Automatic business process generation method and device based on fuzzy demand semantic analysis

Abstract

The invention discloses an automatic business process generation method and device based on fuzzy demand semantic analysis, and the method comprises the steps: carrying out the processing of a natural language business demand through a preset demand understanding agent, and improving the accurate extraction capability of the natural language business demand from a structured process instruction; moreover, a multi-round context memory mechanism is introduced, the multi-round dialogue history and the process construction state of the user are continuously tracked, dynamic iteration and continuous support of process generation are realized, and the response capability of the system to complex multi-round business requirements is enhanced; besides, through multi-agent cooperative driving, an end-to-end automatic closed loop is formed from semantic analysis to process construction, technical mapping and deployment execution, and task scheduling and abnormity self-repairing are realized in combination with a preset control and scheduling agent, so that the intelligent level and robustness of the business process generation system are remarkably improved, and the service process generation efficiency is improved. Dynamic changes and various expressions of business requirements can be met, so that the flexibility and accuracy of business process generation are improved.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to an automated business process generation method and apparatus based on fuzzy demand semantic parsing. Background Technology

[0002] As enterprises accelerate their digitalization and intelligentization processes, business process requirements are increasingly expressed in a more natural and unstructured manner. Users often use colloquial descriptions and a mix of text and graphics to express business goals and operational logic, resulting in highly nested and ambiguous language that poses a significant challenge to system understanding.

[0003] Currently, keyword extraction or template matching methods struggle to accurately identify task intent, conditional relationships, and triggering events. Meanwhile, the types of input data required for business processes are becoming increasingly diverse, evolving into a multimodal, multi-source, and heterogeneous structure. In addition to textual information, voice and images have become important sources of process triggering and status awareness, originating from different terminals, systems, or business platforms.

[0004] Therefore, traditional automated business process generation methods are unable to meet the dynamic changes and diverse expressions of business needs, and suffer from poor flexibility and low accuracy. Summary of the Invention

[0005] This application aims to at least solve the technical problems existing in the prior art. To this end, the first aspect of this application proposes an automated business process generation method based on fuzzy demand semantic parsing, the method comprising: By understanding the pre-defined requirements, the intelligent agent processes the received natural language business requirements and generates structured process instructions corresponding to the natural language business requirements. The structured process instructions are processed by a pre-set intent recognition agent and a pre-set context memory agent to generate intent recognition results. The intent recognition results include business objectives, triggering conditions, key nodes and output requirements. By constructing an intelligent agent through a pre-defined process, the flowchart generation engine is invoked to process the structured process instructions and intent recognition results, generating a preliminary process model. By using a pre-defined software flow mapping agent, target nodes in the initial process model are bound to target software technology components and target data structures, generating configuration files; By pre-deploying intelligent agents based on configuration files, the target business process is deployed to the target execution environment for execution status monitoring and resource scheduling.

[0006] In one possible implementation, a pre-defined demand understanding agent processes the received natural language business requirements to generate structured process instructions corresponding to the natural language business requirements, including: By understanding the pre-defined requirements, the intelligent agent invokes a pre-defined large language model to perform contextual semantic parsing on the natural language business requirements, and extracts a set of candidate semantic units; the set of candidate semantic units includes multiple candidate semantic units. Based on a semantic recognizer enhanced by a pre-defined rule template and a dictionary, semantic annotation is performed on each candidate semantic unit, and typified intermediate results are output. The typified intermediate results include tasks, process nodes, constraints, and triggering events. Based on a pre-defined domain knowledge base, the typified intermediate results are standardized and confirmed to generate structured process instructions.

[0007] In one possible implementation, a pre-defined demand understanding agent invokes a pre-defined large language model to perform contextual semantic parsing of natural language business requirements, extracting a set of candidate semantic units, including: Obtain modal information corresponding to natural language business requirements; wherein, modal information includes at least one of text, tables, images, and speech; By understanding the preset requirements, the intelligent agent calls a preset large language model to determine the target modal information processing method corresponding to the modal information; Based on the target modality information processing method, the contextual semantics of natural language business requirements are parsed to extract a set of candidate semantic units.

[0008] In one possible implementation, the typified intermediate results are standardized and verified based on a preset domain knowledge base to generate structured process instructions, including: The candidate entities in the typed intermediate results are encoded into candidate entity vectors by a preset semantic encoder, and each preset term entity in the preset domain knowledge base is encoded into a standard term vector. The similarity between the candidate entity vectors and the standard term vectors is calculated to obtain a list of similarity scores; Based on the similarity score list, the target term vector that matches the candidate entity vector is determined from each standard term vector; The typified intermediate results are standardized and confirmed based on the current similarity score between the candidate entity vector and the target term vector, and structured process instructions are generated.

[0009] In one possible implementation, the typified intermediate results are standardized and confirmed based on the current similarity score between the candidate entity vector and the target term vector, generating structured process instructions, including: Obtain the preset similarity threshold; If the current similarity score is greater than the preset similarity threshold, the candidate entity corresponding to the candidate entity vector is replaced by the preset term entity corresponding to the target term vector, and a structured process instruction is generated. If the current similarity score is less than or equal to the preset similarity threshold, the preset term entity corresponding to the target term vector will be marked as an ambiguous item, and a structured process instruction will be generated.

[0010] In one possible implementation, a pre-defined process is used to construct an intelligent agent that invokes a flowchart generation engine to process structured process instructions and intent recognition results, generating a preliminary process model, including: The intelligent agent is constructed by calling the flowchart through a pre-defined process to build a domain-specific language description file; the domain-specific language description file includes various types of control structures, including approval, condition, parallel, and rollback. The structured process instructions and intent recognition results are processed based on domain-specific language description files to generate graphical flowcharts. The graphical flowchart can be edited through a preset visual editing interface to generate a preliminary process model.

[0011] In one possible implementation, the method further includes: The system uses preset control and scheduling agents to coordinate and handle exceptions for other agents, and updates the user context memory of the preset context memory agent. Other agents include preset demand understanding agents, preset intent recognition agents, preset context memory agents, preset process construction agents, preset software flow mapping agents, and preset deployment agents. The scheduling and coordination includes managing the execution dependencies and scheduling order among other agents, and the exception handling includes detecting task exceptions of other agents and triggering preset error recovery logic corresponding to the task exceptions.

[0012] The second aspect of this application proposes an automated business process generation device based on fuzzy demand semantic parsing, the device comprising: The requirement understanding module is used to process the received natural language business requirements through a preset requirement understanding intelligent agent and generate structured process instructions corresponding to the natural language business requirements. The intent recognition module is used to process structured process instructions through multi-turn dialogue context provided by a preset intent recognition agent and a preset context memory agent, and generate intent recognition results; wherein, the intent recognition results include business objectives, triggering conditions, key nodes and output requirements; The process construction module is used to call the flowchart generation engine to process the structured process instructions and intent recognition results through the preset process construction intelligent agent, and generate a preliminary process model. The software flow mapping module is used to bind target nodes in the preliminary process model with target software technology components and target data structures through a preset software flow mapping agent, and generate configuration files. The deployment module is used to deploy target business processes to the target execution environment based on configuration files using a pre-defined deployment agent, in order to monitor execution status and schedule resources.

[0013] In one possible implementation, the aforementioned requirement understanding module is specifically used for: By understanding the pre-defined requirements, the intelligent agent invokes a pre-defined large language model to perform contextual semantic parsing on the natural language business requirements, and extracts a set of candidate semantic units; the set of candidate semantic units includes multiple candidate semantic units. Based on a semantic recognizer enhanced by a pre-defined rule template and a dictionary, semantic annotation is performed on each candidate semantic unit, and typified intermediate results are output. The typified intermediate results include tasks, process nodes, constraints, and triggering events. Based on a pre-defined domain knowledge base, the typified intermediate results are standardized and confirmed to generate structured process instructions.

[0014] In one possible implementation, the aforementioned requirement understanding module is further used for: Obtain modal information corresponding to natural language business requirements; wherein, modal information includes at least one of text, tables, images, and speech; By understanding the preset requirements, the intelligent agent calls a preset large language model to determine the target modal information processing method corresponding to the modal information; Based on the target modality information processing method, the contextual semantics of natural language business requirements are parsed to extract a set of candidate semantic units.

[0015] In one possible implementation, the aforementioned requirement understanding module is further used for: The candidate entities in the typed intermediate results are encoded into candidate entity vectors by a preset semantic encoder, and each preset term entity in the preset domain knowledge base is encoded into a standard term vector. The similarity between the candidate entity vectors and the standard term vectors is calculated to obtain a list of similarity scores; Based on the similarity score list, the target term vector that matches the candidate entity vector is determined from each standard term vector; The typified intermediate results are standardized and confirmed based on the current similarity score between the candidate entity vector and the target term vector, and structured process instructions are generated.

[0016] In one possible implementation, the aforementioned requirement understanding module is further used for: Obtain the preset similarity threshold; If the current similarity score is greater than the preset similarity threshold, the candidate entity corresponding to the candidate entity vector is replaced by the preset term entity corresponding to the target term vector, and a structured process instruction is generated. If the current similarity score is less than or equal to the preset similarity threshold, the preset term entity corresponding to the target term vector will be marked as an ambiguous item, and a structured process instruction will be generated.

[0017] In one possible implementation, the above process construction module is specifically used for: The intelligent agent is constructed by calling the flowchart through a pre-defined process to build a domain-specific language description file; the domain-specific language description file includes various types of control structures, including approval, condition, parallel, and rollback. The structured process instructions and intent recognition results are processed based on domain-specific language description files to generate graphical flowcharts. The graphical flowchart can be edited through a preset visual editing interface to generate a preliminary process model.

[0018] In one possible implementation, the aforementioned automated business process generation device based on fuzzy demand semantic parsing is further used for: The system uses preset control and scheduling agents to coordinate and handle exceptions for other agents, and updates the user context memory of the preset context memory agent. Other agents include preset demand understanding agents, preset intent recognition agents, preset context memory agents, preset process construction agents, preset software flow mapping agents, and preset deployment agents. The scheduling and coordination includes managing the execution dependencies and scheduling order among other agents, and the exception handling includes detecting task exceptions of other agents and triggering preset error recovery logic corresponding to the task exceptions.

[0019] A third aspect of this application provides an electronic device comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, a code set, or an instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by the processor to implement the automated business process generation method based on fuzzy demand semantic parsing as described in the first aspect.

[0020] The fourth aspect of this application proposes a computer-readable storage medium storing at least one instruction, at least one program, code set, or instruction set, wherein the at least one instruction, the at least one program, the code set, or the instruction set is loaded and executed by a processor to implement the automated business process generation method based on fuzzy demand semantic parsing as described in the first aspect.

[0021] The embodiments of this application have the following beneficial effects: This application provides an automated business process generation method and apparatus based on fuzzy requirement semantic parsing. The method includes: processing received natural language business requirements through a preset requirement understanding agent to generate structured process instructions corresponding to the natural language business requirements; processing the structured process instructions through a multi-turn dialogue context provided by a preset intent recognition agent and a preset context memory agent to generate intent recognition results; processing the structured process instructions and intent recognition results through a preset process construction agent by calling a flowchart generation engine to generate a preliminary process model; binding target nodes in the preliminary process model with target software technology components and target data structures through a preset software flow mapping agent to generate a configuration file; and deploying the target business process to the target execution environment based on the configuration file through a preset deployment agent for execution status monitoring and resource scheduling. This solution enhances the accuracy of extracting natural language business requirements from pre-defined demand-understanding agents, thereby improving the system's ability to extract structured process instructions. Furthermore, it introduces a multi-turn context memory mechanism to continuously track the user's multi-turn dialogue history and process construction status, enabling dynamic iteration and continuity support for process generation and enhancing the system's responsiveness to complex multi-turn business requirements. In addition, through multi-agent collaborative driving, an end-to-end automated closed loop is formed from semantic parsing to process construction, technology mapping, and deployment execution. Combined with pre-defined control and scheduling agents, task scheduling and anomaly self-healing are achieved, significantly improving the intelligence level and robustness of the agent-driven business process generation system based on fuzzy demand semantic parsing. This allows the system to meet the dynamic changes and diverse expressions of business requirements, improving the flexibility and accuracy of business process generation. Moreover, non-technical users can directly describe business process requirements in natural language, and the system automatically completes process design, component mapping, and deployment, achieving "zero-code" generation and execution of business processes, significantly reducing the modeling threshold and improving the degree of business automation. Attached Figure Description

[0022] Figure 1 A block diagram of a computer device provided in an embodiment of this application; Figure 2 A flowchart illustrating the steps of an automated business process generation method based on fuzzy demand semantic parsing, provided in an embodiment of this application; Figure 3 A flowchart illustrating the steps for generating structured process instructions is provided in an embodiment of this application. Figure 4 A flowchart illustrating the steps for extracting a set of candidate semantic units is provided in this application embodiment; Figure 5 A flowchart illustrating another step in generating structured process instructions provided in this application embodiment; Figure 6A flowchart illustrating another step in generating structured process instructions provided in this application embodiment; Figure 7 A flowchart illustrating the steps for generating a preliminary process model is provided in this embodiment of the application. Figure 8 This is a structural block diagram of an automated business process generation device based on fuzzy demand semantic parsing provided in an embodiment of this application. Detailed Implementation

[0023] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0024] As enterprises accelerate their digitalization and intelligentization processes, business process requirements are increasingly expressed in a more natural and unstructured manner. Users often use colloquial descriptions and mixed text and graphics to express business goals and operational logic, resulting in highly nested and ambiguous language that poses significant challenges to system understanding. Traditional keyword extraction or template matching methods struggle to accurately identify task intent, conditional relationships, and triggering events, necessitating stronger semantic parsing capabilities to support high-precision process modeling. Simultaneously, the types of data input for business process requirements are becoming increasingly diverse, evolving into multimodal and heterogeneous structures. Besides textual information, voice and images have become important sources of process triggering and status awareness, originating from different terminals, systems, or business platforms. This requires systems to possess cross-modal semantic alignment and fusion capabilities, enabling unified data representation, event parsing, and process response to adapt to the complex information flows in real-world business scenarios.

[0025] In the process of building and deploying traditional business processes, developers and enterprises generally face the following challenges: (1) Natural language requirements are vague and lack structured capabilities. Business personnel usually describe process goals and rules in free language, which has a complex semantic structure and high uncertainty. Existing systems mostly rely on shallow methods such as keyword matching and rule engines for parsing, which makes it difficult to identify key elements such as task actions, triggering conditions, and constraint logic, resulting in low system understanding accuracy and high misjudgment rate.

[0026] (2) Process modeling heavily relies on manual participation, resulting in a high development threshold. Most mainstream process platforms adopt visual modeling and explicit configuration methods, requiring engineers with development experience to manually draw flowcharts, configure task nodes and execution logic. Non-technical users find it difficult to directly participate in process creation and maintenance, leading to long process construction cycles and high costs for cross-team collaboration.

[0027] (3) The process automation link is difficult to close the loop and lacks intelligent deployment and evolution capabilities. The current system lacks unified semantic support and intelligent linkage mechanism in each stage such as requirement acquisition, process modeling, component configuration, deployment and execution. Version switching, context migration and dynamic adjustment in process operation still require manual intervention, which cannot achieve end-to-end automation and seriously affects iteration efficiency and system maintainability.

[0028] (4) Insufficient support for multilingual and multimodal business input. In actual business, user input may include multimodal data such as mixed Chinese and English expressions, voice commands, and image materials. Existing systems have limited ability to process unstructured input and lack a unified semantic understanding framework and modality fusion mechanism, resulting in limited business scope and weak adaptability.

[0029] (5) The current automation platform relies on static rule templates for process scheduling and decision-making, which is slow to respond to dynamic environmental changes and lacks an intelligent decision-making mechanism based on context awareness and task dependency modeling. It is difficult to support the abnormal handling, branch jump and collaborative execution of complex processes.

[0030] In view of this, this application proposes an automated business process generation method based on fuzzy requirement semantic parsing. By using a pre-set requirement understanding agent to process natural language business requirements, the method improves the accuracy of extracting natural language business requirements into structured process instructions. Furthermore, a multi-turn context memory mechanism is introduced to continuously track the user's multi-turn dialogue history and process construction status, enabling dynamic iteration and continuity support for process generation and enhancing the system's responsiveness to complex multi-turn business requirements. In addition, through multi-agent collaborative driving, an end-to-end automated closed loop is formed from semantic parsing to process construction, technology mapping, and deployment execution. Combined with a pre-set control and scheduling agent, task scheduling and anomaly self-healing are achieved, significantly improving the intelligence level and robustness of the agent-driven business process generation system based on fuzzy requirement semantic parsing. This allows the system to meet the dynamic changes and diverse expressions of business requirements, improving the flexibility and accuracy of business process generation. Moreover, non-technical users can directly describe business process requirements in natural language, and the system automatically completes process design, component mapping, and deployment, achieving "zero-code" generation and execution of business processes, significantly reducing the modeling threshold and improving the degree of business automation.

[0031] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of embodiments of this disclosure, unless otherwise stated, "a plurality of" means two or more. Furthermore, the use of "based on" or "according to" implies openness and inclusiveness, because processes, steps, calculations, or other actions "based on" or "according to" one or more of the stated conditions or values ​​may in practice be based on additional conditions or beyond the stated values.

[0032] The automated business process generation method based on fuzzy demand semantic parsing provided in this application can be applied to computer equipment (electronic devices). The computer equipment can be a server or a terminal. The server can be a single server or a server cluster composed of multiple servers. This application does not specifically limit this. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices.

[0033] Taking a computer device as an example, Figure 1 A block diagram of a server is shown, such as Figure 1 As shown, the server may include a processor and memory connected via a system bus. The processor provides computing 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 a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. When the computer program is executed by the processor, it implements an automated business process generation method based on fuzzy demand semantic parsing.

[0034] Those skilled in the art will understand that Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the server to which the present application is applied. Optionally, the server may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0035] It should be noted that the execution subject of the embodiments of this application can be a computer device or an automated business process generation device based on fuzzy demand semantic parsing. The following method embodiments will be described with computer device as the execution subject.

[0036] Figure 2 This document provides a flowchart illustrating the steps of an automated business process generation method based on fuzzy requirement semantic parsing, as illustrated in an embodiment of this application. Figure 2As shown, the method includes the following steps: Step 202: The intelligent agent, based on preset requirements, processes the received natural language business requirements and generates structured process instructions corresponding to the natural language business requirements.

[0037] Users can input natural language business requirements through a preset visual editing interface, which supports input in multiple languages ​​and modalities. Natural language business requirements can include business objectives, constraints, etc. Optionally, the preset visual editing interface can also support user identification, input validation, context state maintenance, and multi-turn dialogue management to ensure the continuity and completeness of input.

[0038] After obtaining the user's input of natural language business requirements, a pre-defined requirement-understanding agent can process the received natural language business requirements, thereby generating structured process instructions corresponding to those requirements. In some optional embodiments, such as... Figure 3 As shown, Figure 3 A flowchart illustrating the steps for generating structured process instructions provided in this application embodiment includes: Step 302: By understanding the preset requirements, the intelligent agent calls the preset large language model to perform contextual semantic parsing on the natural language business requirements and extracts a set of candidate semantic units.

[0039] Step 304: Based on the preset rule template and dictionary-enhanced semantic recognizer, perform semantic annotation on each candidate semantic unit and output typified intermediate results.

[0040] Step 306: Standardize and confirm the typified intermediate results based on the preset domain knowledge base, and generate structured process instructions.

[0041] The candidate semantic unit set includes multiple candidate semantic units, and the pre-defined large language model can be denoted as... The input natural language business requirements are denoted as The resulting set of candidate semantic units can be denoted as , ,in, , As candidate semantic units, i The value range is from 1 to n.

[0042] In some alternative embodiments, such as Figure 4 As shown, Figure 4 A flowchart illustrating the steps for extracting a set of candidate semantic units, provided in this application embodiment, includes: Step 402: Obtain modal information corresponding to natural language business requirements.

[0043] Step 404: Understand the intelligent agent by calling the preset large language model to determine the target modal information processing method corresponding to the modal information through preset requirements.

[0044] Step 406: Perform contextual semantic parsing on natural language business requirements based on the target modality information processing method to extract a set of candidate semantic units.

[0045] The modal information can include at least one of text, tables, images, and speech. Users can input natural language business requirements through a preset visual editing interface that supports multiple languages. For example, the natural language business requirement can be "I want to apply for a database server for production environment use, and I need to enable external network access." This allows the intelligent agent to determine the target modality information processing method corresponding to the modal information by invoking a pre-defined large language model based on preset requirements. Then, contextual semantic parsing is performed on the natural language business requirements based on the target modality information processing method to extract a set of candidate semantic units. Specifically, target modality information processing methods corresponding to different modalities of natural language business requirements, such as text, tables, images, and speech, can be pre-set. By supporting semantic parsing and process construction in multilingual environments and integrating multimodal information processing capabilities, the system meets the business needs of cross-language and multi-type data input, improving its versatility and application breadth.

[0046] Next, a semantic recognizer enhanced based on preset rule templates and a dictionary can be used. For the candidate semantic unit set Semantic annotation is performed on each candidate semantic unit in the dataset to output typified intermediate results. The categorized intermediate results include tasks, process nodes, constraints, and triggering events.

[0047] Specifically, ,in, .

[0048] For example, if the user's natural language business requirement is: "After a customer completes registration, the system needs to send a welcome email and initiate a 7-day free trial," then the typed intermediate output may include: { "Task": ["Send a welcome email", "Initiate the free trial process"], "Process Nodes": ["Registration Completed", "Email Sent", "Trial Process Started"], "Constraints": ["Customer completes registration"], Triggered event: ["Registration successful"] } After obtaining the typified intermediate results, they can be standardized and validated based on a preset domain knowledge base to generate structured process instructions. In some optional embodiments, such as... Figure 5 As shown, Figure 5 A flowchart illustrating another step in generating structured process instructions provided in this application embodiment includes: Step 502: Encode the candidate entities in the typed intermediate results into candidate entity vectors using a preset semantic encoder, and encode each preset term entity in the preset domain knowledge base into a standard term vector.

[0049] Step 504: Calculate the similarity between the candidate entity vector and the standard term vector to obtain a list of similarity scores.

[0050] Step 506: Based on the similarity score list, determine the target term vector that matches the candidate entity vector from each standard term vector.

[0051] Step 508: Standardize and confirm the typified intermediate results based on the current similarity score between the candidate entity vector and the target term vector, and generate structured process instructions.

[0052] In this context, candidate entities in the typed intermediate results are denoted as... Each preset term entity in the preset domain knowledge base is denoted as The predefined semantic encoder can be a bidirectional encoder representation from Transformers (BERT). This predefined semantic encoder can encode candidate entities and predefined term entities into high-dimensional semantic vectors, thus obtaining the candidate entity vector. Vector of Standard Terminology .

[0053] Next, for each standard term vector, the similarity between the candidate entity vector and the standard term vector can be calculated, resulting in a similarity score list composed of multiple current similarity scores. Current similarity score It can be calculated using formula (1).

[0054] (1) This allows for the determination of target term vectors that match candidate entity vectors from a list of standard term vectors, based on a similarity score. Finally, the typified intermediate results are standardized and confirmed based on the current similarity score between the candidate entity vectors and the target term vectors, generating structured process instructions.

[0055] In some alternative embodiments, such as Figure 6As shown, Figure 6 Another flowchart for generating structured process instructions provided in this application embodiment includes: Step 602: Obtain the preset similarity threshold.

[0056] Step 604: If the current similarity score is greater than the preset similarity threshold, the candidate entity corresponding to the candidate entity vector is replaced by the preset term entity corresponding to the target term vector, and a structured process instruction is generated.

[0057] Step 606: If the current similarity score is less than or equal to the preset similarity threshold, then the preset term entity corresponding to the target term vector is marked as an ambiguous item, and a structured process instruction is generated.

[0058] Among them, the preset similarity threshold The settings can be customized in advance. Optionally, they can be adaptively adjusted according to the complexity of the corpus and the tolerance of business. For example, the preset similarity threshold can be set between 0.75 and 0.9.

[0059] If the current similarity score is greater than the preset similarity threshold, the match is considered valid. The candidate entity corresponding to the candidate entity vector can be replaced by the preset term entity corresponding to the target term vector, and a structured process instruction is generated. If the current similarity score is less than or equal to the preset similarity threshold, the preset term entity corresponding to the target term vector is marked as an ambiguous item, and a subsequent manual confirmation process is triggered before generating a structured process instruction.

[0060] For example, in an enterprise's IT service management scenario, employees frequently need to initiate service processes such as "activate VPN," "request database," and "allocate virtual machine." These processes typically involve multi-department approvals, security compliance verification, automated execution, and logging. After processing, the following structured process instructions can be obtained: { "flow_name": "Database resource application process", "actors": ["employees", "line managers", "IT security", "automation systems"], "steps": [ {"name": "Submitted Application", "actor": "Employee"}, {"name": "Manager Approval", "actor": "Line Manager"}, { "name": "Security Review", "actor": "IT security", "condition": "The resource is in a production environment or involves external network access" }, { "name": "Resource Allocation", "actor": "automation system", "action": "Execute database deployment script" }, { "name": "Notifications and Records", "actor": "system", "action": "Send deployment notification + write operation log" } ], "constraints": { "auto_notify": true, "record_all_steps": true } } In this embodiment, by combining contextual semantic understanding of a large language model, semantic recognition based on rules and entity dictionaries, and standardized verification of a domain knowledge base, high-precision extraction of business process tasks, nodes, constraints, and events is achieved, greatly enhancing the accuracy and professionalism of structured instructions.

[0061] Step 204: Process the structured process instructions using the multi-turn dialogue context provided by the preset intent recognition agent and the preset context memory agent to generate intent recognition results.

[0062] The intent recognition result includes business objectives, triggering conditions, key nodes, and output requirements, which may include format requirements for the output content. A pre-defined context memory agent can model the context of the current input. For example, this process may include: identifying the current user as an "IT process administrator" with process publishing permissions; determining the conversation context as "database service automation application"; loading the user's last three historical application process templates as supplementary input prompts; and constructing structured context prompts to guide the pre-defined large language model in generating stable output.

[0063] Optionally, the pre-defined context memory agent can continuously record user multi-turn inputs, interaction history, intermediate states of process construction, and interaction data between agents. It also maintains a vector database and context cache to achieve multi-turn conversation association and semantic consistency maintenance. Furthermore, it can provide real-time context query and historical data support for the pre-defined requirement understanding agent and the pre-defined intent recognition agent, assisting in business process optimization and iterative generation. By introducing the pre-defined context memory agent, continuous maintenance and dynamic updates of multi-turn dialogue states and task history are achieved, supporting requirement continuation and process reuse in complex business scenarios. This effectively overcomes the limitations of traditional one-time business process generation, improving the system's adaptability and intelligence level.

[0064] Next, the structured process instructions can be processed using the multi-turn dialogue context provided by the preset intent recognition agent and the preset context memory agent to generate intent recognition results. Specifically, the preset intent recognition agent can combine the multi-turn dialogue context provided by the context memory agent to perform deep semantic understanding and functional decomposition, thereby identifying business objectives, triggering conditions, key nodes, and output requirements, forming a complete business semantic structure, and finally outputting a complete intent recognition result for further processing by the preset process construction agent.

[0065] Step 206: The intelligent agent is constructed through a preset process. The flowchart generation engine is called to process the structured process instructions and intent recognition results to generate a preliminary process model.

[0066] Among them, the pre-defined process construction agent can generate a preliminary process model based on structured process instructions and intent recognition results. In some optional embodiments, such as... Figure 7 As shown, Figure 7 A flowchart illustrating the steps for generating a preliminary process model, as provided in this application embodiment, includes: Step 702: Construct an intelligent agent by calling a flowchart through a preset process to build a domain-specific language description file.

[0067] Step 704: Process the structured process instructions and intent recognition results based on the domain-specific language description file to generate a graphical flowchart.

[0068] Step 706: Edit the graphical flowchart using the preset visual editing interface to generate a preliminary process model.

[0069] Specifically, a pre-defined process can be used to construct an intelligent agent that invokes a flowchart to build a Domain Specific Language (DSL) description file. This DSL description file includes various types of control structures, including but not limited to approval, conditional, parallel, and rollback structures. Based on this DSL description file, structured process instructions and intent recognition results can be processed to generate a graphical flowchart, which supports interactive modification.

[0070] Next, the graphical flowchart can be edited through a preset visual editing interface to generate a preliminary process model. Optionally, users can modify the graphical flowchart by changing the node order, branch conditions, and executor types through the preset visual editing interface. The system can automatically update the user's editing operations to ultimately obtain the preliminary process model.

[0071] In this embodiment, a process modeling and visualization platform based on a domain-specific language is used to support interactive process editing, parameter adjustment, and exception rollback. Through a closed-loop user feedback mechanism, process optimization is driven, enabling human-machine collaborative construction and intelligent iteration, which greatly improves the user experience and process execution quality.

[0072] Step 208: Bind the target nodes in the preliminary process model to the target software technology components and target data structures through the preset software flow mapping agent, and generate a configuration file.

[0073] Specifically, the pre-defined software flow mapping agent can automatically and accurately map target sections in the abstract preliminary process model to target software technology components and target data structures based on a pre-defined semantic matching algorithm. It then binds these sections to the corresponding target software technology components and target data structures, thereby automatically generating configuration files that conform to the target runtime environment and automatically performing syntax and compatibility checks. Furthermore, it supports user-defined parameters for target software technology components, performs runtime environment adaptation testing, and dynamically adjusts the configuration to meet platform requirements.

[0074] Step 210: Deploy the target business process to the target execution environment using a pre-defined deployment agent based on the configuration file, for execution status monitoring and resource scheduling.

[0075] Among them, the pre-deployed intelligent agent can automatically complete the software flow deployment by relying on containerization technology, deploy the target business process to the pre-deployed process orchestration platform, and achieve end-to-end integration. That is, the target business process can be deployed to the target execution environment based on the configuration file, so as to monitor the execution status and schedule resources.

[0076] In addition, it can perform automated integration testing and log monitoring, tracking process status and anomalies in real time. It also leverages a pre-set intelligent anomaly diagnosis module to trigger self-repair or rollback mechanisms, ensuring high availability and stability of the process. Furthermore, it can feed back deployment results, execution status, and anomaly information to a pre-set control and scheduling agent, supporting real-time display on the user interface.

[0077] Optionally, the system can collect the execution logs of the target business process in real time and display the execution status. If the deployment of the target business process fails, such as due to insufficient resources or script exceptions, the system can automatically trigger a rollback process and notify the user of the error details.

[0078] In some optional embodiments, preset control and scheduling agents can be used to coordinate and handle exceptions for other agents, and to update the user context memory of the preset context memory agent. These other agents may include preset demand understanding agents, preset intent recognition agents, preset context memory agents, preset process construction agents, preset software flow mapping agents, and preset deployment agents. Coordination includes managing execution dependencies and scheduling order among other agents, and exception handling includes detecting task anomalies in other agents and triggering preset error recovery logic corresponding to the anomalies.

[0079] Specifically, the preset control and scheduling agent plays a core role in the entire system. It can manage the execution dependencies and scheduling order among various agents; dynamically detect task failures, interruptions, or inconsistencies in state; trigger preset error recovery logic, such as model switching, prompt word rewriting, and process rollback; and update user context memory to achieve long-term knowledge retention.

[0080] In this embodiment, the method includes: processing received natural language business requirements through a preset requirement understanding agent to generate structured process instructions corresponding to the natural language business requirements; processing the structured process instructions through a preset intent recognition agent and a preset context memory agent providing multi-turn dialogue context to generate intent recognition results; processing the structured process instructions and intent recognition results through a preset process construction agent calling a flowchart generation engine to generate a preliminary process model; binding target nodes in the preliminary process model with target software technology components and target data structures through a preset software flow mapping agent to generate a configuration file; and deploying the target business process to the target execution environment based on the configuration file through a preset deployment agent for execution status monitoring and resource scheduling. This solution enhances the accuracy of extracting natural language business requirements from pre-defined demand-understanding agents, thereby improving the system's ability to extract structured process instructions. Furthermore, it introduces a multi-turn context memory mechanism to continuously track the user's multi-turn dialogue history and process construction status, enabling dynamic iteration and continuity support for process generation and enhancing the system's responsiveness to complex multi-turn business requirements. In addition, through multi-agent collaborative driving, an end-to-end automated closed loop is formed from semantic parsing to process construction, technology mapping, and deployment execution. Combined with pre-defined control and scheduling agents, task scheduling and anomaly self-healing are achieved, significantly improving the intelligence level and robustness of the agent-driven business process generation system based on fuzzy demand semantic parsing. This allows the system to meet the dynamic changes and diverse expressions of business requirements, improving the flexibility and accuracy of business process generation. Moreover, non-technical users can directly describe business process requirements in natural language, and the system automatically completes process design, component mapping, and deployment, achieving "zero-code" generation and execution of business processes, significantly reducing the modeling threshold and improving the degree of business automation.

[0081] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0082] Figure 8 This is a structural block diagram of an automated business process generation device based on fuzzy demand semantic parsing, provided in an embodiment of this application.

[0083] like Figure 8 As shown, the automated business process generation device 800 based on fuzzy demand semantic parsing includes: The requirement understanding module 802 is used to process the received natural language business requirements through a preset requirement understanding intelligent agent and generate structured process instructions corresponding to the natural language business requirements.

[0084] The intent recognition module 804 is used to process the structured process instructions through a multi-turn dialogue context provided by a preset intent recognition agent and a preset context memory agent, and generate intent recognition results; wherein, the intent recognition results include business objectives, triggering conditions, key nodes and output requirements.

[0085] The process construction module 806 is used to call the flowchart generation engine through a preset process construction agent to process the structured process instructions and the intent recognition results to generate a preliminary process model.

[0086] The software flow mapping module 808 is used to bind the target nodes, target software technology components, and target data structures in the preliminary process model through a preset software flow mapping intelligent agent, and generate a configuration file.

[0087] The deployment module 810 is used to deploy the target business process to the target execution environment based on the configuration file through a preset deployment agent, so as to monitor the execution status and schedule resources.

[0088] Regarding the apparatus in the above embodiments, the specific methods by which each module performs its operations have been described in detail in the embodiments related to the method, and will not be elaborated upon here. Each module in the above-described automated business process generation apparatus based on fuzzy demand semantic parsing can be implemented entirely or partially through software, hardware, or a combination thereof. Each module 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 of each module.

[0089] In one embodiment of this application, a computer device is provided, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: By understanding the pre-defined requirements, the intelligent agent processes the received natural language business requirements and generates structured process instructions corresponding to the natural language business requirements. The structured process instructions are processed by a pre-set intent recognition agent and a pre-set context memory agent to generate intent recognition results. The intent recognition results include business objectives, triggering conditions, key nodes and output requirements. By constructing an intelligent agent through a pre-defined process, the flowchart generation engine is invoked to process the structured process instructions and intent recognition results, generating a preliminary process model. By using a pre-defined software flow mapping agent, target nodes in the initial process model are bound to target software technology components and target data structures, generating configuration files; By pre-deploying intelligent agents based on configuration files, the target business process is deployed to the target execution environment for execution status monitoring and resource scheduling.

[0090] In one embodiment of this application, the processor further performs the following steps when executing the computer program: By understanding the pre-defined requirements, the intelligent agent invokes a pre-defined large language model to perform contextual semantic parsing on the natural language business requirements, and extracts a set of candidate semantic units; the set of candidate semantic units includes multiple candidate semantic units. Based on a semantic recognizer enhanced by a pre-defined rule template and a dictionary, semantic annotation is performed on each candidate semantic unit, and typified intermediate results are output. The typified intermediate results include tasks, process nodes, constraints, and triggering events. Based on a pre-defined domain knowledge base, the typified intermediate results are standardized and confirmed to generate structured process instructions.

[0091] In one embodiment of this application, the processor further performs the following steps when executing the computer program: Obtain modal information corresponding to natural language business requirements; wherein, modal information includes at least one of text, tables, images, and speech; By understanding the preset requirements, the intelligent agent calls a preset large language model to determine the target modal information processing method corresponding to the modal information; Based on the target modality information processing method, the contextual semantics of natural language business requirements are parsed to extract a set of candidate semantic units.

[0092] In one embodiment of this application, the processor further performs the following steps when executing the computer program: The candidate entities in the typed intermediate results are encoded into candidate entity vectors by a preset semantic encoder, and each preset term entity in the preset domain knowledge base is encoded into a standard term vector. The similarity between the candidate entity vectors and the standard term vectors is calculated to obtain a list of similarity scores; Based on the similarity score list, the target term vector that matches the candidate entity vector is determined from each standard term vector; The typified intermediate results are standardized and confirmed based on the current similarity score between the candidate entity vector and the target term vector, and structured process instructions are generated.

[0093] In one embodiment of this application, the processor further performs the following steps when executing the computer program: Obtain the preset similarity threshold; If the current similarity score is greater than the preset similarity threshold, the candidate entity corresponding to the candidate entity vector is replaced by the preset term entity corresponding to the target term vector, and a structured process instruction is generated. If the current similarity score is less than or equal to the preset similarity threshold, the preset term entity corresponding to the target term vector will be marked as an ambiguous item, and a structured process instruction will be generated.

[0094] In one embodiment of this application, the processor further performs the following steps when executing the computer program: The intelligent agent is constructed by calling the flowchart through a pre-defined process to build a domain-specific language description file; the domain-specific language description file includes various types of control structures, including approval, condition, parallel, and rollback. The structured process instructions and intent recognition results are processed based on domain-specific language description files to generate graphical flowcharts. The graphical flowchart can be edited through a preset visual editing interface to generate a preliminary process model.

[0095] In one embodiment of this application, preset control and scheduling agents are used to coordinate and handle exceptions for other agents, and to update the user context memory of the preset context memory agent. The other agents include a preset demand understanding agent, a preset intent recognition agent, a preset context memory agent, a preset process construction agent, a preset software flow mapping agent, and a preset deployment agent. The scheduling and coordination includes managing the execution dependencies and scheduling order among other agents, and the exception handling includes detecting task exceptions of other agents and triggering preset error recovery logic corresponding to the task exceptions.

[0096] The computer device provided in this application embodiment has a similar implementation principle and technical effect to the above method embodiment, and will not be described again here.

[0097] In one embodiment of this application, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, it performs the following steps: By understanding the pre-defined requirements, the intelligent agent processes the received natural language business requirements and generates structured process instructions corresponding to the natural language business requirements. The structured process instructions are processed by a pre-set intent recognition agent and a pre-set context memory agent to generate intent recognition results. The intent recognition results include business objectives, triggering conditions, key nodes and output requirements. By constructing an intelligent agent through a pre-defined process, the flowchart generation engine is invoked to process the structured process instructions and intent recognition results, generating a preliminary process model. By using a pre-defined software flow mapping agent, target nodes in the initial process model are bound to target software technology components and target data structures, generating configuration files; By pre-deploying intelligent agents based on configuration files, the target business process is deployed to the target execution environment for execution status monitoring and resource scheduling.

[0098] In one embodiment of this application, the computer program, when executed by a processor, further performs the following steps: By understanding the pre-defined requirements, the intelligent agent invokes a pre-defined large language model to perform contextual semantic parsing on the natural language business requirements, and extracts a set of candidate semantic units; the set of candidate semantic units includes multiple candidate semantic units. Based on a semantic recognizer enhanced by a pre-defined rule template and a dictionary, semantic annotation is performed on each candidate semantic unit, and typified intermediate results are output. The typified intermediate results include tasks, process nodes, constraints, and triggering events. Based on a pre-defined domain knowledge base, the typified intermediate results are standardized and confirmed to generate structured process instructions.

[0099] In one embodiment of this application, the computer program, when executed by a processor, further performs the following steps: Obtain modal information corresponding to natural language business requirements; wherein, modal information includes at least one of text, tables, images, and speech; By understanding the preset requirements, the intelligent agent calls a preset large language model to determine the target modal information processing method corresponding to the modal information; Based on the target modality information processing method, the contextual semantics of natural language business requirements are parsed to extract a set of candidate semantic units.

[0100] In one embodiment of this application, the computer program, when executed by a processor, further performs the following steps: The candidate entities in the typed intermediate results are encoded into candidate entity vectors by a preset semantic encoder, and each preset term entity in the preset domain knowledge base is encoded into a standard term vector. The similarity between the candidate entity vectors and the standard term vectors is calculated to obtain a list of similarity scores; Based on the similarity score list, the target term vector that matches the candidate entity vector is determined from each standard term vector; The typified intermediate results are standardized and confirmed based on the current similarity score between the candidate entity vector and the target term vector, and structured process instructions are generated.

[0101] In one embodiment of this application, the computer program, when executed by a processor, further performs the following steps: Obtain the preset similarity threshold; If the current similarity score is greater than the preset similarity threshold, the candidate entity corresponding to the candidate entity vector is replaced by the preset term entity corresponding to the target term vector, and a structured process instruction is generated. If the current similarity score is less than or equal to the preset similarity threshold, the preset term entity corresponding to the target term vector will be marked as an ambiguous item, and a structured process instruction will be generated.

[0102] In one embodiment of this application, the computer program, when executed by a processor, further performs the following steps: The intelligent agent is constructed by calling the flowchart through a pre-defined process to build a domain-specific language description file; the domain-specific language description file includes various types of control structures, including approval, condition, parallel, and rollback. The structured process instructions and intent recognition results are processed based on domain-specific language description files to generate graphical flowcharts. The graphical flowchart can be edited through a preset visual editing interface to generate a preliminary process model.

[0103] In one embodiment of this application, preset control and scheduling agents are used to coordinate and handle exceptions for other agents, and to update the user context memory of the preset context memory agent. The other agents include a preset demand understanding agent, a preset intent recognition agent, a preset context memory agent, a preset process construction agent, a preset software flow mapping agent, and a preset deployment agent. The scheduling and coordination includes managing the execution dependencies and scheduling order among other agents, and the exception handling includes detecting task exceptions of other agents and triggering preset error recovery logic corresponding to the task exceptions.

[0104] The computer-readable storage medium provided in this embodiment is similar in principle and technical effect to the method embodiment described above, and will not be repeated here.

[0105] 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 can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various 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 RAMbus dynamic RAM (RDRAM), etc.

[0106] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0107] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. An automatic business process generation method based on fuzzy requirement semantic parsing, characterized in that, The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement; An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement; An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement; An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement; An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement.

2. The method of claim 1, wherein, The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement.

3. The method of claim 2, wherein, The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement.

4. The method according to claim 2 or 3, characterized in that, The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to the natural language business requirement. The method comprises: An agent for understanding requirements processes a received natural language business requirement to generate a structured process instruction corresponding to The type intermediate result is standardized and confirmed according to the current similarity score between the candidate entity vector and the target term vector, and the structured process instruction is generated.

5. The method of claim 4, wherein, The standardized confirmation of the type intermediate result according to the current similarity score between the candidate entity vector and the target term vector, and the generation of the structured process instruction, comprise: obtaining a preset similarity threshold; if the current similarity score is greater than the preset similarity threshold, replacing the candidate entity corresponding to the candidate entity vector with a preset term entity corresponding to the target term vector, and generating the structured process instruction; if the current similarity score is less than or equal to the preset similarity threshold, marking the preset term entity corresponding to the target term vector as an ambiguous item, and generating the structured process instruction.

6. The method according to any one of claims 1 to 3, characterized in that, The preliminary process model is generated by processing the structured process instruction and the intent recognition result through the preset process construction agent calling the process graph generation engine, comprising: a domain-specific language description file is constructed through the preset process construction agent, wherein the domain-specific language description file includes multiple types of control structures, and the multiple types include approval, condition, parallel, and rollback; the structured process instruction and the intent recognition result are processed based on the domain-specific language description file to generate a graphical process graph; the graphical process graph is edited through a preset visual editing interface to generate the preliminary process model.

7. The method according to any one of claims 1 to 3, characterized in that, The method further comprises: the user context memory of the preset context memory agent is updated by scheduling and coordinating other agents and handling exceptions through a preset control and scheduling agent, wherein the other agents include the preset intent recognition agent, the preset context memory agent, the preset process construction agent, the preset software stream mapping agent, and the preset deployment agent; the scheduling and coordination includes managing the execution dependency and scheduling order between the other agents; the exception handling includes detecting task exception conditions of the other agents and triggering a preset error recovery logic corresponding to the task exception conditions.

8. An apparatus for automatic business process generation based on fuzzy requirement semantic parsing, characterized in that, The device comprises: a demand understanding module for processing a received natural language business demand through a preset demand understanding agent to generate a structured process instruction corresponding to the natural language business demand; an intent recognition module for processing the structured process instruction through a multi-round dialogue context provided by a preset intent recognition agent and a preset context memory agent to generate an intent recognition result; wherein the intent recognition result includes a business target, a trigger condition, a key node, and an output requirement; a process construction module for processing the structured process instruction and the intent recognition result through a preset process construction agent calling a process graph generation engine to generate a preliminary process model; a software flow mapping module, configured to bind the target node in the preliminary process model with a target software technology component and a target data structure by a preset software flow mapping agent, and generate a configuration file; a deployment module, configured to deploy the target business process to a target execution environment based on the configuration file by a preset deployment agent, to perform state monitoring and resource scheduling.

9. An electronic device, comprising: The electronic device includes a processor and a memory, and the memory stores at least one instruction, at least one program, a code set or an instruction set, which are loaded and executed by the processor to implement the automatic business process generation method based on fuzzy requirement semantic parsing according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The storage medium stores at least one instruction, at least one program, a code set or an instruction set, which are loaded and executed by the processor to implement the automatic business process generation method based on fuzzy requirement semantic parsing according to any one of claims 1-7.

Images