Interactive business process intelligent modeling system driven by meta-model

Abstract

The invention relates to the technical field of business process modeling, in particular to a meta-model-driven interactive business process intelligent modeling method and system. The method comprises the following steps: constructing a meta-model framework containing core elements such as activities, events and gateways, and establishing a meta-model constraint rule base to define a structure legality rule of a BPMN2.0 specification; based on a domain ontology Schema and the domain exclusive large language model after instruction fine tuning, extracting a business entity and a process relationship thereof from the unstructured text; carrying out information integrity verification and logic consistency detection on the extracted BPMN elements by utilizing a constraint satisfaction algorithm and Petri net reachability analysis; displaying a difference label of the process sketch through an interactive verification interface, and realizing semantic correction and model iteration of a user on the process elements based on a natural language feedback mechanism; an XML code is generated by adopting a template-driven code generation strategy, and a flow chart is generated through a BPMN parser. According to the method, the business process modeling efficiency can be remarkably improved, and the manual intervention proportion in the modeling process is reduced.

Description

Technical Field

[0001] This invention relates to the field of business process modeling technology, and in particular to a meta-model-driven interactive intelligent business process modeling method, system, and computer-readable storage medium. Background Technology

[0002] Business process modeling, as a core component of enterprise digital transformation, is widely used in heavily regulated industries such as finance, manufacturing, and healthcare. Existing technologies utilize a collaborative approach involving meta-model-driven approaches, BPMN standardization, and intelligent agent technology to construct a technical system encompassing requirements analysis and model generation. Specifically, this system covers the entire process from unstructured text parsing and meta-model constraint verification to XML code generation, including key steps such as BPMN 2.0 specification adaptation, domain ontology semantic mapping, and human-machine collaborative verification. However, with increasing business complexity, traditional methods are no longer sufficient to meet the dual requirements of cross-domain dynamic adaptation and semantic consistency assurance.

[0003] Existing business process modeling methods directly rely on human experience-driven meta-model expansion without establishing a dynamic integration mechanism with the domain ontology. This can lead to poor cross-domain adaptability or low modeling efficiency, thereby affecting the model's business compliance and technical specifications. Traditional BPMN modeling depends on manual drawing using design tools, which suffers from low efficiency in verifying complex processes, manual coding required for model-to-code conversion, and still requires more than 70% manual intervention during model iteration. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] Therefore, the first objective of this invention is to propose a model-driven intelligent modeling method for interactive business processes, characterized by comprising: S1. Construct a metamodel framework containing core elements such as activities, events, and gateways, and establish a metamodel constraint rule base to define the structural legality rules of the BPMN2.0 specification. S2, based on the domain ontology schema and the domain-specific large language model after instruction fine-tuning, extracts business entities and their process relationships from unstructured text, and maps the business entities to BPMN element types; S3 utilizes constraint satisfaction algorithms and Petri net reachability analysis to perform information integrity verification and logical consistency checks on the extracted BPMN elements, thereby completing the verification of the information integrity of the meta-model and generating a missing information completion scheme based on the knowledge base and large language model. S4 displays the differences in the flowchart through an interactive verification interface, and enables users to make semantic corrections to the flowchart elements and iterate the model based on a natural language feedback mechanism. S5 uses a template-driven code generation strategy to generate XML code and generates flowcharts through a BPMN parser.

[0006] In one embodiment of the present invention, S1 further includes: S11 uses formal modeling tools to parse the meta-model structure of the BPMN 2.0 specification and generates a meta-meta-model framework that includes activities, events, gateways, connection objects, and swimlanes. S12, Establish a meta-model constraint rule base as the semantic constraint of BPMN, wherein the meta-model constraint rule base includes control flow integrity rules and structural legality rules.

[0007] In one embodiment of the present invention, S2 further includes: S21, Construct the business concept ontology of the target domain through ontology modeling tools, and define the schema structure of business object attributes, behavioral characteristics and interaction relationships; S22 employs a text parsing strategy based on prompt engineering, combined with a domain-specific large language model fine-tuned from instructions, to extract business entities and process relationships that conform to BPMN element types from user requirement documents.

[0008] In one embodiment of the present invention, S3 further includes: S31, develop a verification engine based on constraint satisfaction algorithm, and use the verification engine in parallel to automatically verify whether the extracted elements meet the topological constraints of the meta-model in order to detect information integrity; S32, Construct a process logic consistency detection module, and use the Petri net transformation method to perform reachability analysis and deadlock detection on the process structure; S33 acquires existing industry standard process templates, enterprise historical process data, and business rule bases. It matches standard process elements in the domain knowledge graph through entity connection technology and generates missing element completion solutions through the contextual inference capability of the large language model to construct the domain knowledge graph.

[0009] In one embodiment of the present invention, S4 further includes: S41 constructs an interactive element verification interface and uses difference visualization technology to highlight process nodes that may be missing or contradictory. S42 implements a correction mechanism based on natural language feedback, allowing users to directly modify the semantic description of process elements through text input.

[0010] In one embodiment of the present invention, S5 further includes: S51, construct a mapping rule library from BPMN elements to BPMN2.0 XML Schema, including conversion rules such as element identifier generation, attribute set filling, and association relationship encoding, and use a template-driven code generation strategy combined with the XMI serialization algorithm to generate standard BPMN XML files; S52, based on the generated standard BPMN XML file, constructs a DOM tree structure of process elements through the BPMN parser, and applies an automatic layout algorithm to render and generate the flowchart.

[0011] In one embodiment of the present invention, it further includes: After generating the standard BPMN XML file, the standard BPMN XML file is verified by human and machine verification through XML Schema validation and XPath expression detection functions.

[0012] To achieve the above objectives, a second aspect of the present invention provides a business process automation modeling apparatus, comprising: The metamodel building module is used to build a meta-metamodel framework containing core elements such as activities, events, and gateways, and to establish a metamodel constraint rule library to define the structural legality rules of the BPMN2.0 specification. The business entity extraction and mapping module is used to extract business entities and their process relationships from unstructured text based on the domain ontology schema and the domain-specific large language model after instruction fine-tuning, and to map the business entities to BPMN element types. The process verification module is used to perform information integrity verification and logical consistency detection on the extracted BPMN elements using constraint satisfaction algorithms and Petri net reachability analysis, and generate a process sketch that conforms to the meta-model constraints. The interactive verification and correction module is used to display the difference annotations of the process sketch through an interactive verification interface, and realize the semantic correction of process elements and model iteration by users based on the natural language feedback mechanism. The flowchart generation module uses a template-driven code generation strategy to generate XML code and then uses a BPMN parser to generate flowcharts.

[0013] In one embodiment of the present invention, the metamodel building module is further configured to: The meta-model structure of the BPMN 2.0 specification is analyzed using formal modeling tools to generate a meta-meta-model framework that includes activities, events, gateways, connection objects, and swimlanes. A meta-model constraint rule base is established as the semantic constraint of the BPMN. The meta-model constraint rule base includes control flow integrity rules and structural legality rules.

[0014] To achieve the above objectives, a third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0015] The methods, systems, and storage media of this invention can significantly improve the efficiency of business process modeling, reduce the proportion of manual intervention in the modeling process, and enhance cross-industry adaptability and semantic consistency through the joint driving of metamodel and domain ontology, effectively reducing model error rate and realizing interpretability and real-time verification of the modeling process.

[0016] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0017] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a meta-model-driven intelligent modeling method for interactive business processes according to an embodiment of the present invention; Figure 2 This is a structural diagram of a meta-model-driven interactive business process intelligent modeling system according to an embodiment of the present invention. Detailed Implementation

[0018] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] The following description, with reference to the accompanying drawings, describes an embodiment of the present invention. Example 1 Figure 1 This is a flowchart of a meta-model-driven intelligent modeling method for interactive business processes according to an embodiment of the present invention.

[0021] like Figure 1 As shown, the metamodel-driven intelligent modeling method for interactive business processes includes the following steps: S1. Construct a meta-meta-model framework containing core elements such as activities, events, and gateways, and establish a meta-model constraint rule library to define the structural legality rules of the BPMN2.0 specification.

[0022] Further, step S1 includes: S11 uses formal modeling tools to parse the meta-model structure of the BPMN 2.0 specification and generates a meta-meta-model framework that includes activities, events, gateways, connection objects, and swimlanes. S12, Establish a meta-model constraint rule base as the semantic constraint of BPMN, wherein the meta-model constraint rule base includes control flow integrity rules and structural legality rules.

[0023] Specifically, in some implementations, constructing a joint driving framework that includes the BPMN metamodel specification and a domain ontology knowledge base is a core technical step in this invention for achieving intelligent business process modeling. This step formally parses the BPMN 2.0 metamodel structure and establishes a dynamic constraint rule base, thereby providing structured and semantic foundational support for subsequent process element identification, semantic mapping, and model generation.

[0024] At the technical implementation level, this step first formally models the meta-model in the BPMN 2.0 standard. Using EMF (Eclipse Modeling Framework) or similar modeling tools, the meta-elements of BPMN (such as `Activity`, `Event`, `Gateway`, `SequenceFlow`, `Lane`, etc.) and their semantic constraints (such as inheritance, aggregation, association, etc.) are transformed into a computable meta-metamodel structure. Furthermore, by defining a constraint rule base for the meta-model, static validation of the flow structure is achieved. For example, the control flow integrity rule requires that each `Activity` must contain at least one `Incoming` and one `Outgoing` `SequenceFlow`; the structural validity rule requires that `ExclusiveGateway`s must appear in pairs, and their child nodes must satisfy specific topological relationships.

[0025] At the parameter level, the constraint rule base can be semantically described using OWL (Web Ontology Language) or SHACL (Shapes Constraint Language), supporting dynamic verification by the SPARQL query and inference engine. The rule base can be updated daily or weekly to adapt to the evolution of business rules. Furthermore, the rule matching accuracy must reach over 95% to ensure the compliance of the model during the generation phase.

[0026] At the application level, this framework can be widely used in industries with high requirements for process compliance, such as finance, manufacturing, and healthcare. For example, in financial risk control process modeling, the system needs to simultaneously meet the structural specifications of BPMN 2.0 and the business logic constraints of the Basel Accords. Through the joint driving framework, automatic mapping of business terms (such as "credit assessment" and "approval nodes") to BPMN elements can be achieved, improving modeling efficiency and accuracy.

[0027] In terms of technical effectiveness, this step significantly improves the automation and cross-domain adaptability of process modeling by semantically fusing the BPMN metamodel with the domain ontology knowledge base. Simultaneously, the introduction of a dynamic constraint rule base allows for dual structural and semantic verification of the model before generation, effectively reducing subsequent verification and correction costs and improving the robustness and maintainability of the overall modeling system.

[0028] S2, based on the domain ontology schema and the domain-specific large language model after instruction fine-tuning, extracts business entities and their process relationships from unstructured text, and maps the business entities to BPMN element types.

[0029] Further, step S2 includes: S21, Construct the business concept ontology of the target domain through ontology modeling tools, and define the schema structure of business object attributes, behavioral characteristics and interaction relationships; S22 employs a text parsing strategy based on prompt engineering, combined with a domain-specific large language model fine-tuned from instructions, to extract business entities and process relationships that conform to BPMN element types from user requirement documents.

[0030] Specifically, in step 2, this invention uses a domain-specific large language model (LLM) fine-tuned with instructions to perform semantic parsing on unstructured text, and combines it with entity linking technology to map the extracted business entities and process relationships to the metamodel specification and domain ontology schema. This step is the core link in realizing automated business process modeling, and its technical implementation is based on the deep integration of natural language processing (NLP) and knowledge graph technologies.

[0031] In some implementations, this step first involves designing structured input templates through prompt engineering, transforming user-provided business requirement documents (such as text descriptions, meeting minutes, process specifications, etc.) into instruction formats that the model can understand. Upon receiving input, the model, based on its instruction tuning within a specific domain (such as finance, manufacturing, logistics, etc.), can identify and extract key business entities (such as "order approval" and "inventory verification") and their semantic relationships (such as "sequential flow" and "message flow"). During tuning, the model is trained on a large amount of annotated business process text, learning domain-specific terminology, process patterns, and semantic structures, thereby improving its extraction accuracy from unstructured text.

[0032] Furthermore, entity linking technology is used to match extracted business entities with standard concepts in the domain ontology schema. For example, after the model identifies "order approval," it uses an entity linking algorithm to semantically align it with "OrderApprovalActivity" defined in the ontology, ensuring that the extraction result conforms to the predefined metamodel structure. This process typically relies on semantic similarity calculations (such as BERTScore and Cosine Similarity) and contextual semantic reasoning to achieve high-precision entity mapping.

[0033] In application scenarios, this step is widely applicable to areas such as enterprise process digital transformation, business process automation (RPA) configuration, and compliance process modeling. For example, in the financial industry, users can describe in natural language that "after a customer submits a loan application, it needs to go through three steps: risk control review, credit approval, and contract signing." The system will automatically identify and map these steps to the corresponding BPMN activity nodes and match them with standard process elements in the ontology schema through entity links, ensuring the semantic consistency and executability of the model.

[0034] The technical benefits of this step are that it significantly improves the accuracy of business entity identification and the completeness of process relationship extraction, reduces the need for repeated confirmation of business terms in manual modeling, and enhances the interpretability and cross-system compatibility of the model by semantic alignment with the domain ontology, laying a solid foundation for subsequent metamodel verification and XML generation.

[0035] S3 utilizes constraint satisfaction algorithms and Petri net reachability analysis to perform information integrity verification and logical consistency checks on the extracted BPMN elements, thereby completing the verification of the information integrity of the meta-model, and generating a missing information completion scheme based on the knowledge base and the large language model.

[0036] Further, step S3 includes: S31, develop a verification engine based on constraint satisfaction algorithm, and use the verification engine in parallel to automatically verify whether the extracted elements meet the topological constraints of the meta-model in order to detect information integrity; S32, Construct a process logic consistency detection module, and use the Petri net transformation method to perform reachability analysis and deadlock detection on the process structure; S33 acquires existing industry standard process templates, enterprise historical process data, and business rule bases. It matches standard process elements in the domain knowledge graph through entity connection technology and generates missing element completion solutions through the contextual inference capability of the large language model to construct the domain knowledge graph.

[0037] Specifically, in step 3-1, the system performs structural integrity verification on the business process elements extracted from unstructured text based on the Constraint Satisfaction Problem (CSP) algorithm. This algorithm formalizes the topological constraints in the BPMN 2.0 metamodel into logical constraints, constructing a CSP framework containing variables, constraints, and a solver, thereby verifying whether the extraction results meet the integrity requirements of the process structure. Specifically, variables correspond to nodes in the process (such as activities, events, gateways, etc.), and constraints include control flow integrity rules (such as "each activity must contain a clear input-output sequence flow") and structural validity rules (such as "exclusive gateways must appear in pairs"), which can be represented as logical expressions or connectivity conditions in graph theory.

[0038] In some implementations, the CSP solver combines backtracking search with heuristic variable selection strategies (such as Minimum Remaining Values, MRV) to improve verification efficiency. The system sets a maximum number of iterations. With timeout threshold This ensures real-time responsiveness even in complex processes. If validation fails, the system will output the set of nodes that violate constraints. And mark the inconsistencies in its topological relationships.

[0039] Furthermore, this step plays a crucial "quality control" role in the entire modeling process, ensuring that the process elements extracted from the text structurally conform to the BPMN 2.0 specification, providing a foundation for subsequent semantic consistency detection and automatic completion. Through this mechanism, the system can effectively reduce the model reconstruction cost caused by structural errors and improve the automation and reliability of the modeling process.

[0040] S4 displays the differences in the flowchart through an interactive verification interface, and enables users to make semantic corrections to the flowchart elements and iterate the model based on a natural language feedback mechanism.

[0041] Further, step S4 includes: S41 constructs an interactive element verification interface and uses difference visualization technology to highlight process nodes that may be missing or contradictory. S42 implements a correction mechanism based on natural language feedback, allowing users to directly modify the semantic description of process elements through text input.

[0042] Specifically, in step 4, the system develops an interactive element verification interface and uses difference visualization technology to highlight potentially missing or contradictory process nodes, thereby guiding users to make accurate semantic corrections. This step is a key link in achieving a closed loop of human-machine collaboration in the entire modeling process, and its technical implementation is based on the deep integration of a graphical user interface (GUI) and a semantic difference detection algorithm.

[0043] In some implementations, discrepancy visualization technology identifies potential mismatches or missing nodes by comparing the semantic consistency between machine-generated process elements and the original business text. Specifically, the system employs semantic similarity-based matching algorithms, such as BERT or RoBERTa sentence embedding models, to calculate the semantic distance between the text description and the process node labels. When the semantic similarity is below a preset threshold (e.g., ...), the system identifies the missing nodes. When a node is identified as "pending verification," the system marks it as such and highlights it with a red border or a bright background in the visualization interface. Furthermore, the system supports context-based process node reachability analysis. If a node is determined to be unreachable or at risk of deadlock in Petri net reachability analysis, a warning icon will be displayed to alert the user.

[0044] S5 uses a template-driven code generation strategy to generate XML code and generates flowcharts through a BPMN parser.

[0045] Further, step S5 includes: S51, construct a mapping rule library from BPMN elements to BPMN2.0 XML Schema, including conversion rules such as element identifier generation, attribute set filling, and association relationship encoding, and use a template-driven code generation strategy combined with the XMI serialization algorithm to generate standard BPMN XML files; S52, based on the generated standard BPMN XML file, constructs a DOM tree structure of process elements through the BPMN parser, and applies an automatic layout algorithm to render and generate the flowchart.

[0046] This step involves establishing a data channel between the semantic difference detection module and the visualization rendering engine. The difference detection module outputs the semantic confidence matrix of the nodes. ,in This represents the number of process nodes. For the number of text descriptions, Indicates the first The node and the first The matching degree between each text description. The visualization engine dynamically renders a flowchart based on this matrix, allowing users to view the matching text, confidence scores, and possible alternatives by clicking or hovering.

[0047] In practical applications, this step is widely used in industries with high requirements for process compliance, such as finance, manufacturing, and healthcare. For example, in financial risk control process modeling, the system can automatically identify issues such as "missing approval nodes" or "incomplete compliance clauses," and guide business personnel to supplement or adjust them through an interactive interface. This technology significantly improves the transparency and controllability of the modeling process, allowing human feedback to directly affect the semantic structure of the model, thereby reducing model reconstruction costs and improving modeling quality and business adaptability.

[0048] The meta-model-driven intelligent business process modeling method of this invention can significantly improve the automation level and cross-domain adaptability of business process modeling. It achieves end-to-end generation from unstructured text to standard BPMN model through the joint driving of meta-model and domain ontology, and enhances the semantic integrity of the model and the interpretability of human-computer collaboration.

[0049] Example 2 Figure 2 This is a structural diagram of a meta-model-driven interactive business process intelligent modeling system according to an embodiment of the present invention.

[0050] like Figure 2 As shown, the metamodel-driven interactive business process intelligent modeling system includes: The metamodel building module is used to build a meta-metamodel framework containing core elements such as activities, events, and gateways, and to establish a metamodel constraint rule library to define the structural legality rules of the BPMN2.0 specification. The business entity extraction and mapping module is used to extract business entities and their process relationships from unstructured text based on the domain ontology schema and the domain-specific large language model after instruction fine-tuning, and to map the business entities to BPMN element types. The process verification module is used to perform information integrity verification and logical consistency detection on the extracted BPMN elements using constraint satisfaction algorithms and Petri net reachability analysis, and generate a process sketch that conforms to the meta-model constraints. The interactive verification and correction module is used to display the difference annotations of the process sketch through an interactive verification interface, and realize the semantic correction of process elements and model iteration by users based on the natural language feedback mechanism. The flowchart generation module uses a template-driven code generation strategy to generate XML code and then uses a BPMN parser to generate flowcharts.

[0051] Furthermore, the metamodel building module is also used for: The meta-model structure of the BPMN 2.0 specification is analyzed using formal modeling tools to generate a meta-meta-model framework that includes activities, events, gateways, connection objects, and swimlanes. A meta-model constraint rule base is established as the semantic constraint of the BPMN. The meta-model constraint rule base includes control flow integrity rules and structural legality rules.

[0052] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned meta-model-driven interactive business process intelligent modeling method.

[0053] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0054] Furthermore, 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 technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A model-driven intelligent modeling method for interactive business processes, characterized in that, include: S1. Construct a metamodel framework containing core elements such as activities, events, and gateways, and establish a metamodel constraint rule base to define the structural legality rules of the BPMN2.0 specification. S2, based on the domain ontology schema and the domain-specific large language model after instruction fine-tuning, extracts business entities and their process relationships from unstructured text, and maps the business entities to BPMN element types; S3 utilizes constraint satisfaction algorithms and Petri net reachability analysis to perform information integrity verification and logical consistency checks on the extracted BPMN elements, and generates missing information completion schemes based on knowledge bases and large language models. S4 displays the differences in the flowchart through an interactive verification interface, and enables users to make semantic corrections to the flowchart elements and iterate the model based on a natural language feedback mechanism. S5 uses a template-driven code generation strategy to generate XML code and generates flowcharts through a BPMN parser.

2. The method as described in claim 1, characterized in that, S1 further includes: S11 uses formal modeling tools to parse the meta-model structure of the BPMN 2.0 specification and generates a meta-meta-model framework that includes activities, events, gateways, connection objects, and swimlanes. S12, Establish a meta-model constraint rule base as the semantic constraint of BPMN, wherein the meta-model constraint rule base includes control flow integrity rules and structural legality rules.

3. The method as described in claim 1, characterized in that, S2 further includes: S21, Construct the business concept ontology of the target domain through ontology modeling tools, and define the schema structure of business object attributes, behavioral characteristics and interaction relationships; S22 employs a text parsing strategy based on prompt engineering, combined with a domain-specific large language model fine-tuned from instructions, to extract business entities and process relationships that conform to BPMN element types from user requirement documents.

4. The method as described in claim 1, characterized in that, S3 further includes: S31, develop a verification engine based on constraint satisfaction algorithm, and use the verification engine in parallel to automatically verify whether the extracted elements meet the topological constraints of the meta-model in order to detect information integrity; S32, Construct a process logic consistency detection module, and use the Petri net transformation method to perform reachability analysis and deadlock detection on the process structure; S33 acquires existing industry standard process templates, enterprise historical process data, and business rule bases. It matches standard process elements in the domain knowledge graph through entity connection technology and generates missing element completion solutions through the contextual inference capability of the large language model to construct the domain knowledge graph.

5. The method as described in claim 1, characterized in that, S4 further includes: S41 constructs an interactive element verification interface and uses difference visualization technology to highlight process nodes that may be missing or contradictory. S42 implements a correction mechanism based on natural language feedback, allowing users to directly modify the semantic description of process elements through text input.

6. The method as described in claim 1, characterized in that, The S5 also includes: S51, construct a mapping rule library from BPMN elements to BPMN2.0 XML Schema, including conversion rules such as element identifier generation, attribute set filling, and association relationship encoding, and use a template-driven code generation strategy combined with the XMI serialization algorithm to generate standard BPMN XML files; S52, based on the generated standard BPMN XML file, constructs a DOM tree structure of process elements through the BPMN parser, and applies an automatic layout algorithm to render and generate the flowchart.

7. The method as described in claim 1, characterized in that, Also includes: After generating the standard BPMN XML file, the standard BPMN XML file is verified by human and machine verification through XML Schema validation and XPath expression detection functions.

8. A business process automation modeling device, characterized in that, include: The metamodel building module is used to build a meta-metamodel framework containing core elements such as activities, events, and gateways, and to establish a metamodel constraint rule library to define the structural legality rules of the BPMN2.0 specification. The business entity extraction and mapping module is used to extract business entities and their process relationships from unstructured text based on the domain ontology schema and the domain-specific large language model after instruction fine-tuning, and to map the business entities to BPMN element types. The process verification module is used to perform information integrity verification and logical consistency detection on the extracted BPMN elements using constraint satisfaction algorithms and Petri net reachability analysis, and generate a process sketch that conforms to the meta-model constraints. The interactive verification and correction module is used to display the difference annotations of the process sketch through an interactive verification interface, and realize the semantic correction of process elements and model iteration by users based on the natural language feedback mechanism. The flowchart generation module uses a template-driven code generation strategy to generate XML code and then uses a BPMN parser to generate flowcharts.

9. The system method as described in claim 8, characterized in that, The metamodel building module is also used for: The meta-model structure of the BPMN 2.0 specification is analyzed using formal modeling tools to generate a meta-meta-model framework that includes activities, events, gateways, connection objects, and swimlanes. A meta-model constraint rule base is established as the semantic constraint of the BPMN. The meta-model constraint rule base includes control flow integrity rules and structural legality rules.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as claimed in any one of claims 1-8.

Images