration, premium calculation, and payment confirmation functions. The first intelligent agent can then send this product prototype generation request to a second intelligent agent.

[0018] S22: Use the second intelligent agent to perform intent recognition on the product prototype generation request, obtain the functional requirements corresponding to the product prototype generation request, send the functional requirements to the first intelligent agent, and send the functional requirements to the third intelligent agent based on the first intelligent agent.

[0019] In this embodiment, functional requirements can refer to information extracted from the product prototype generation request that describes the specific business capabilities that the target insurance application should possess.

[0020] For example, the second intelligent agent performs intent recognition on a product prototype generation request to generate a cancer insurance application process for 50-year-old users, including health declaration, premium calculation, and payment confirmation functions. The resulting functional requirements are: target user: 50 years old, insurance type: cancer insurance, core process steps: health declaration, premium calculation, and payment confirmation. This functional requirement is then sent to the first intelligent agent, which forwards it to the third intelligent agent.

[0021] S23: Using a third intelligent agent, based on the functional requirements and combined with a preset knowledge graph, generate a flowchart to achieve the functional requirements, send the flowchart to the first intelligent agent, and then send the flowchart to the fourth intelligent agent based on the first intelligent agent.

[0022] In this embodiment, the third intelligent agent can refer to an intelligent agent responsible for generating a flowchart based on functional requirements and a preset knowledge graph. The nodes of the preset knowledge graph represent insurance steps, and the attributes of the nodes represent the execution information of the corresponding insurance steps. The execution information can refer to the data fields required to complete the insurance step, the type of output result, and the business rules that the step depends on. For example, for the health disclosure node, its execution information may include: the input fields are whether there is high blood pressure, whether there is diabetes, and whether there is heart disease, and the output result is the health disclosure conclusion (pass / fail). The business rule is that if any answer is yes, the conclusion is failure. The edges between nodes represent the execution order relationship between the corresponding insurance steps. For example, the edge from the health disclosure node to the underwriting judgment node indicates that the underwriting judgment step is entered after the health disclosure is completed. The flowchart can refer to a directed graph structure composed of insurance step nodes, the execution information of insurance step nodes, and the connections between steps, used to describe the complete business execution path required to complete the functional requirements.

[0023] For example, based on the functional requirements of the third intelligent agent, the target user is 50 years old, the insurance type is cancer insurance, and the core process steps are health declaration, premium calculation, and payment confirmation. Combined with a preset knowledge graph, the generated flowchart includes: Start node → Health declaration node → Underwriting judgment node (branching based on health declaration results) → Branch 1 (pass) points to premium calculation node → Premium calculation node points to payment confirmation node → Payment confirmation node points to end node; Branch 2 (fail) points to rejection notification node → Rejection notification node points to end node. This flowchart is then sent to the first intelligent agent, which forwards it to the fourth intelligent agent.

[0024] Optionally, a third intelligent agent is used to generate a flowchart for implementing the functional requirements, based on the functional requirements and a pre-defined knowledge graph, including: Using a third-party intelligent agent, entity parsing is performed on the functional requirements to obtain the requirement entities; Based on the demand entity, the node corresponding to each demand entity is matched from the preset knowledge graph; Based on the nodes corresponding to all demand entities, the attributes of the nodes, and the edges between the corresponding nodes, determine the execution path between the nodes corresponding to all demand entities. A flowchart is generated based on the insurance steps represented by each node in all execution paths and the execution order relationship between the insurance steps represented by the edges between the corresponding nodes.

[0025] Among them, the demand entity can refer to the elements extracted from functional requirements that are related to the insurance business process.

[0026] For example, based on the third-party intelligent agent's analysis of the target user's functional requirements (age 50, insurance type: cancer insurance), and the core process steps (health declaration, premium calculation, payment confirmation), the resulting requirement entities can include health declaration, premium calculation, and payment confirmation. Based on the health declaration entity, a health declaration node is matched from a pre-defined knowledge graph; based on the premium calculation entity, a premium calculation node is matched from the pre-defined knowledge graph; based on the payment confirmation entity, a payment confirmation node is matched from the pre-defined knowledge graph. Finally, based on the health declaration node, premium stage node, and payment confirmation node, and considering all requirement entities... The edge relationships between corresponding nodes (e.g., the knowledge graph stores edges from the health disclosure node to the underwriting judgment node, edges from the underwriting judgment node to the premium calculation node and the rejection notification node via conditional branches, and edges from the premium calculation node to the payment confirmation node) determine the execution path as: health disclosure node → underwriting judgment node → (if passed) premium calculation node → payment confirmation node → end, or (if failed) rejection notification node → end; based on the insurance steps represented by each node in this execution path, the execution information of the insurance steps, and the execution order relationships represented by the edges between nodes, a corresponding flowchart is generated.

[0027] S24: Using the fourth intelligent agent, based on the preset rule base, the quality of the generated flowchart is checked, the check result is obtained, and the check result is sent to the first intelligent agent.

[0028] S25: Based on the first agent, determine whether the inspection result meets the preset conditions. If the inspection result meets the preset conditions, send the flowchart to the fifth agent.

[0029] In this embodiment, the fourth intelligent agent can refer to the intelligent agent responsible for checking the flowchart based on the preset rule base. The preset rule base can refer to the set of rules stored in advance to determine whether the flowchart meets the business logic requirements and industry standard constraints. The check result can refer to the structured information output by the fourth intelligent agent after checking the generation quality of the flowchart, which indicates whether there are problems in the flowchart and what problems exist. The preset conditions can refer to the standards used to determine whether the check result is qualified.

[0030] Specifically, a fourth intelligent agent can be used to check the flowchart based on the rules stored in the preset rule base, obtain the check result, and send the check result to the first intelligent agent. The first intelligent agent then determines whether the check result meets the preset conditions. If the check result meets the preset conditions, the flowchart is sent to the fifth intelligent agent.

[0031] S26: Using the fifth intelligent agent, based on the flowchart and combined with the preset specification library, generate a target product prototype corresponding to the target insurance application, send the target product prototype to the first intelligent agent, and send the target product prototype to the user based on the first intelligent agent.

[0032] In this embodiment, the fifth intelligent agent can refer to the intelligent agent responsible for generating a product prototype based on the flowchart and a preset specification library. The preset specification library can refer to a collection of page templates, component styles, layout rules and interaction specifications that are pre-stored to map flowchart nodes to specific pages. The target product prototype can refer to the product prototype of the target insurance application.

[0033] Optionally, a fifth intelligent agent is used to generate a prototype of the target product corresponding to the target insurance application based on the flowchart and a pre-defined specification library, including: Using the fifth intelligent agent, the flowchart is converted into the corresponding interactive page according to the page specifications in the preset specification library; Use the interactive page as a prototype of the target product.

[0034] Interactive pages can refer to front-end interfaces with interactive functions such as clickability, navigation, and input, allowing users to simulate real business operation processes on these interfaces.

[0035] For example, using the fifth intelligent agent, based on the flowchart above which includes health disclosure, underwriting judgment, premium calculation, and payment confirmation, and combined with the page specifications in the preset specification library: the health disclosure node is converted into an interactive page containing a list of health disclosure questions and a yes / no selection button; the premium calculation node is converted into an interactive page displaying the premium amount and calculation details; and the payment confirmation node is converted into an interactive page containing a payment button and a compliance statement. The generated target product prototype contains the interactive pages connected in sequence above. Users can click on each page to complete the simulated insurance application process. The target product prototype is then sent to the first intelligent agent, which then sends it to the user.

[0036] In this embodiment, the design and generation of insurance product prototypes are automatically realized through the collaborative work of the first, second, third, fourth, and fifth intelligent agents in the insurance product prototype generation platform, improving generation efficiency. The first intelligent agent acts as a central coordinator, uniformly receiving and distributing user requests, and managing the task flow and state synchronization between intelligent agents, thereby improving the overall coordination and task execution efficiency among multiple intelligent agents. The second intelligent agent performs intent recognition to extract functional requirements, enabling subsequent intelligent agents to accurately execute downstream tasks based on these requirements without deviating from user needs. The third intelligent agent generates flowcharts using a preset knowledge graph, ensuring the professionalism of the generation process. The fourth intelligent agent performs checks using a preset rule base, ensuring the compliance and completeness of the process. The fifth intelligent agent designs the product prototype from the flowchart based on a preset specification library, ensuring the standardization and consistency of the design.

[0037] like Figure 3The diagram shown is a flowchart illustrating a product prototype generation method based on an intelligent agent according to Embodiment 3 of the present invention. In step S24, a fourth intelligent agent is used to check the generation quality of the flowchart based on a preset rule base, and the check result is obtained. This may include the following steps: S31: Using the fourth intelligent agent, based on the rules in the preset rule base, perform a compliance check on the flowchart and obtain the compliance check result.

[0038] S32: Based on the rules in the preset rule base, perform an integrity check on the flowchart and obtain the integrity check result.

[0039] S33: The results of compliance checks and integrity checks shall be considered as inspection results.

[0040] In this embodiment, the compliance check result can refer to the result of the compliance check on the flowchart, which may include the conclusion information and corresponding compliance check suggestions on whether the flowchart meets the preset industry regulations, business policies and regulatory requirements. The integrity check result can refer to the result of the integrity check on the flowchart, which may include whether there are missing branch steps, unreachable steps and isolated steps in the flowchart.

[0041] Specifically, a fourth intelligent agent performs a compliance check on the flowchart based on rules in a preset rule base, verifying whether the flow logic between each step node in the flowchart complies with the compliance requirements of the insurance industry (e.g., after health disclosure, underwriting judgment must be initiated; if underwriting fails, payment confirmation cannot proceed directly). A compliance check result is generated. Simultaneously, an integrity check is performed on the flowchart based on rules in the preset rule base, verifying whether the flowchart's topological structure is complete (e.g., whether there is a unique starting step and at least one ending step; whether each step has at least one path reachable from the starting step; whether each branch points to an existing subsequent step; whether there is an infinite loop that cannot be terminated). A integrity check result is generated. The compliance check result and the integrity check result are combined as the final check result, used for subsequent judgments on whether preset conditions are met.

[0042] Optionally, based on whether the detection and inspection results of the first intelligent agent meet preset conditions, including: Based on the first intelligent agent, it detects whether the compliance check results meet the preset compliance conditions and whether the integrity check results meet the preset integrity conditions. If the compliance check result meets the preset compliance conditions and the integrity check result meets the preset integrity conditions, then the check result is determined to meet the preset conditions.

[0043] Among them, the preset compliance conditions can refer to the criteria used to determine whether the compliance check results are qualified, that is, the compliance check results do not contain erroneous information that violates insurance industry regulations or business rules. The preset integrity conditions can refer to the criteria used to determine whether the integrity check results are qualified, that is, the integrity check results do not contain erroneous information about defects in the flowchart topology.

[0044] Specifically, if the compliance check result meets the preset compliance conditions (i.e., no compliance errors) and the integrity check result meets the preset integrity conditions (i.e., no integrity errors), then it is determined that the check result meets the preset conditions, and the first intelligent agent sends the flowchart to the fifth intelligent agent for subsequent processing.

[0045] Optionally, after checking whether the compliance check results meet the preset compliance conditions and whether the integrity check results meet the preset integrity conditions, the method further includes: If the compliance check result is found to be inconsistent with the preset compliance conditions, it is determined that the check result does not meet the preset conditions, and the compliance check result and flowchart are sent to the third intelligent agent. Using a third intelligent agent, based on the compliance check results and combined with a preset knowledge graph, the flowchart is revised to obtain a revised flowchart, which is then sent to the first intelligent agent. If the integrity check result is found to meet the preset integrity conditions, the corrected flowchart is sent to the fifth agent based on the first agent. Using a fifth intelligent agent, based on the flowchart and a pre-defined specification library, a product prototype for the target insurance application is generated, which also includes: Using a fifth intelligent agent, a product prototype for the target insurance application is generated based on the revised flowchart and a pre-defined specification library.

[0046] Specifically, after the first intelligent agent checks whether the compliance check results meet the preset compliance conditions and whether the integrity check results meet the preset integrity conditions, if it detects that the compliance check results do not meet the preset compliance conditions (i.e., the compliance check results contain erroneous information that violates insurance industry regulations or business rules), the first intelligent agent determines that the check results do not meet the preset conditions and sends the compliance error information and flowchart in the compliance check results to the third intelligent agent. Based on the compliance error information and combined with the preset knowledge graph, the third intelligent agent corrects the non-compliant step nodes or the flow relationships between nodes in the flowchart. For example, it deletes non-compliant step nodes, adjusts the execution order of nodes, supplements missing compliance nodes, or modifies flow branches that do not meet regulatory requirements, thereby generating a corrected flowchart. The corrected flowchart is then sent to the first intelligent agent. If the integrity check results meet the preset integrity conditions, the first intelligent agent sends the corrected flowchart to the fifth intelligent agent. The fifth intelligent agent then uses the corrected flowchart and the preset specification library to generate a target product prototype corresponding to the target insurance application.

[0047] Optionally, after checking whether the compliance check results meet the preset compliance conditions and whether the integrity check results meet the preset integrity conditions, the method further includes: If the integrity check result is found to be inconsistent with the preset integrity conditions, then the integrity check result and flowchart are sent to the third intelligent agent. Using a third intelligent agent, based on the integrity check results and combined with a preset knowledge graph, the flowchart is supplemented to obtain a supplemented flowchart, which is then sent to the first intelligent agent. If the compliance check results meet the preset compliance conditions, the supplemented flowchart will be sent to the fifth intelligent agent based on the first intelligent agent. Using a fifth intelligent agent, based on the flowchart and a pre-defined specification library, a product prototype for the target insurance application is generated, which also includes: Using a fifth intelligent agent, based on the supplemented flowchart and a pre-defined specification library, a product prototype for the target insurance application is generated.

[0048] Specifically, after the first agent checks whether the compliance check result meets the preset compliance conditions and whether the integrity check result meets the preset integrity conditions, if the integrity check result does not meet the preset integrity conditions (i.e., the integrity check result contains error information indicating a defect in the flowchart topology), the first agent determines that the check result does not meet the preset conditions and sends the topology defect information and the flowchart in the integrity check result to the third agent. The third agent supplements the flowchart based on the topology defect information and a preset knowledge graph. For example, for isolated nodes, it connects them to the correct execution path based on the edge relationships between the node and other nodes in the preset knowledge graph. For cases where a starting node is missing, it supplements the flowchart based on the preset knowledge graph. The knowledge graph identifies the default starting step of the process and adds the corresponding starting node. For cases where a termination node is missing, it adds the corresponding termination node based on the natural end point of the process in the preset knowledge graph. For unreachable branches, it supplements the missing intermediate nodes or adjusts the branch conditions according to the business rules in the preset knowledge graph. For infinite loops, it breaks the loop and supplements a reasonable exit path according to the process specifications in the preset knowledge graph, thereby generating a supplemented flowchart. The supplemented flowchart is then sent to the first intelligent agent. If the compliance check result meets the preset compliance conditions, the first intelligent agent sends the supplemented flowchart to the fifth intelligent agent. The fifth intelligent agent, based on the supplemented flowchart and in conjunction with the preset specification library, generates a target product prototype corresponding to the target insurance application.

[0049] Optionally, if the compliance check result does not meet the preset compliance conditions and the integrity check result does not meet the preset integrity conditions, then it is determined that the check results do not meet the preset conditions. In this case, the flowchart, compliance check result, and integrity check result are sent to a third intelligent agent. The third intelligent agent then modifies the flowchart based on the compliance check result and a preset knowledge graph to obtain a modified flowchart. Based on the integrity check result and the preset knowledge graph, the modified flowchart is supplemented to obtain a supplemented and modified flowchart. The supplemented and modified flowchart is sent to a first intelligent agent. Based on the first intelligent agent, the supplemented and modified flowchart is sent to a fifth intelligent agent. The fifth intelligent agent then generates a product prototype of the target insurance application based on the supplemented and modified flowchart and a preset specification library.

[0050] In this embodiment, when the compliance check fails to meet preset conditions, the third intelligent agent corrects the non-compliant steps or flow relationships in the flowchart based on the compliance error information, ensuring that the flowchart complies with the regulations and business rules of the insurance industry. When the integrity check fails to meet preset conditions, the third intelligent agent supplements the flowchart based on the integrity error information, solving topological defects such as isolated nodes, unreachable branches, and missing start and end nodes. This avoids regenerating the entire flowchart in a general way, reducing unnecessary computational overhead and repetitive work. At the same time, it ensures that the corrected or supplemented flowchart meets both compliance and integrity requirements, thus providing reliable input for the fifth intelligent agent to generate a high-quality and usable target product prototype.

[0051] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0052] like Figure 4 As shown in Embodiment 4, an agent-based product prototype generation device is provided. This device is applied to an insurance product prototype generation platform, which includes a first agent, a second agent, a third agent, a fourth agent, and a fifth agent. This agent-based product prototype generation device corresponds one-to-one with the agent-based product prototype generation method described in the previous embodiments. The agent-based product prototype generation device includes a receiving module 41, an intent recognition module 42, a process generation module 43, a rule checking module 44, a check and judgment module 45, and a prototype generation module 46. Detailed descriptions of each functional module are as follows: The receiving module 41 is used to receive a product prototype generation request from a user for a target insurance application from the first intelligent agent and send the product prototype generation request to the second intelligent agent. The intent recognition module 42 is used to use the second intelligent agent to perform intent recognition on the product prototype generation request, obtain the functional requirements corresponding to the product prototype generation request, send the functional requirements to the first intelligent agent, and send the functional requirements to the third intelligent agent based on the first intelligent agent. The process generation module 43 is used to use the third intelligent agent to generate a flowchart to realize the functional requirements based on the functional requirements and combined with a preset knowledge graph, send the flowchart to the first intelligent agent, and send the flowchart to the fourth intelligent agent based on the first intelligent agent. The rule checking module 44 is used to use the fourth intelligent agent to check the generation quality of the flowchart based on a preset rule base, obtain the check result, and send the check result to the first intelligent agent. The inspection and judgment module 45 is used to determine whether the inspection result meets the preset conditions based on the first intelligent agent. If the inspection result meets the preset conditions, the flowchart is sent to the fifth intelligent agent. The prototype generation module 46 is used by the fifth intelligent agent to generate a target product prototype corresponding to the target insurance application according to the flowchart and in combination with the preset specification library, send the target product prototype to the first intelligent agent, and send the target product prototype to the user based on the first intelligent agent.

[0053] Optionally, the above process generation module 43 includes: The first parsing unit is used to use the third intelligent agent to perform entity parsing on the functional requirements to obtain the requirement entity; The first matching unit is used to match the node corresponding to each demand entity from the preset knowledge graph according to the demand entity; The path identification unit is used to determine the execution path between nodes corresponding to all demand entities based on the nodes corresponding to all demand entities, the attributes of the nodes, and the edges between the corresponding nodes. The process generation unit is used to generate the flowchart based on the insurance steps represented by each node in all execution paths and the execution order relationship between the insurance steps represented by the edges between the corresponding nodes.

[0054] Optionally, the rule checking module 44 mentioned above includes: The compliance check unit is used to perform a compliance check on the flowchart based on the rules in the preset rule base using the fourth intelligent agent, and obtain the compliance check result. The integrity check unit is used to perform integrity checks on the flowchart based on the rules in the preset rule base and obtain integrity check results. A combination unit is used to combine the compliance check result and the integrity check result into the check result.

[0055] Optionally, the above-mentioned inspection and judgment module 45 includes: The condition checking unit is used to detect, based on the first intelligent agent, whether the compliance check result meets the preset compliance conditions, and whether the integrity check result meets the preset integrity conditions; The first determining unit is configured to determine that the inspection result meets the preset conditions if the compliance inspection result is detected to meet the preset compliance conditions and the integrity inspection result meets the preset integrity conditions.

[0056] Optionally, the above-mentioned inspection and judgment module 45 further includes: The second determining unit is configured to determine that the compliance check result does not meet the preset compliance conditions if the compliance check result is detected to be non-compliant, and to send the compliance check result and the flowchart to the third intelligent agent. The correction unit is used to use the third intelligent agent to correct the flowchart based on the compliance check results and the preset knowledge graph, so as to obtain a corrected flowchart, and send the corrected flowchart to the first intelligent agent. The correction sending unit is used to send the corrected flowchart to the fifth agent based on the first agent if the integrity check result is detected to meet the preset integrity condition. The aforementioned prototype generation module 46 also includes: The correction generation unit is used by the fifth intelligent agent to generate a product prototype of the target insurance application based on the corrected flowchart and a preset specification library.

[0057] Optionally, the above-mentioned inspection and judgment module 45 further includes: The third determining unit is configured to determine that the integrity check result does not meet the preset integrity condition if the integrity check result is detected to be unsatisfactory, and send the integrity check result and the flowchart to the third intelligent agent. The supplementary unit is used to supplement the flowchart using the third intelligent agent based on the integrity check result and the preset knowledge graph, to obtain a supplemented flowchart, and then send the supplemented flowchart to the first intelligent agent. A supplementary sending unit is used to send the supplemented flowchart to the fifth intelligent agent based on the first intelligent agent if the compliance check result meets the preset compliance conditions. The aforementioned prototype generation module 46 also includes: The supplementary generation unit is used by the fifth intelligent agent to generate a product prototype of the target insurance application based on the supplemented flowchart and a preset specification library.

[0058] Optionally, the above-mentioned prototype generation module 46 includes: The conversion unit is used by the fifth intelligent agent to convert the flowchart into a corresponding interactive page according to the page specifications in the preset specification library. A forming unit is used to transform the interactive page into a prototype of the target product.

[0059] Specific limitations regarding the agent-based product prototype generation device can be found in the limitations of the intelligent question-answering method described above, and will not be repeated here. Each module in the aforementioned agent-based product prototype generation device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0060] Embodiment 5 of the present invention provides a computer device, which may be a server, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, network interface, and database connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores product prototype generation requests, preset knowledge graphs, preset rule bases, and preset specification bases. The network interface communicates with external terminals via a network connection. When the computer program is executed by the processor, it implements an agent-based product prototype generation method.

[0061] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the agent-based product prototype generation method described in the above embodiments, for example... Figure 2 As shown in S21-S26, or Figure 3 As shown, to avoid repetition, it will not be described again here. Alternatively, when the processor executes a computer program, it implements the functions of each module / unit in this embodiment of the data management device, for example, Figure 4 The functions of the receiving module 41, intent recognition module 42, process generation module 43, rule checking module 44, check and judgment module 45, and prototype generation module 46 shown are not described again here to avoid repetition.

[0062] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the agent-based product prototype generation method described above, for example... Figure 2 As shown in S21-S26, or Figure 3 As shown, to avoid repetition, it will not be described again here. Alternatively, when the processor executes a computer program, it implements the functions of each module / unit in this embodiment of the data management device, for example, Figure 4The functions of the receiving module 41, intent recognition module 42, process generation module 43, rule checking module 44, check and judgment module 45, and prototype generation module 46 shown are not described again here to avoid repetition.

[0063] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0064] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0065] If any software tools or components not belonging to this company appear in the embodiments of this invention, they are merely illustrative examples and do not represent actual use. All user personal information involved in the embodiments of this invention is authorized (with knowledge and consent) by the relevant parties or fully authorized by all parties, and the executing entity may obtain it through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with the relevant laws and regulations of the relevant countries and regions, and do not violate public order and good morals.

[0066] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A product prototype generation method based on intelligent agents, characterized in that, The product prototype generation method is applied to an insurance product prototype generation platform, which includes a first intelligent agent, a second intelligent agent, a third intelligent agent, a fourth intelligent agent, and a fifth intelligent agent, comprising: Based on the first intelligent agent receiving the user's product prototype generation request for the target insurance application, the product prototype generation request is sent to the second intelligent agent. The second intelligent agent is used to perform intent recognition on the product prototype generation request, obtain the functional requirements corresponding to the product prototype generation request, send the functional requirements to the first intelligent agent, and send the functional requirements to the third intelligent agent based on the first intelligent agent. Using the third intelligent agent, based on the functional requirements and combined with a preset knowledge graph, a flowchart for implementing the functional requirements is generated, and the flowchart is sent to the first intelligent agent. Based on the first intelligent agent, the flowchart is then sent to the fourth intelligent agent. The fourth intelligent agent checks the generation quality of the flowchart based on a preset rule base, obtains the check result, and sends the check result to the first intelligent agent. Based on the first agent's determination of whether the inspection result meets the preset conditions, if the inspection result meets the preset conditions, the flowchart is sent to the fifth agent. Using the fifth intelligent agent, according to the flowchart and in conjunction with the preset specification library, a target product prototype corresponding to the target insurance application is generated, and the target product prototype is sent to the first intelligent agent. Based on the first intelligent agent, the target product prototype is sent to the user.

2. The product prototype generation method based on intelligent agents as described in claim 1, characterized in that, The step of using the third intelligent agent to generate a flowchart for implementing the functional requirements, based on the functional requirements and in conjunction with a preset knowledge graph, includes: Using the third intelligent agent, entity parsing is performed on the functional requirements to obtain the requirement entity; Based on the demand entity, nodes corresponding to each demand entity are matched from the preset knowledge graph. The nodes in the preset knowledge graph represent insurance steps, the attributes of the nodes represent the execution information of the corresponding insurance steps, and the edges between the nodes represent the execution order relationship between the corresponding insurance steps. Based on the nodes corresponding to all demand entities, the attributes of the nodes, and the edges between the corresponding nodes, determine the execution path between the nodes corresponding to all demand entities. The flowchart is generated based on the insurance steps represented by each node in all execution paths and the execution order relationship between the insurance steps represented by the edges between the corresponding nodes.

3. The product prototype generation method based on intelligent agents as described in claim 1, characterized in that, The fourth intelligent agent, based on a preset rule base, checks the generation quality of the flowchart and obtains the check results, including: Using the fourth intelligent agent, based on the rules in the preset rule base, a compliance check is performed on the flowchart to obtain the compliance check result; Based on the rules in the preset rule base, the flowchart is subjected to an integrity check to obtain the integrity check result; The compliance check results and the integrity check results are referred to as the check results.

4. The product prototype generation method based on intelligent agents as described in claim 3, characterized in that, The step of detecting whether the inspection result meets preset conditions based on the first intelligent agent includes: Based on the first intelligent agent, it is detected whether the compliance check result meets the preset compliance conditions, and whether the integrity check result meets the preset integrity conditions; If the compliance check result is found to meet the preset compliance condition, and the integrity check result meets the preset integrity condition, then it is determined that the check result meets the preset condition.

5. The product prototype generation method based on intelligent agents as described in claim 4, characterized in that, After detecting whether the compliance check result meets the preset compliance conditions and whether the integrity check result meets the preset integrity conditions, the method further includes: If the compliance check result is detected to not meet the preset compliance conditions, then it is determined that the check result does not meet the preset conditions, and the compliance check result and the flowchart are sent to the third intelligent agent; Using the third intelligent agent, based on the compliance check results and combined with the preset knowledge graph, the flowchart is modified to obtain a modified flowchart, and the modified flowchart is sent to the first intelligent agent. If the integrity check result is found to meet the preset integrity condition, the corrected flowchart is sent to the fifth agent based on the first agent; The step of using the fifth intelligent agent to generate a product prototype of the target insurance application based on the flowchart and a preset specification library also includes: Using the fifth intelligent agent, a product prototype of the target insurance application is generated based on the revised flowchart and a preset specification library.

6. The product prototype generation method based on intelligent agents as described in claim 4, characterized in that, After detecting whether the compliance check result meets the preset compliance conditions and whether the integrity check result meets the preset integrity conditions, the method further includes: If the integrity check result is detected to not meet the preset integrity condition, then it is determined that the check result does not meet the preset condition, and the integrity check result and the flowchart are sent to the third intelligent agent; Using the third intelligent agent, based on the integrity check result and combined with the preset knowledge graph, the flowchart is supplemented to obtain a supplemented flowchart, and the supplemented flowchart is sent to the first intelligent agent. If the compliance check result meets the preset compliance conditions, the supplemented flowchart is sent to the fifth intelligent agent based on the first intelligent agent; The step of using the fifth intelligent agent to generate a product prototype of the target insurance application based on the flowchart and a preset specification library also includes: Using the fifth intelligent agent, based on the supplemented flowchart and in conjunction with a preset specification library, a product prototype of the target insurance application is generated.

7. The product prototype generation method based on intelligent agents as described in claim 1, characterized in that, The step of using the fifth intelligent agent to generate a target product prototype corresponding to the target insurance application, based on the flowchart and a preset specification library, includes: Using the fifth intelligent agent, the flowchart is converted into a corresponding interactive page according to the page specifications in the preset specification library; The interactive page is the prototype of the target product.

8. A product prototype generation device based on intelligent agents, characterized in that, The product prototype generation device is applied to an insurance product prototype generation platform, which includes a first intelligent agent, a second intelligent agent, a third intelligent agent, a fourth intelligent agent, and a fifth intelligent agent, comprising: The receiving module is used to receive a product prototype generation request from a user for a target insurance application from the first intelligent agent and send the product prototype generation request to the second intelligent agent. The intent recognition module is used to use the second intelligent agent to perform intent recognition on the product prototype generation request, obtain the functional requirements corresponding to the product prototype generation request, send the functional requirements to the first intelligent agent, and send the functional requirements to the third intelligent agent based on the first intelligent agent. The process generation module is used to use the third intelligent agent to generate a flowchart to implement the functional requirements based on the functional requirements and combined with a preset knowledge graph, send the flowchart to the first intelligent agent, and send the flowchart to the fourth intelligent agent based on the first intelligent agent. The rule checking module is used to use the fourth intelligent agent to check the generation quality of the flowchart based on a preset rule base, obtain the check result, and send the check result to the first intelligent agent. The inspection and judgment module is used to determine whether the inspection result meets the preset conditions based on the first intelligent agent. If the inspection result meets the preset conditions, the flowchart is sent to the fifth intelligent agent. The prototype generation module is used by the fifth intelligent agent to generate a target product prototype corresponding to the target insurance application according to the flowchart and in conjunction with a preset specification library, and sends the target product prototype to the first intelligent agent, and sends the target product prototype to the user based on the first intelligent agent.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the agent-based product prototype generation method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the agent-based product prototype generation method as described in any one of claims 1 to 7.

Images