Business process automatic modeling method and device, electronic equipment and medium

Abstract

The invention discloses a business process automatic modeling method and device, electronic equipment and a medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: receiving and analyzing current multi-modal input description data in real time in a business process automatic modeling system architecture to obtain current multi-modal input intermediate representation description data; carrying out extraction through an entity relationship process extraction method, and generating a current process topological knowledge graph in combination with a process modeling standard adaptation method; verifying through a business rule atlas verification and identification method, and generating a current intermediate representation parameter with a semantic role; and through a dynamic model generation method, describing a current intermediate representation parameter with a semantic role, and generating and feeding back a current automatic modeling business process in combination with a historical business process modeling template. The problems that manual operation is highly relied on, and static characteristics and data input forms are rigid are solved, the flexibility and automation of business process modeling are improved, and the burden of manual processing is relieved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a business process automation modeling method, device, electronic device and medium. Background Art

[0002] Business processes can generally be generated through manual modeling or semi-automated rule-driven modeling. However, this approach suffers from low efficiency and inflexibility. With the development of artificial intelligence, automated modeling of business processes is crucial.

[0003] In the process of implementing the present invention, the inventors discovered the following deficiencies in the prior art: Currently, manual modeling requires manual conversion of collected unstructured requirements into textual descriptions, followed by manual mapping using modeling tools and dragging related graphical elements to construct business processes. This method relies heavily on manual operations, resulting in both inefficiencies and inaccuracies. Semi-automated rule-driven modeling, which generates business processes through partial automation based on a predefined rule base, is limited by the static nature of the rule base and the rigidity of the input format. Furthermore, it only supports structured data input, resulting in significant deficiencies in flexibility and scalability. Summary of the Invention

[0004] The present invention provides a business process automation modeling method, device, electronic equipment and medium to improve the flexibility and automation of business process modeling.

[0005] According to one aspect of the present invention, a business process automation modeling method is provided, which includes:

[0006] In a pre-established business process automation modeling system architecture, the multimodal input parsing module receives and parses the current multimodal input description data in real time to obtain the current multimodal input intermediate representation description data;

[0007] The business process automation modeling system architecture includes a scenario service layer, a basic service layer, and a large language model layer; the scenario service layer includes a multimodal input parsing module; the basic service layer includes a dynamic model generation module; the large language model layer includes a knowledge graph verification module and a large language model semantic understanding module;

[0008] The entity relationship process extraction method in the semantic understanding module of the large language model is used to extract the current multimodal input intermediate representation description data, and combined with the pre-built process modeling standard adaptation method, the current process topology knowledge graph is generated;

[0009] The current process topology knowledge graph is verified by the business rule graph verification and identification method corresponding to the knowledge graph verification module, and the current intermediate representation parameters with semantic roles are generated;

[0010] The dynamic model generation method corresponding to the dynamic model generation module is used to describe the current intermediate representation parameters with semantic roles, and combined with the historical business process modeling template, the current automated modeling business process is generated, and the current automated modeling business process is fed back to the user.

[0011] According to another aspect of the present invention, a business process automation modeling device is provided, comprising:

[0012] A current multimodal input intermediate representation description data determination module is used to receive and parse the current multimodal input description data in real time through the multimodal input parsing module in a pre-established business process automation modeling system architecture to obtain the current multimodal input intermediate representation description data;

[0013] The business process automation modeling system architecture includes a scenario service layer, a basic service layer, and a large language model layer; the scenario service layer includes a multimodal input parsing module; the basic service layer includes a dynamic model generation module; the large language model layer includes a knowledge graph verification module and a large language model semantic understanding module;

[0014] A current process topology knowledge graph generation module is used to extract the current multimodal input intermediate representation description data through the entity relationship process extraction method in the large language model semantic understanding module, and generate the current process topology knowledge graph in combination with a pre-built process modeling standard adaptation method;

[0015] A current intermediate representation parameter generation module with semantic roles is used to verify the current process topology knowledge graph through the business rule graph verification and identification method corresponding to the knowledge graph verification module, and generate the current intermediate representation parameter with semantic roles;

[0016] The current automated modeling business process generation and feedback module is used to describe the current intermediate representation parameters with semantic roles through the dynamic model generation method corresponding to the dynamic model generation module, and combine the historical business process modeling template to generate the current automated modeling business process, and feedback the current automated modeling business process to the user.

[0017] According to another aspect of the present invention, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the business process automation modeling method described in any embodiment of the present invention is implemented.

[0018] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the business process automation modeling method described in any embodiment of the present invention when executed.

[0019] The technical solution of the embodiment of the present invention is to receive and parse the current multimodal input description data in real time through the multimodal input parsing module in a pre-established business process automation modeling system architecture to obtain the current multimodal input intermediate representation description data; extract the current multimodal input intermediate representation description data through the entity relationship process extraction method in the large language model semantic understanding module, and generate the current process topology knowledge graph in combination with the pre-built process modeling standard adaptation method; verify the current process topology knowledge graph through the business rule graph verification and recognition method corresponding to the knowledge graph verification module, and generate the current semantic role intermediate representation parameters; describe the current semantic role intermediate representation parameters through the dynamic model generation method corresponding to the dynamic model generation module, and generate the current automated modeling business process in combination with the historical business process modeling template, and feedback the current automated modeling business process to the user. This solves the problems of high reliance on manual operation, static characteristics and rigid data input form, improves the flexibility, scalability and automation of business process modeling, reduces the burden of manual processing, and improves the accuracy of business process automation modeling.

[0020] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 This is a flowchart of a business process automation modeling method provided in accordance with the first embodiment of the present invention;

[0023] Figure 2 This is a detailed flowchart of a business process automation modeling method provided according to the second embodiment of the present invention;

[0024] Figure 3 This is a schematic diagram of the structure of a business process automation modeling device provided according to the third embodiment of the present invention;

[0025] Figure 4 It is a structural diagram of an electronic device provided according to the fourth embodiment of the present invention. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "target", "current", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0028] It is worth noting that in the technical solution of this application, the information collected is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with the relevant laws, regulations and standards of relevant countries and regions, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse; if the user chooses to refuse, the expert decision-making process will be entered.

[0029] Example 1

[0030] Figure 1 A flowchart of a business process automation modeling method is provided for embodiment 1 of the present invention. This embodiment is applicable to situations where business processes are automatically modeled. The method can be executed by a business process automation modeling device, which can be implemented in the form of hardware and / or software.

[0031] Correspondingly, such as Figure 1 As shown, the method includes:

[0032] S110 . In a pre-established business process automation modeling system architecture, the multimodal input parsing module receives and parses the current multimodal input description data in real time to obtain the current multimodal input intermediate representation description data.

[0033] Among them, the business process automation modeling system architecture includes a scenario service layer, a basic service layer and a large language model layer; the scenario service layer includes a multimodal input parsing module; the basic service layer includes a dynamic model generation module; the large language model layer includes a knowledge graph verification module and a large language model semantic understanding module.

[0034] In this embodiment, a business process automation modeling system architecture is constructed. The architecture is a lightweight architecture deployed in cloud devices. The automated modeling operations of business processes are realized through the scenario service layer, basic service layer and large language model layer in the system architecture.

[0035] The scenario service layer primarily serves as a front-end service, receiving and parsing the current multimodal input description data. The basic service layer primarily serves as a back-end service, providing foundational service support for data analysis, knowledge question-and-answer services, and more. It serves as a bridge between the scenario service layer and the large language model layer, receiving conversation requests from the scenario service layer, invoking the large language model layer, and returning generated content. The large language model layer primarily serves as a back-end service, providing infrastructure functions such as user management, role management, and permission management. It also provides database and knowledge base configuration related to the large model scenario service, session history, and invocation services for big data infrastructure services.

[0036] The current multimodal input description data may be input data in multiple modal forms. Specifically, the current multimodal input description data may include multimodal input description data such as voice, sketch, text, or table.

[0037] The current multimodal input intermediate representation description data can be obtained by parsing the multimodal input data to obtain the data parsed intermediate representation corresponding to each modality. The current multimodal input intermediate representation description data can be recognized by the business process automation modeling system architecture, thereby improving the automated modeling of business processes.

[0038] Specifically, for voice input description data, a speech transcription model can first transcribe it into text, and then a large language extraction model can be used to identify key operation nodes to obtain an intermediate data parsing representation. For example, for form input description data, fields such as approval levels or role permissions can be parsed to identify key operation nodes and obtain an intermediate data parsing representation.

[0039] Optionally, the multimodal input parsing module receives and parses the current multimodal input description data in real time to obtain the current multimodal input intermediate representation description data, including: parsing the current multimodal input description data through a preset unified intermediate representation method to obtain the data parsing intermediate representation corresponding to each modality; performing feature alignment processing on each data parsing intermediate representation through a preset cross-modal alignment loss function to obtain each data parsing feature alignment intermediate representation; and fusing each of the data parsing feature alignment intermediate representations to obtain the current multimodal input intermediate representation description data.

[0040] The unified intermediate representation method can be a unified intermediate representation that can be obtained by processing multimodal input data separately. The data parsing intermediate representation can be an intermediate representation obtained from different modalities using a specified intermediate representation method. The cross-modal alignment loss function can be a loss function that can align features across multiple intermediate representations. The data parsing feature alignment intermediate representation can be the parameters of each intermediate representation after feature alignment.

[0041] For example, assuming that the current multimodal input description data includes speech, sketch and text, a unified intermediate representation method can be used to obtain the speech data parsing intermediate representation, the sketch data parsing intermediate representation and the text data parsing intermediate representation respectively.

[0042] Furthermore, through the cross-modal alignment loss function, feature alignment processing is performed on the intermediate representation of speech data parsing, the intermediate representation of sketch data parsing, and the intermediate representation of text data parsing, respectively, to obtain the speech data parsing feature alignment intermediate representation, the sketch data parsing feature alignment intermediate representation, and the text data parsing feature alignment intermediate representation, respectively.

[0043] Accordingly, by fusing the intermediate representations of speech data parsing feature alignment, sketch data parsing feature alignment, and text data parsing feature alignment, we can further obtain the intermediate representation description data of the current multimodal input. Furthermore, we can store the intermediate representation description data of the current multimodal input in the knowledge base and reference architecture asset library.

[0044] The advantage of this setting is that by unifying the intermediate representation method and processing the cross-modal alignment loss function to obtain the current multimodal input intermediate representation description data, it is possible to better process multimodal data and improve the flexibility and scalability of business process modeling.

[0045] S120. Extract the current multimodal input intermediate representation description data through the entity relationship process extraction method in the large language model semantic understanding module, and generate the current process topology knowledge graph in combination with the pre-built process modeling standard adaptation method.

[0046] The entity-relationship process extraction method can be a method for extracting entities, entity relationships, and process logic from data. The process modeling standard adaptation method can be a method for adapting the process modeling methodology to the historical modeling standard.

[0047] Among them, the current process topology knowledge graph can be a graph that uses the form of a knowledge graph to represent the relationship between entities between multimodal input data and process logic and other information.

[0048] Optionally, the entity relationship process extraction method in the large language model semantic understanding module is used to extract the current multimodal input intermediate representation description data, and combined with the pre-built process modeling standard adaptation method, the current process topology knowledge graph is generated, including: extracting the current multimodal input intermediate representation description data through the entity relationship process extraction method to obtain each current entity, current entity relationship and current process logic; according to the process modeling standard adaptation method, the process flow meta-semantics in the historical modeling standard are adapted to the process modeling methodology to obtain the process flow graph meta-semantic logical constraints; combined with the process flow graph meta-semantic logical constraints, the knowledge graphs are constructed for each current entity, current entity relationship and current process logic respectively to generate the current process topology knowledge graph.

[0049] The flowchart meta-semantic logical constraints may be the conversion of standard flowchart meta-semantics into computable logical constraints, and may specifically include node dependencies or state transition rules.

[0050] In this embodiment, it is first necessary to extract the current multimodal input intermediate representation description data to obtain each current entity, current entity relationship and current process logic. The number of current entities, current entity relationships and current process logic here is at least one, so as to further generate the current process topology knowledge graph based on the obtained flow chart meta-semantic logical constraints.

[0051] For example, assuming that the current entity includes the front-end input stage, the review stage and the quality inspection stage, and also includes the input personnel, the review personnel and the quality inspection personnel, different current entities include different entity relationships. For example, after the review by the reviewer, it needs to be sent to the quality inspector for review, but for different businesses, quality inspection may not be required after the review, so different process logics correspond.

[0052] The advantage of this setting is that the current process topology knowledge graph is generated through the entity relationship process extraction method and the process modeling standard adaptation method. This can obtain a more refined and accurate current process topology knowledge graph, which can better serve the construction of business processes and improve the accuracy of business process automation modeling.

[0053] S130. Verify the current process topology knowledge graph through the business rule graph verification and identification method corresponding to the knowledge graph verification module, and generate the current intermediate representation parameters with semantic roles.

[0054] The current intermediate representation parameter with a semantic role may be an intermediate representation parameter including entity elements and process elements.

[0055] Optionally, the business rule graph verification and identification method corresponding to the knowledge graph verification module is used to verify the current process topology knowledge graph and generate current intermediate representation parameters with semantic roles, including: obtaining and obtaining standard business rules and business rule graphs corresponding to the standard business rules based on business rule standard data; verifying the compliance of the current process topology knowledge graph through business rule graph verification and identification methods according to the business rule graph; if the current process topology knowledge graph meets the compliance requirements, identifying process elements and entity elements of the current multimodal input intermediate representation description data to generate current intermediate representation parameters with semantic roles.

[0056] The business rule standard data may be industry compliance rules corresponding to different industries, and may also include a business term mapping verb vocabulary.

[0057] In this embodiment, the current process topology knowledge graph can be matched with each subgraph in the business rule graph and a compliance check can be performed to determine whether the current process topology knowledge graph meets compliance requirements. Furthermore, the current multimodal input intermediate representation description data needs to be identified for process and entity elements. For example, by identifying process and entity elements such as approval levels and remote processing branches, the current semantic role intermediate representation parameters can be obtained.

[0058] The benefits of this setting are: verification of the compliance of the current process topology knowledge graph, and verification through the generation process of the current intermediate representation parameters with semantic roles, which can ensure that the current process topology knowledge graph meets the compliance requirements, and then generate more accurate intermediate representation parameters, in order to better automate the modeling of business processes and improve the accuracy of the current automated modeling business processes.

[0059] S140. Describe the current intermediate representation parameters with semantic roles through the dynamic model generation method corresponding to the dynamic model generation module, and generate the current automated modeling business process in combination with the historical business process modeling template, and feedback the current automated modeling business process to the user.

[0060] The dynamic model generation method may be a method for dynamically and automatically modeling a business process. The historical business process modeling template may be a template for generating multiple business process models based on different modeling data. The current automated modeling business process may be in an Extensible Markup Language file format.

[0061] In this embodiment, the current semantic role intermediate representation parameters are analyzed to calculate similarity with each of the pre-built historical business process modeling templates. Based on the similarity, the template with the highest similarity is selected and used to automatically model the business process for the current semantic role intermediate representation parameters. Furthermore, the current automated business process can be modeled and feedback processing can be performed.

[0062] Optionally, the scenario service layer includes an interactive feedback module; after the current automated modeling business process is fed back to the user, it also includes: receiving the current user correction instruction in real time through the interactive feedback module, parsing the current user correction instruction, and obtaining the correction instruction parsing result; according to the correction instruction parsing result, and in combination with the business rule library, the current automated modeling business process is automatically adjusted to obtain and feedback the current automated modeling business adjustment process.

[0063] In this embodiment, after obtaining the current automated modeling business process and providing feedback to the user, the user can parse the extensible markup language file format based on the received current automated modeling business process, and adjust the generated current business process model through the drag and drop interface, and analyze the current business process model to record correction actions, such as changing serial approval to parallel approval.

[0064] Furthermore, based on the recorded correction actions, the current user correction instructions are generated and fed back, and then based on the correction instructions, the logical conflict nodes between the business rule library and the current automated modeling business process can be compared in real time, and the current automated modeling business process can be automatically readjusted to achieve the optimization processing operation of the current automated modeling business process. The current automated modeling business adjustment process can be obtained, which can realize the automated modeling and optimization adjustment operations of the business process more flexibly and accurately.

[0065] In addition, the correction actions recorded by the users may be clustered to automatically generate different correction action candidate rules.

[0066] The advantage of this setting is that the current automated modeling business process can be automatically adjusted and corrected. In this way, the current automated modeling business process can be corrected through the current user correction instructions received. In this way, the accuracy of the current automated modeling business process is higher, which can improve the rationality of resource allocation.

[0067] The technical solution of the embodiment of the present invention is to receive and parse the current multimodal input description data in real time through the multimodal input parsing module in a pre-established business process automation modeling system architecture to obtain the current multimodal input intermediate representation description data; extract the current multimodal input intermediate representation description data through the entity relationship process extraction method in the large language model semantic understanding module, and generate the current process topology knowledge graph in combination with the pre-built process modeling standard adaptation method; verify the current process topology knowledge graph through the business rule graph verification and recognition method corresponding to the knowledge graph verification module, and generate the current semantic role intermediate representation parameters; describe the current semantic role intermediate representation parameters through the dynamic model generation method corresponding to the dynamic model generation module, and generate the current automated modeling business process in combination with the historical business process modeling template, and feedback the current automated modeling business process to the user. This solves the problems of high reliance on manual operation, static characteristics and rigid data input form, improves the flexibility, scalability and automation of business process modeling, reduces the burden of manual processing, and improves the accuracy of business process automation modeling.

[0068] Example 2

[0069] Figure 2 This is a detailed flowchart of a business process automation modeling method provided according to Example 2 of the present invention. This embodiment is refined based on the above embodiments. In this embodiment, the dynamic model generation method corresponding to the dynamic model generation module is described, the current intermediate representation parameters with semantic roles are described, and combined with the historical business process modeling template, the current automated modeling business process is generated for further refinement.

[0070] S210: In a pre-established business process automation modeling system architecture, the multimodal input parsing module receives and parses the current multimodal input description data in real time to obtain the current multimodal input intermediate representation description data.

[0071] S220. Extract the current multimodal input intermediate representation description data through the entity relationship process extraction method in the large language model semantic understanding module, and generate the current process topology knowledge graph in combination with the pre-built process modeling standard adaptation method.

[0072] S230. Verify the current process topology knowledge graph through the business rule graph verification and identification method corresponding to the knowledge graph verification module, and generate the current intermediate representation parameters with semantic roles.

[0073] S240: Acquire historical modeling data, and generate the historical business process modeling template through a preset automatic clustering algorithm.

[0074] Among them, historical modeling data can be descriptive data of processes in different fields. For example, historical modeling data can be financial risk control process data or credit approval process data.

[0075] In this embodiment, the historical business process modeling template may be a modeling template obtained by clustering different historical modeling data. For example, the historical business process modeling template may include a financial risk control process modeling template or a credit approval process modeling template.

[0076] S250 , combining the historical business process modeling template and describing the current intermediate representation parameters with semantic roles through a dynamic model generation method to determine a target business process modeling template.

[0077] Optionally, the target business process modeling template is determined by describing the current intermediate representation parameters with semantic roles through a dynamic model generation method in combination with the historical business process modeling template, including: describing the current intermediate representation parameters with semantic roles through a dynamic model generation method to obtain the description results of the current intermediate representation parameters with semantic roles; calculating the template matching similarities between the current intermediate representation parameter description results with semantic roles and each template in the historical business process modeling template through a preset similarity matching calculation method to obtain the matching similarities of each current template; among the matching similarities of each current template, selecting the template corresponding to the largest current template matching similarity and determining it as the target business process modeling template.

[0078] Among them, the current parameter description result with semantic role intermediate representation can be the parameter description result of generating the purpose, definition and scope description of the optimal process structure and process elements (such as business areas, business activities and business tasks, etc.) using a large language model.

[0079] In this embodiment, it is necessary to first calculate the template matching similarity between the current intermediate representation parameter description result with semantic roles and each template in the historical business process modeling template. The different template matching similarities corresponding to each template can be calculated, and then the template with the highest template matching similarity can be selected, indicating that it is the most matched with the current intermediate representation parameter description result with semantic roles, and then the automated modeling operation can be performed according to the target business process modeling template.

[0080] The advantage of this setting is that the target business process modeling template is determined by calculating the similarity with each template in the historical business process modeling template, which can save the time of business process automatic modeling and improve the accuracy, efficiency and flexibility of business process automatic modeling.

[0081] S260: Model the current intermediate representation parameters with semantic roles according to the target business process modeling template to generate the current automated modeling business process.

[0082] In this embodiment, after the target business process modeling template is determined, the current semantic role intermediate representation parameters are filled and automatically modeled according to the template to obtain the current automatically modeled business process.

[0083] The advantage of this setting is that the process of generating the current automated modeling business process through a dynamic model generation method improves the accuracy and efficiency of automatic modeling of business processes, reduces labor costs, and increases the reuse rate of templates.

[0084] S270: Feedback the current automated modeling business process to the user.

[0085] The technical solution of the embodiment of the present invention determines the target business process modeling template by performing similarity calculation with each template in the historical business process modeling template, and generates the process of the current automated modeling business process through a dynamic model generation method. This can save the time of business process automated modeling, improve the accuracy, efficiency and flexibility of business process automated modeling, reduce labor costs, and increase the reuse rate of templates.

[0086] Example 3

[0087] Figure 3 This is a schematic diagram of the structure of a business process automation modeling device provided in the third embodiment of the present invention. The business process automation modeling device provided in this embodiment can be implemented by software and / or hardware, and can be configured in a terminal device or server to implement a business process automation modeling method in the embodiment of the present invention. Figure 3 As shown, the device includes: a current multimodal input intermediate representation description data determination module 310, a current process topology knowledge graph generation module 320, a current semantic role intermediate representation parameter generation module 330 and a current automated modeling business process generation and feedback module 340.

[0088] The current multimodal input intermediate representation description data determination module 310 is configured to receive and parse the current multimodal input description data in real time through the multimodal input parsing module in a pre-established business process automation modeling system architecture to obtain the current multimodal input intermediate representation description data.

[0089] The business process automation modeling system architecture includes a scenario service layer, a basic service layer, and a large language model layer; the scenario service layer includes a multimodal input parsing module; the basic service layer includes a dynamic model generation module; the large language model layer includes a knowledge graph verification module and a large language model semantic understanding module;

[0090] The current process topology knowledge graph generation module 320 is used to extract the current multimodal input intermediate representation description data through the entity relationship process extraction method in the large language model semantic understanding module, and generate the current process topology knowledge graph in combination with the pre-built process modeling standard adaptation method;

[0091] The current semantic role intermediate representation parameter generation module 330 is used to verify the current process topology knowledge graph through the business rule graph verification and identification method corresponding to the knowledge graph verification module, and generate the current semantic role intermediate representation parameters;

[0092] The current automated modeling business process generation and feedback module 340 is used to describe the current intermediate representation parameters with semantic roles through the dynamic model generation method corresponding to the dynamic model generation module, and combine the historical business process modeling template to generate the current automated modeling business process, and feedback the current automated modeling business process to the user.

[0093] The technical solution of the embodiment of the present invention is to receive and parse the current multimodal input description data in real time through the multimodal input parsing module in a pre-established business process automation modeling system architecture to obtain the current multimodal input intermediate representation description data; extract the current multimodal input intermediate representation description data through the entity relationship process extraction method in the large language model semantic understanding module, and generate the current process topology knowledge graph in combination with the pre-built process modeling standard adaptation method; verify the current process topology knowledge graph through the business rule graph verification and recognition method corresponding to the knowledge graph verification module, and generate the current semantic role intermediate representation parameters; describe the current semantic role intermediate representation parameters through the dynamic model generation method corresponding to the dynamic model generation module, and generate the current automated modeling business process in combination with the historical business process modeling template, and feedback the current automated modeling business process to the user. This solves the problems of high reliance on manual operation, static characteristics and rigid data input form, improves the flexibility, scalability and automation of business process modeling, reduces the burden of manual processing, and improves the accuracy of business process automation modeling.

[0094] Based on the above embodiments, the scene service layer includes an interactive feedback module.

[0095] On the basis of the above embodiments, it also includes a current automated modeling business adjustment process determination and feedback module, which can be specifically used to, after the current automated modeling business process is fed back to the user, receive the current user correction instruction in real time through the interactive feedback module, parse the current user correction instruction, and obtain the correction instruction parsing result; according to the correction instruction parsing result, and in combination with the business rule library, automatically adjust the current automated modeling business process to obtain and feedback the current automated modeling business adjustment process.

[0096] Based on the above embodiments, the current multimodal input intermediate representation description data determination module 310 can be specifically used to: parse and process the current multimodal input description data through a pre-set unified intermediate representation method to obtain the data parsing intermediate representation corresponding to each modality; perform feature alignment processing on each data parsing intermediate representation through a pre-set cross-modal alignment loss function to obtain each data parsing feature aligned intermediate representation; and fuse the data parsing feature aligned intermediate representations to obtain the current multimodal input intermediate representation description data.

[0097] Based on the above embodiments, the current process topology knowledge graph generation module 320 can be specifically used to: extract the current multimodal input intermediate representation description data through the entity relationship process extraction method to obtain each current entity, current entity relationship and current process logic; according to the process modeling standard adaptation method, adapt the process flow diagram meta-semantics in the historical modeling standard to the process modeling methodology to obtain the process diagram meta-semantic logical constraints; combine the process diagram meta-semantic logical constraints to construct knowledge graphs for each current entity, current entity relationship and current process logic to generate the current process topology knowledge graph.

[0098] Based on the above embodiments, the current intermediate representation parameter generation module 330 with semantic roles can be specifically used to: obtain and obtain standard business rules and business rule graphs corresponding to the standard business rules based on business rule standard data; verify the compliance of the current process topology knowledge graph through business rule graph verification and identification methods according to the business rule graph; if the current process topology knowledge graph meets the compliance requirements, identify the process elements and entity elements of the current multimodal input intermediate representation description data to generate the current intermediate representation parameters with semantic roles.

[0099] Based on the above embodiments, the current automated modeling business process generation and feedback module 340 can specifically include: a historical business process modeling template generation unit, which can be specifically used to: obtain historical modeling data, and generate the historical business process modeling template through a pre-set automatic clustering algorithm; a target business process modeling template determination unit, which can be specifically used to: combine the historical business process modeling template, and use a dynamic model generation method to describe the current intermediate representation parameters with semantic roles to determine the target business process modeling template; the current automated modeling business process generation unit can be specifically used to: model the current intermediate representation parameters with semantic roles according to the target business process modeling template to generate the current automated modeling business process.

[0100] On the basis of the above embodiments, the target business process modeling template determination unit can also be specifically used to: describe the current intermediate representation parameters with semantic roles through a dynamic model generation method to obtain the current intermediate representation parameter description results with semantic roles; calculate the template matching similarities between the current intermediate representation parameter description results with semantic roles and each template in the historical business process modeling template through a preset similarity matching calculation method to obtain each current template matching similarity; among each current template matching similarity, select the template corresponding to the largest current template matching similarity and determine it as the target business process modeling template.

[0101] The business process automation modeling device provided in the embodiment of the present invention can execute the business process automation modeling method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0102] Example 4

[0103] Figure 4 A schematic diagram of the structure of an electronic device 10 that can be used to implement the fourth embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0104] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0105] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0106] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors that run machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the business process automation modeling method.

[0107] In some embodiments, the business process automation modeling method can be implemented as a computer program tangibly embodied in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the business process automation modeling method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to execute the business process automation modeling method in any other appropriate manner (e.g., via firmware).

[0108] The method includes: in a pre-established business process automation modeling system architecture, receiving and parsing current multimodal input description data in real time through the multimodal input parsing module to obtain current multimodal input intermediate representation description data; wherein, the business process automation modeling system architecture includes a scenario service layer, a basic service layer and a large language model layer; the scenario service layer includes a multimodal input parsing module; the basic service layer includes a dynamic model generation module; the large language model layer includes a knowledge graph verification module and a large language model semantic understanding module; through the entity relationship process extraction method in the large language model semantic understanding module, The current multimodal input intermediate representation description data is extracted, and combined with the pre-built process modeling standard adaptation method, the current process topology knowledge graph is generated; the current process topology knowledge graph is verified by the business rule graph verification and identification method corresponding to the knowledge graph verification module, and the current intermediate representation parameters with semantic roles are generated; the current intermediate representation parameters with semantic roles are described by the dynamic model generation method corresponding to the dynamic model generation module, and combined with the historical business process modeling template, the current automated modeling business process is generated, and the current automated modeling business process is fed back to the user.

[0109] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0110] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0111] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0112] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0113] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0114] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0115] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0116] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

[0117] Example 5

[0118] The fifth embodiment of the present invention also provides a computer-readable storage medium, wherein the computer-readable instructions are used to execute a business process automation modeling method when executed by a computer processor, the method comprising: in a pre-established business process automation modeling system architecture, receiving and parsing the current multimodal input description data in real time through the multimodal input parsing module to obtain the current multimodal input intermediate representation description data; wherein the business process automation modeling system architecture includes a scenario service layer, a basic service layer and a large language model layer; the scenario service layer includes a multimodal input parsing module; the basic service layer includes a dynamic model generation module; the large language model layer includes a knowledge graph verification module and a large language model semantic understanding module. Module; through the entity relationship process extraction method in the large language model semantic understanding module, the current multimodal input intermediate representation description data is extracted, and combined with the pre-built process modeling standard adaptation method, the current process topology knowledge graph is generated; through the business rule graph verification and identification method corresponding to the knowledge graph verification module, the current process topology knowledge graph is verified, and the current intermediate representation parameters with semantic roles are generated; through the dynamic model generation method corresponding to the dynamic model generation module, the current intermediate representation parameters with semantic roles are described, and combined with the historical business process modeling template, the current automated modeling business process is generated, and the current automated modeling business process is fed back to the user.

[0119] Of course, the computer-readable storage medium provided in the embodiment of the present invention has computer-executable instructions that are not limited to the method operations described above, and can also execute related operations in the business process automation modeling provided in any embodiment of the present invention.

[0120] Through the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented with the help of software and necessary general-purpose hardware, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0121] It is worth noting that in the above-mentioned embodiment of business process automation modeling, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0122] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A business process automation modeling method, characterized in that: include: In a pre-established business process automation modeling system architecture, the multimodal input parsing module receives and parses the current multimodal input description data in real time to obtain the current multimodal input intermediate representation description data; The business process automation modeling system architecture includes a scenario service layer, a basic service layer, and a large language model layer; the scenario service layer includes a multimodal input parsing module; the basic service layer includes a dynamic model generation module; the large language model layer includes a knowledge graph verification module and a large language model semantic understanding module; The entity relationship process extraction method in the semantic understanding module of the large language model is used to extract the current multimodal input intermediate representation description data, and combined with the pre-built process modeling standard adaptation method, the current process topology knowledge graph is generated; The current process topology knowledge graph is verified by the business rule graph verification and identification method corresponding to the knowledge graph verification module, and the current intermediate representation parameters with semantic roles are generated; The dynamic model generation method corresponding to the dynamic model generation module is used to describe the current intermediate representation parameters with semantic roles, and combined with the historical business process modeling template, the current automated modeling business process is generated, and the current automated modeling business process is fed back to the user.

2. The method according to claim 1, characterized in that The scene service layer includes an interactive feedback module; After providing feedback of the current automated modeling business process to the user, the method further includes: The interactive feedback module receives the current user correction instruction in real time, analyzes the current user correction instruction, and obtains the correction instruction analysis result; According to the analysis result of the correction instruction and in combination with the business rule library, the current automated modeling business process is automatically adjusted to obtain and feed back the current automated modeling business adjustment process.

3. The method according to claim 1, characterized in that The multimodal input parsing module receives and parses the current multimodal input description data in real time to obtain the current multimodal input intermediate representation description data, including: Parsing the current multimodal input description data using a pre-set unified intermediate representation method to obtain data parsing intermediate representations corresponding to each modality; Through the pre-set cross-modal alignment loss function, feature alignment processing is performed on each data parsing intermediate representation to obtain the feature-aligned intermediate representation of each data parsing; The data parsing features are aligned with the intermediate representation and fused to obtain the current multimodal input intermediate representation description data.

4. The method according to claim 3, characterized in that The entity relationship process extraction method in the semantic understanding module of the large language model is used to extract the current multimodal input intermediate representation description data, and combined with the pre-built process modeling standard adaptation method to generate the current process topology knowledge graph, including: Extracting the current multimodal input intermediate representation description data using the entity relationship process extraction method to obtain each current entity, current entity relationship, and current process logic; According to the process modeling standard adaptation method, the process modeling methodology is adapted to the process flow meta-semantics in the historical modeling standard to obtain the process flow diagram meta-semantic logical constraints; Combined with the semantic logical constraints of the process graph elements, knowledge graphs are constructed for each current entity, current entity relationship and current process logic to generate the current process topology knowledge graph.

5. The method according to claim 4, characterized in that The business rule graph verification and identification method corresponding to the knowledge graph verification module verifies the current process topology knowledge graph and generates the current semantic role intermediate representation parameters, including: Obtaining and obtaining standard business rules and a business rule graph corresponding to the standard business rules based on business rule standard data; According to the business rule graph, the compliance of the current process topology knowledge graph is verified by a business rule graph verification and identification method; If the current process topology knowledge graph meets the compliance requirements, the process elements and entity elements of the current multimodal input intermediate representation description data are identified to generate the current intermediate representation parameters with semantic roles.

6. The method according to claim 5, characterized in that The dynamic model generation method corresponding to the dynamic model generation module is used to describe the current intermediate representation parameters with semantic roles, and combined with the historical business process modeling template to generate the current automated modeling business process, including: Acquire historical modeling data and generate the historical business process modeling template through a pre-set automatic clustering algorithm; In combination with the historical business process modeling template, the current intermediate representation parameters with semantic roles are described through a dynamic model generation method to determine the target business process modeling template; The current intermediate representation parameters with semantic roles are modeled according to the target business process modeling template to generate the current automated modeling business process.

7. The method according to claim 6, characterized in that The method of combining the historical business process modeling template and describing the current intermediate representation parameters with semantic roles through a dynamic model generation method to determine the target business process modeling template includes: The current intermediate representation parameters with semantic roles are described by a dynamic model generation method to obtain the description result of the current intermediate representation parameters with semantic roles; By using a preset similarity matching calculation method, the template matching similarity between the current semantic role intermediate representation parameter description result and each template in the historical business process modeling template is calculated to obtain the current template matching similarity; Among the current template matching similarities, a template corresponding to the greatest current template matching similarity is selected and determined as the target business process modeling template.

8. A business process automation modeling device, characterized in that: include: A current multimodal input intermediate representation description data determination module is used to receive and parse the current multimodal input description data in real time through the multimodal input parsing module in a pre-established business process automation modeling system architecture to obtain the current multimodal input intermediate representation description data; The business process automation modeling system architecture includes a scenario service layer, a basic service layer, and a large language model layer; the scenario service layer includes a multimodal input parsing module; the basic service layer includes a dynamic model generation module; the large language model layer includes a knowledge graph verification module and a large language model semantic understanding module; A current process topology knowledge graph generation module is used to extract the current multimodal input intermediate representation description data through the entity relationship process extraction method in the large language model semantic understanding module, and generate the current process topology knowledge graph in combination with a pre-built process modeling standard adaptation method; A current intermediate representation parameter generation module with semantic roles is used to verify the current process topology knowledge graph through the business rule graph verification and identification method corresponding to the knowledge graph verification module, and generate the current intermediate representation parameter with semantic roles; The current automated modeling business process generation and feedback module is used to describe the current intermediate representation parameters with semantic roles through the dynamic model generation method corresponding to the dynamic model generation module, and combine the historical business process modeling template to generate the current automated modeling business process, and feedback the current automated modeling business process to the user.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, it implements a business process automation modeling method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement a business process automation modeling method according to any one of claims 1 to 7 when executed.

Images