Workflow configuration method and device, equipment and medium

Abstract

The invention provides a workflow configuration method and device, equipment and a medium, and can be applied to the technical field of artificial intelligence. The method comprises the following steps: in response to a received workflow configuration demand, determining a target historical workflow matched with the workflow configuration demand from a plurality of historical workflows; according to the workflow configuration requirement and the target historical workflow, searching configuration knowledge information of a plurality of preset nodes stored in a configuration knowledge base to obtain a target configuration knowledge information set; and processing the workflow configuration requirement and the target configuration knowledge information set by utilizing a pre-training language model to obtain a workflow configuration file, so as to configure the workflow based on the workflow configuration file.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of artificial intelligence, and more particularly to a workflow configuration method, device, equipment and medium. BACKGROUND

[0002] A workflow configuration platform generally allows users to quickly build an application or an automation workflow by dragging and connecting different function nodes. After a user performs visual configuration of a workflow on the workflow configuration platform, the workflow configuration platform automatically generates a configuration file of the workflow and executes the application or the workflow according to the configuration file.

[0003] However, when configuring a workflow using the workflow configuration platform, the user generally needs to learn the functions of various nodes provided by the platform and manually add, configure and connect the nodes one by one, which is low in efficiency and accuracy. SUMMARY

[0004] In view of the above problems, the present disclosure provides a workflow configuration method, device, equipment and medium.

[0005] According to a first aspect of the present disclosure, a workflow configuration method is provided, comprising: in response to receiving a workflow configuration requirement, determining a target historical workflow matching the workflow configuration requirement from a plurality of historical workflows; retrieving configuration knowledge information of a plurality of preset nodes respectively stored in a configuration knowledge base according to the workflow configuration requirement and the target historical workflow to obtain a target configuration knowledge information set; processing the workflow configuration requirement and the target configuration knowledge information set by using a pre-trained language model to obtain a workflow configuration file, so as to configure a workflow based on the workflow configuration file.

[0006] According to an embodiment of the present disclosure, the retrieving of the configuration knowledge information of the plurality of preset nodes respectively stored in the configuration knowledge base according to the workflow configuration requirement and the target historical workflow to obtain the target configuration knowledge information set comprises: determining a plurality of first target nodes included in the target historical workflow; retrieving the configuration knowledge information of the plurality of first target nodes respectively from the configuration knowledge base to obtain a first target configuration knowledge information subset matching the target historical workflow; retrieving the configuration knowledge information of the plurality of preset nodes respectively based on the workflow configuration requirement to obtain a second target configuration knowledge information subset matching the workflow configuration requirement; and obtaining the target configuration knowledge information set according to the first target configuration knowledge information subset and the second target configuration knowledge information subset.

[0007] According to an embodiment of the present disclosure, the configuration knowledge information comprises node function information and node dependency information; and the retrieving of the configuration knowledge information of each of the plurality of preset nodes based on the workflow configuration requirement to obtain a second target configuration knowledge information subset matched with the workflow configuration requirement comprises: matching the workflow configuration requirement with the node function information of each of the preset nodes to determine at least one second target node whose node function information is matched with the workflow configuration requirement; determining, according to the node dependency information of each of the second target nodes, a third target node from the plurality of preset nodes that has a dependency relationship with the second target node; and obtaining the second target configuration knowledge information subset according to the configuration knowledge information of each of the second target nodes and each of the third target nodes.

[0008] According to an embodiment of the present disclosure, the determining of a target historical workflow matched with the workflow configuration requirement from a plurality of historical workflows comprises: performing keyword extraction on the workflow configuration requirement to obtain a scene keyword and a function keyword; and determining, according to the scene keyword and the function keyword, a target historical workflow matched with the workflow configuration requirement from the plurality of historical workflows.

[0009] According to an embodiment of the present disclosure, the determining of a target historical workflow matched with the workflow configuration requirement from a plurality of historical workflows according to the scene keyword and the function keyword comprises: obtaining scene information and function information of each of the historical workflows; determining a first matching degree of the workflow configuration information and each of the historical workflows according to the scene keyword and the scene information of each of the historical workflows; determining a second matching degree of the workflow configuration information and each of the historical workflows according to the function keyword and the function information of each of the historical workflows; and determining a target historical workflow from the plurality of historical workflows according to the first matching degree and the second matching degree of the workflow configuration information and each of the historical workflows.

[0010] According to an embodiment of the present disclosure, the processing of the workflow configuration requirement and the target configuration knowledge information set by using a pre-trained language model to obtain a workflow configuration file comprises: inputting the workflow configuration requirement and the target configuration knowledge information set into a first prompt template to obtain first prompt information, the first prompt information being used to guide the pre-trained language model to generate a workflow configuration file according to the workflow configuration requirement and the target configuration knowledge information set; inputting the first prompt information into the pre-trained language model to generate an initial workflow configuration file; performing format verification on the initial workflow configuration file to obtain a verification result; and in a case where the verification result indicates that the initial workflow configuration file passes the verification, determining that the initial workflow configuration file is the workflow configuration file.

[0011] According to an embodiment of the present disclosure, the workflow configuration method further comprises: in the case that the above-mentioned check result represents that the above-mentioned initial workflow configuration file fails to pass the check, inputting the above-mentioned check result and the above-mentioned initial workflow configuration file into a second prompt template to obtain second prompt information, the above-mentioned second prompt information being used to guide the above-mentioned pre-trained language model to modify the format of the above-mentioned initial workflow configuration file according to the above-mentioned check result; and inputting the above-mentioned second prompt information into the above-mentioned pre-trained language model to obtain the above-mentioned workflow configuration file.

[0012] The second aspect of the present disclosure provides a workflow configuration device, characterized in that the device comprises: a history matching module, configured to determine a target historical workflow matching a workflow configuration requirement from a plurality of historical workflows in response to receiving the workflow configuration requirement; a knowledge retrieval module, configured to retrieve configuration knowledge information of each of a plurality of preset nodes stored in a configuration knowledge base according to the workflow configuration requirement and the target historical workflow to obtain a target configuration knowledge information set; and a configuration generation module, configured to process the workflow configuration requirement and the target configuration knowledge information set by using a pre-trained language model to obtain a workflow configuration file, so as to configure a workflow based on the workflow configuration file.

[0013] The third aspect of the present disclosure provides an electronic device, comprising: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above-mentioned method.

[0014] The fourth aspect of the present disclosure further provides a computer-readable storage medium having a computer program or instructions stored thereon, wherein the computer program or instructions are executed by a processor to implement the steps of the above-mentioned method.

[0015] The fifth aspect of the present disclosure further provides a computer program product comprising a computer program or instructions, wherein the computer program or instructions are executed by a processor to implement the steps of the above-mentioned method.

[0016] According to an embodiment of the present disclosure, by acquiring a target configuration knowledge information set related to a workflow configuration requirement, simulating a user's learning process of specific use and configuration method of each node provided by the platform, and generating a workflow configuration file by using a pre-trained language model, the user does not need to manually add, configure and connect nodes, thereby automatically generating a workflow configuration file and improving the generation efficiency of the workflow configuration file. Moreover, by taking a target historical workflow as reference information retrieved by the knowledge base for enhancement, the influence caused by the user's missing information when describing the workflow configuration requirement is reduced, the relevant target configuration knowledge information can be more comprehensively obtained from the knowledge base, and the accuracy of generating the workflow configuration file by using the pre-trained language model is improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] The above and other objects, features and advantages of the present disclosure will become more apparent from the following description when taken in conjunction with the accompanying drawings, in which:

[0018] Figure 1 An application scenario diagram of a workflow configuration method, apparatus, device and medium according to an embodiment of the present disclosure is schematically shown;

[0019] Figure 2 A flowchart of a workflow configuration method according to an embodiment of the present disclosure is schematically shown;

[0020] Figure 3 A schematic diagram of a configuration knowledge base retrieval according to an embodiment of the present disclosure is schematically shown

[0021] Figure 4 A flowchart of generating a workflow configuration file according to an embodiment of the present disclosure is schematically shown;

[0022] Figure 5 A structural block diagram of a workflow configuration apparatus according to an embodiment of the present disclosure is schematically shown; and

[0023] Figure 6 A block diagram of an electronic device suitable for implementing a workflow configuration method according to an embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0024] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present disclosure. In the following detailed description, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. However, it will be apparent to one skilled in the art that one or more embodiments can be practiced without these specific details. In addition, in the following description, descriptions of well-known structures and techniques have been omitted to avoid unnecessarily obscuring the concept of the present disclosure.

[0025] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used herein, the terms "comprises", "comprising", "includes", "including" and the like are, where used, meant to be inclusive, but not limiting in any way.

[0026] All terms used herein including technical and scientific terms have the meanings commonly understood by one of ordinary skill in the art unless otherwise defined. It should be noted that the terms used herein are to be interpreted as having a meaning that is consistent with the understanding of those terms by those skilled in the art, and should not be interpreted in an idealized or overly formal sense.

[0027] In the case of using expressions similar to "at least one of A, B, and C, etc.", it should be generally interpreted as including any one of A, B, and C, etc. (for example, "a system having at least one of A, B, and C" should include but not be limited to a system having A alone, a system having B alone, a system having C alone, a system having both A and B, a system having both A and C, a system having both B and C, and / or a system having A, B, and C, etc.).

[0028] In the technical solutions of the disclosure, the user information (including but not limited to user personal information, user image information, user equipment information, such as location information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved are information and data authorized by the user or authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of related data comply with relevant laws, regulations and standards, necessary security measures are taken, do not violate public order and good customs, and provide corresponding operation portal for user to choose authorization or refusal.

[0029] In the scenario of using personal information for automated decision-making, the method, device and system provided by the embodiments of the disclosure all provide corresponding operation portal for the user to choose to agree or refuse the automated decision-making result; if the user chooses to refuse, the expert decision-making process is entered. The expression "automated decision-making" here refers to the activity of automatically analyzing, evaluating the behavior habits, interests and hobbies, or economic, health, credit status, etc. of an individual through a computer program and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by personnel who are engaged in a certain field of work, have specialized experience, knowledge and skills, and reach a certain professional level.

[0030] A workflow orchestration platform usually allows users to quickly build and deploy applications or automated workflows by dragging and connecting different function nodes such as interface calls, data processing, logical judgments, large model calls, etc. These platforms usually use a specific domain-specific language (Domain Specific Language, DSL) or configuration file to describe the node composition, node parameters, and connection relationship between nodes of the entire application. After the user visualizes the arrangement on the platform, the platform will automatically generate or update this configuration file and execute the application according to the configuration file.

[0031] However, when using the workflow orchestration platform, the user needs to learn the operation method of the platform, the specific purpose and configuration method of each node, and how to reasonably organize the nodes to achieve the expected function, so as to have a deep understanding of the various node functions provided by the platform, which has a high learning cost. Moreover, during the workflow configuration process, the user also needs to manually add, configure and connect nodes one by one, and for complex applications, the number of nodes can be very large, the arrangement process is time-consuming and laborious, and the accuracy is low.

[0032] Embodiments of the present disclosure provide a workflow configuration method, comprising: in response to receiving a workflow configuration requirement, determining a target historical workflow matching the workflow configuration requirement from a plurality of historical workflows; retrieving configuration knowledge information of a plurality of preset nodes respectively stored in a configuration knowledge base according to the workflow configuration requirement and the target historical workflow, to obtain a target configuration knowledge information set; processing the workflow configuration requirement and the target configuration knowledge information set by using a pre-trained language model, to obtain a workflow configuration file, so as to configure the workflow based on the workflow configuration file.

[0033] Embodiments of the present disclosure simulate the user's learning process of the specific purpose and configuration method of each node provided by the platform by obtaining the target configuration knowledge information set related to the workflow configuration requirement, and generate the workflow configuration file by using the pre-trained language model, without the user manually adding, configuring and connecting nodes one by one, thereby automatically generating the workflow configuration file and improving the generation efficiency of the workflow configuration file. Moreover, by taking the target historical workflow as reference information for knowledge base retrieval enhancement, the influence caused by the user's missing information when describing the workflow configuration requirement is reduced, the relevant target configuration knowledge information can be more comprehensively obtained from the knowledge base, and the accuracy of generating the workflow configuration file by using the pre-trained language model is improved.

[0034] Figure 1 An application scenario diagram of the workflow configuration method, apparatus, device and medium according to embodiments of the present disclosure is schematically shown.

[0035] As Figure 1As shown, the application scenarios according to this embodiment can include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 can include various connection types, such as wired, wireless communication links, or optical fiber cables, and the like.

[0036] A user can use the first terminal device 101, the second terminal device 102, the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, and the like. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, and the like (only as examples).

[0037] The first terminal device 101, the second terminal device 102, the third terminal device 103 can be various electronic devices with display screens and supporting web browsing, including but not limited to smartphones, tablet computers, laptop computers, desktop computers, and the like.

[0038] The server 105 can be a server providing various services, such as a background management server supporting a website browsed by a user using the first terminal device 101, the second terminal device 102, the third terminal device 103 (only as an example). The background management server can analyze and process received user requests and the like, and feed back the processing results (such as web pages, information, or data, and the like obtained or generated according to user requests) to the terminal device.

[0039] It should be noted that the workflow configuration method provided by the embodiments of the present disclosure can generally be executed by the server 105. Accordingly, the workflow configuration apparatus provided by the embodiments of the present disclosure can generally be arranged in the server 105. The workflow configuration method provided by the embodiments of the present disclosure can also be executed by a server or a server cluster different from the server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or the server 105. Accordingly, the workflow configuration apparatus provided by the embodiments of the present disclosure can also be arranged in a server or a server cluster different from the server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or the server 105.

[0040] It should be understood that Figure 1 The number of terminal devices, networks, and servers in the above description is only illustrative. According to the implementation needs, there can be any number of terminal devices, networks, and servers.

[0041] For example, a user can use the first terminal device 101, the second terminal device 102, the third terminal device to send a workflow configuration requirement to the server 105 through the network 104, and the server 105 determines a target historical workflow matching the workflow configuration requirement from a plurality of historical workflows in response to receiving the workflow configuration requirement; retrieves configuration knowledge information of a plurality of preset nodes respectively stored in a configuration knowledge base according to the workflow configuration requirement and the target historical workflow to obtain a target configuration knowledge information set; processes the workflow configuration requirement and the target configuration knowledge information set by using a pre-trained language model to obtain a workflow configuration file, so as to configure the workflow based on the workflow configuration file.

[0042] The following will be based on Figure 1 The scenario described is configured by Figures 2-4 The workflow configuration method of the disclosed embodiment is described in detail.

[0043] Figure 2 The flowchart of the workflow configuration method according to the embodiment of the present disclosure is schematically shown.

[0044] As Figure 2 indicated, the workflow configuration method of this embodiment includes operation S210 to operation S230.

[0045] In operation S210, a target historical workflow matching the workflow configuration requirement is determined from a plurality of historical workflows in response to receiving the workflow configuration requirement.

[0046] In operation S220, configuration knowledge information of a plurality of preset nodes respectively stored in a configuration knowledge base is retrieved according to the workflow configuration requirement and the target historical workflow to obtain a target configuration knowledge information set.

[0047] In operation S230, the workflow configuration requirement and the target configuration knowledge information set are processed by using a pre-trained language model to obtain a workflow configuration file, so as to configure the workflow based on the workflow configuration file.

[0048] According to the embodiment of the present disclosure, the workflow configuration requirement can include functional requirements of a workflow to be generated, etc. In some embodiments, an interactive interface can be included in the workflow configuration platform, and the user can input the workflow configuration requirement in the form of natural language in the interactive interface.

[0049] According to the embodiment of the present disclosure, after receiving the workflow configuration requirement, since the workflow configuration requirement can not accurately and comprehensively represent the actual requirement of the user, the target historical workflow related to the workflow configuration requirement can be obtained as reference information to improve the comprehensiveness of the reference information of the generated workflow.

[0050] According to an embodiment of the present disclosure, a plurality of historical workflows that have been configured can be stored in the workflow configuration platform, and information such as function information and scene information of each historical workflow can be stored, so as to directly call the historical workflows.

[0051] In some embodiments, when the target historical workflow is determined, the plurality of historical workflows can be screened according to the workflow configuration requirement and the function information and scene information of each historical workflow, to obtain the target historical workflow that matches the function and scene represented by the workflow configuration requirement.

[0052] According to an embodiment of the present disclosure, after the target historical workflow is determined, the target historical workflow can be used as reference information for searching the configuration knowledge base, so as to more comprehensively search the related target configuration knowledge information set from the configuration knowledge base.

[0053] According to an embodiment of the present disclosure, a plurality of preset nodes of different function types can be configured in the workflow configuration platform, including data input nodes, data processing nodes, logical judgment nodes, large language model calling nodes, interface integration nodes, knowledge base searching nodes, and output nodes.

[0054] According to an embodiment of the present disclosure, the configuration knowledge base can store configuration knowledge information of each of the plurality of preset nodes of the workflow configuration platform. The configuration knowledge information may, for example, include detailed descriptions of each preset node, such as function description, input parameters, output parameters, configuration options, and usage examples of each preset node.

[0055] In another embodiment, the configuration knowledge base can also store complete format specifications, syntax structures, required field and optional field descriptions of the configuration file for describing the workflow, as well as correct and typical configuration file examples.

[0056] According to an embodiment of the present disclosure, when the configuration knowledge base is searched using the workflow configuration requirement and the target workflow, the target node related to the workflow configuration requirement and the target workflow can be determined, and the configuration knowledge information of the target node is determined as the target configuration knowledge information set.

[0057] According to an embodiment of the present disclosure, the configuration knowledge information of the target node, the complete format specifications, syntax structures, required field and optional field descriptions of the configuration file in the configuration knowledge base, and the correct and typical configuration file examples can also be combined into the target configuration knowledge information set, so as to improve the comprehensiveness of the target configuration knowledge information set.

[0058] According to an embodiment of the present disclosure, after the target configuration knowledge information set is obtained, the workflow configuration requirement and the target configuration knowledge information set can be input into the pre-trained language model to generate the workflow configuration file.

[0059] According to an embodiment of the present disclosure, the pre-trained language model can be obtained by fine-tuning a large language model using the sample workflow configuration requirement, the sample target configuration knowledge information set, and the sample workflow configuration file. Since the large language model has strong natural language understanding capability, the large language model can be used to understand the target configuration knowledge information set, simulate the user's understanding of the functions and configuration methods of each node, and convert the workflow configuration requirement into a workflow.

[0060] According to an embodiment of the present disclosure, after generating the workflow configuration file, the workflow configuration platform can directly parse the workflow configuration file and automatically render the corresponding workflow. The user can fine-tune or directly run the workflow based on the workflow.

[0061] According to an embodiment of the present disclosure, by obtaining the target configuration knowledge information set related to the workflow configuration requirement, simulating the user's learning process of the specific purposes and configuration methods of each node provided by the platform, and generating the workflow configuration file by using the pre-trained language model, the user does not need to manually add, configure, and connect nodes one by one, thereby automatically generating the workflow configuration file and improving the generation efficiency of the workflow configuration file. Moreover, by taking the target historical workflow as reference information for knowledge base retrieval enhancement, the influence caused by the missing information of the user when describing the workflow configuration requirement is reduced, the related target configuration knowledge information can be more comprehensively obtained from the knowledge base, and the accuracy of generating the workflow configuration file by using the pre-trained language model is improved.

[0062] According to an embodiment of the present disclosure, determining the target historical workflow matching the workflow configuration requirement from the plurality of historical workflows includes: extracting keywords in the workflow configuration requirement to obtain scene keywords and function keywords; and determining the target historical workflow matching the workflow configuration requirement from the plurality of historical workflows according to the scene keywords and the function keywords.

[0063] According to an embodiment of the present disclosure, when determining the target historical workflow matching the workflow configuration requirement, the key information in the workflow configuration requirement can be extracted to match the target historical workflow by using the key information.

[0064] According to an embodiment of the present disclosure, the key information of the workflow configuration requirement can include scene keywords and function keywords. Specifically, the scene keywords can represent the application scenario of the to-be-generated workflow, such as being applied to enterprise attendance management, and the function keywords can represent the actual function of the to-be-generated workflow, such as leave process.

[0065] According to an embodiment of the present disclosure, when keyword extraction is performed on the workflow configuration requirement, a preset keyword library can be used to match the scenario keywords and the function keywords in the workflow configuration requirement.

[0066] According to another embodiment of the present disclosure, a large language model can also be used to extract the scenario keywords and the function keywords in the workflow configuration requirement, and no limitation is made in this regard.

[0067] According to an embodiment of the present disclosure, after the scenario keywords and the function keywords are determined, a target historical workflow that matches the scenario keywords and the function keywords can be matched from a plurality of historical workflows.

[0068] According to an embodiment of the present disclosure, by using the scenario keywords and the function keywords, a target historical workflow that matches the workflow configuration requirement can be matched from a plurality of historical workflows, so that the matching degree of the target historical workflow and the workflow configuration requirement is higher, and thus the accuracy of configuration knowledge retrieval using the target historical workflow is improved.

[0069] According to an embodiment of the present disclosure, determining a target historical workflow that matches the workflow configuration requirement from a plurality of historical workflows according to the scenario keywords and the function keywords includes: obtaining scenario information and function information of each historical workflow respectively; determining a first matching degree of the workflow configuration information and each historical workflow according to the scenario keywords and the scenario information of each historical workflow; determining a second matching degree of the workflow configuration information and each historical workflow according to the function keywords and the function information of each historical workflow; and determining the target historical workflow from the plurality of historical workflows according to the first matching degree and the second matching degree of the workflow configuration information and each historical workflow.

[0070] According to an embodiment of the present disclosure, the scenario information and the function information of each historical workflow can be obtained from a workflow configuration platform, and a target historical workflow that matches the scenario keywords and the function keywords can be determined from a plurality of historical workflows based on the scenario information and the function information of each historical workflow.

[0071] According to an embodiment of the present disclosure, the matching degree of the scenario keywords and the scenario information of the historical workflow can be determined as a first matching degree, and the matching degree of the function keywords and the function information of the historical workflow can be determined as a second matching degree.

[0072] According to embodiments of the present disclosure, different weights can be respectively set for the first matching degree and the second matching degree to represent that the scene and the function have different importance. In some embodiments, the first number of scene keywords and the second number of function keywords can be compared, and when the first number is greater than the first number, it indicates that the user emphasizes the scene information in the workflow configuration requirement, and therefore a higher weight is assigned to the first matching degree.

[0073] According to embodiments of the present disclosure, the comprehensive matching degree of the workflow configuration requirement and each historical workflow can be determined according to the first matching degree and the second matching degree, and the historical workflow with the highest comprehensive matching degree with the workflow configuration requirement can be selected as the target historical workflow.

[0074] According to embodiments of the present disclosure, by determining the historical workflow similar to both the scene and the function as the target workflow, the matching degree of the determined target workflow and the workflow configuration requirement is improved, and the accuracy of the target configuration knowledge information set obtained by searching the target workflow is further improved.

[0075] According to embodiments of the present disclosure, the configuration knowledge base can be constructed by the following way: obtaining unstructured documents and semi-structured documents of the workflow configuration platform, such as detailed description documents of each preset node, and performing text cleaning and chunking on these documents to obtain text blocks; performing vector embeddings on the text blocks to generate vector indexes for efficient semantic similarity retrieval. Structured information such as the hierarchical structure of DSL and the fixed association mode between nodes is extracted and organized.

[0076] According to embodiments of the present disclosure, according to the workflow configuration requirement and the target historical workflow, the configuration knowledge information of each of the plurality of preset nodes stored in the configuration knowledge base is searched to obtain a target configuration knowledge information set, including: determining a plurality of first target nodes included in the target historical workflow; retrieving the configuration knowledge information of each of the plurality of first target nodes from the configuration knowledge base to obtain a first target configuration knowledge information subset matched with the target historical workflow; based on the workflow configuration requirement, searching the configuration knowledge information of each of the plurality of preset nodes to obtain a second target configuration knowledge information subset matched with the workflow configuration requirement; and obtaining the target configuration knowledge information set according to the first target configuration knowledge information subset and the second target configuration knowledge information subset.

[0077] According to embodiments of the present disclosure, when the configuration knowledge base is searched using the workflow configuration requirement and the target historical workflow, the workflow configuration requirement and the target historical workflow can be searched respectively.

[0078] According to an embodiment of the present disclosure, since the target historical workflow matches the workflow configuration requirement, when a workflow corresponding to the workflow configuration requirement is generated, a plurality of first target nodes included in the target historical workflow can be used, and thus configuration knowledge information of each of the plurality of first target nodes can be retrieved to obtain a first target configuration knowledge information subset.

[0079] According to an embodiment of the present disclosure, after the configuration knowledge base is retrieved by using the target historical workflow, the configuration knowledge base can be retrieved by using the workflow configuration requirement. In some embodiments, configuration knowledge information matched with the workflow configuration requirement can be retrieved as a second target configuration knowledge information subset.

[0080] According to an embodiment of the present disclosure, after the first target configuration knowledge information subset and the second target configuration knowledge subset are obtained, a union of the first target configuration knowledge information subset and the second target configuration knowledge subset can be determined as a target configuration knowledge information set to ensure comprehensiveness of the target configuration knowledge information set.

[0081] According to an embodiment of the present disclosure, by respectively retrieving a first target configuration knowledge information set matched with the target historical workflow and a second target configuration knowledge information set matched with the workflow configuration requirement, and determining the target configuration knowledge information set based on the first target configuration knowledge information set and the second target configuration knowledge information set, the target configuration knowledge information set obtained is more comprehensive, and thus the accuracy of the generated workflow configuration file is improved.

[0082] According to an embodiment of the present disclosure, the configuration knowledge information includes node function information and node dependency information; based on the workflow configuration requirement, configuration knowledge information of each of a plurality of preset nodes is retrieved to obtain a second target configuration knowledge information subset matched with the workflow configuration requirement, including: matching the workflow configuration requirement with node function information of each preset node to determine at least one second target node whose node function information matches the workflow configuration requirement; according to node dependency information of each second target node, determining a third target node having a dependency relationship with the second target node from the plurality of preset nodes; and according to configuration knowledge information of each second target node and each third target node, obtaining the second target configuration knowledge information subset.

[0083] According to an embodiment of the present disclosure, the node function information can be used to represent information such as a function that can be implemented by the preset node. When the configuration knowledge base is retrieved by using the workflow configuration requirement, the retriever can be used to match the workflow configuration requirement with node function information of each preset node to find configuration knowledge information most relevant to functions, processes, and node types in the workflow configuration requirement.

[0084] For example, if the workflow configuration requirement mentions "knowledge base search", the retriever finds the definition of "knowledge base search node", the parameter configuration such as knowledge base identification parameter name, query input parameter name, and the representation in the configuration file. If the workflow configuration requirement describes a conditional branch, the retriever finds the configuration method of "conditional judgment node".

[0085] According to an embodiment of the present disclosure, the node dependency information can be used to represent other preset nodes having a dependency relationship with the preset node. Considering the dependency relationship between nodes, only the second target node can not be able to generate a complete workflow, so the third target node having a dependency relationship with the second target node can be determined at the same time, and the configuration knowledge information of each second target node and each third target node is retrieved at the same time to obtain a second target configuration information subset, so as to ensure the comprehensiveness of the second target configuration information subset.

[0086] According to an embodiment of the present disclosure, by determining the second target node involved in the workflow configuration requirement, determining the third target node having a dependency relationship with the second target node, and retrieving the configuration knowledge information of each second target node and each third target node at the same time to obtain a second target configuration information subset, the comprehensiveness of the second target configuration information subset is ensured, thereby improving the comprehensiveness of the target configuration information set, and further improving the accuracy of the generated workflow configuration file.

[0087] Figure 3 A schematic diagram of retrieving a configuration knowledge base according to an embodiment of the present disclosure is schematically shown.

[0088] As shown in Figure 3 , the first target node 320 included in the target historical workflow 310 is determined, and the configuration knowledge information of the first target node 320 is retrieved in the configuration knowledge base D1 to obtain a first target configuration knowledge information subset 330. The second target node 350 matched with the workflow configuration requirement 340 is determined, and the third target node 360 having a dependency relationship with the second target node 350 is determined. The configuration knowledge information of the second target node 350 and the third target node 360 is retrieved in the configuration knowledge base D1 to obtain a second target configuration knowledge information subset 370. According to the first target configuration knowledge information subset 330 and the second target configuration knowledge information subset 370, a target configuration knowledge information set 380 is generated.

[0089] According to an embodiment of the present disclosure, the workflow configuration requirement and the target configuration knowledge information set are processed by using a pre-trained language model to obtain a workflow configuration file, including: inputting the workflow configuration requirement and the target configuration knowledge information set into a first prompt template to obtain first prompt information, the first prompt information being used to guide the pre-trained language model to generate the workflow configuration file according to the workflow configuration requirement and the target configuration knowledge information set; inputting the first prompt information into the pre-trained language model to generate an initial workflow configuration file; performing format checking on the initial workflow configuration file to obtain a checking result; and in a case where the checking result represents that the initial workflow configuration file passes the checking, determining that the initial workflow configuration file is the workflow configuration file.

[0090] According to an embodiment of the present disclosure, when guiding the pre-trained language model to generate the workflow configuration file, the pre-trained language model can be driven by a prompt word engineering. Specifically, the workflow configuration requirement and the target configuration knowledge information can be input into a first prompt template that is pre-set to obtain first prompt information, and then the first prompt information is used to guide the pre-trained language model to generate a workflow configuration file that meets the workflow configuration requirement according to the target configuration knowledge information.

[0091] According to an embodiment of the present disclosure, the first prompt template explicitly instructs the pre-trained language model to convert the workflow generation requirement into a configuration file that can be recognized by the workflow configuration platform, and instructs the pre-trained language model to output according to a specific syntax and structure of the configuration file, for example, requiring to output a valid lightweight data interchange format configuration file.

[0092] According to an embodiment of the present disclosure, the first prompt template can be designed in a step-by-step thinking or chain-of-thought manner, guiding the pre-trained language model to first decompose the workflow generation requirement into logical steps, then select appropriate nodes for each step, then configure node parameters, and finally determine the connection between nodes.

[0093] According to an embodiment of the present disclosure, the first prompt template can also be designed in a few-shot learning manner, containing a small amount of high-quality examples of “workflow configuration requirement-corresponding configuration file segment” to guide the pre-trained language model to better understand the conversion task.

[0094] According to an embodiment of the present disclosure, the pre-trained language model receives the first prompt information and processes as follows: first, understands the workflow configuration requirement and decomposes the workflow configuration requirement into a series of operation steps conforming to the logic of the workflow configuration platform in combination with the target configuration knowledge information set. Second, selects the most suitable node for each operation step and attempts to fill in the configuration parameters of the node according to the workflow configuration requirement and the target configuration information set. For parameters that are not explicitly mentioned by the user but are required by the node, the pre-trained language model can use default values, placeholders, or obtain them by asking the user. Third, determine the connection relationship between nodes according to the execution order, conditional branching, data flow direction and the like described in the workflow configuration information, and correctly represent these connections in the generated configuration file, for example, which node's output is connected to which node's input. Finally, generate an initial workflow configuration file conforming to the specification of the target platform and structured.

[0095] According to an embodiment of the present disclosure, in order to ensure that the initial workflow configuration file is correctly formatted and can be normally deployed in the workflow configuration platform, the initial workflow configuration file can be format-verified before being output, and in the case that the initial workflow configuration file passes the verification, the initial workflow configuration file is determined to be a workflow configuration file and the workflow configuration file is output.

[0096] According to an embodiment of the present disclosure, by generating the first prompt information by using the enhanced first prompt template, the pre-trained language model can better understand the workflow configuration requirement and the target configuration knowledge set, so as to improve the accuracy of the generated initial workflow configuration file, and after generating the initial workflow configuration file, by performing format verification on the initial workflow configuration file, the format accuracy of the workflow configuration file can be ensured.

[0097] According to an embodiment of the present disclosure, the workflow configuration method further comprises: in the case that the verification result characterizes that the initial workflow configuration file does not pass the verification, inputting the verification result and the initial workflow configuration file into a second prompt template to obtain second prompt information, the second prompt information being used to guide the pre-trained language model to modify the format of the initial workflow configuration file according to the verification result; inputting the second prompt information into the pre-trained language model to obtain the workflow configuration file.

[0098] According to an embodiment of the present disclosure, in the case that the initial workflow configuration file does not pass the verification, the pre-trained language model can be used to regenerate a workflow configuration file conforming to the format requirement.

[0099] According to an embodiment of the present disclosure, the check result and the initial workflow configuration file can be input into a second prompt template to obtain second prompt information, so that when the second prompt information is input into the pre-trained language model, the pre-trained language model can correct the initial workflow configuration file according to the format problem represented by the check result, to obtain a workflow configuration file meeting the format requirement.

[0100] According to an embodiment of the present disclosure, by correcting the initial workflow configuration file that fails the check by using the pre-trained language model, the obtained workflow configuration file can meet the format requirement, and the accuracy of the workflow configuration file is improved.

[0101] Figure 4 A flowchart of generating a workflow configuration file according to an embodiment of the present disclosure is schematically shown.

[0102] As shown in Figure 4 generating a workflow configuration file includes operation S410 to operation S490.

[0103] In operation S410, a target historical workflow matching the workflow configuration requirement is determined from a plurality of historical workflows.

[0104] In operation S420, the configuration knowledge base is searched according to the workflow configuration requirement and the target historical workflow to obtain a target configuration information set.

[0105] In operation S430, the workflow configuration requirement and the target configuration information set are input into a first prompt template to obtain first prompt information.

[0106] In operation S440, the first prompt information is input into a pre-trained language model to generate an initial workflow configuration file.

[0107] In operation S450, the initial workflow configuration file is subjected to format check to obtain a check result.

[0108] In operation S460, it is determined whether the check result represents that the initial workflow configuration file passes the check. In the case that the check result represents that the initial workflow configuration file fails the check, operation S470 is performed, otherwise operation S490 is performed.

[0109] In operation S470, the check result and the initial workflow configuration file are input into a second prompt template to obtain second prompt information.

[0110] In operation S480, the second prompt information is input into the pre-trained language model to obtain a workflow configuration file.

[0111] In operation S490, the initial workflow configuration file is determined as the workflow configuration file.

[0112] Based on the above workflow configuration method, the present disclosure further provides a workflow configuration apparatus. The following will be described in combination with Figure 5 The apparatus is described in detail.

[0113] Figure 5 The structural block diagram of the workflow configuration apparatus according to the embodiment of the present disclosure is schematically shown.

[0114] As Figure 5 shown, the workflow configuration apparatus 500 of the embodiment includes a history matching module 510, a knowledge retrieval module 520 and a configuration generation module 530.

[0115] The history matching module 510 is configured to determine a target historical workflow matching the workflow configuration requirement from a plurality of historical workflows in response to receiving the workflow configuration requirement. In an embodiment, the history matching module 510 can be configured to perform the operation S210 described above, which will not be described herein again.

[0116] The knowledge retrieval module 520 is configured to retrieve configuration knowledge information of each of a plurality of preset nodes stored in the configuration knowledge base according to the workflow configuration requirement and the target historical workflow, to obtain a target configuration knowledge information set. In an embodiment, the knowledge retrieval module 520 can be configured to perform the operation S220 described above, which will not be described herein again.

[0117] The configuration generation module 530 is configured to process the workflow configuration requirement and the target configuration knowledge information set by using a pre-trained language model, to obtain a workflow configuration file, so as to configure the workflow based on the workflow configuration file. In an embodiment, the configuration generation module 530 can be configured to perform the operation S230 described above, which will not be described herein again.

[0118] According to the embodiment of the present disclosure, the knowledge retrieval module 520 includes a node determination sub-module, a first knowledge determination sub-module, a second knowledge determination sub-module and a target knowledge determination sub-module.

[0119] The node determination sub-module is configured to determine a plurality of first target nodes included in the target historical workflow.

[0120] The first knowledge determination sub-module is configured to retrieve configuration knowledge information of each of the plurality of first target nodes from the configuration knowledge base, to obtain a first target configuration knowledge information subset matching the target historical workflow.

[0121] The second knowledge determination sub-module is configured to retrieve configuration knowledge information of each of the plurality of preset nodes based on the workflow configuration requirement, to obtain a second target configuration knowledge information subset matching the workflow configuration requirement.

[0122] The target knowledge determination sub-module is configured to determine the target configuration knowledge information set according to the first target configuration knowledge information subset and the second target configuration knowledge information subset.

[0123] According to an embodiment of the present disclosure, the configuration knowledge information comprises node function information and node dependency information; and the second knowledge determination sub-module comprises a function determination unit, a dependency determination unit, and a knowledge determination unit.

[0124] The function determination unit is configured to match the workflow configuration requirement with the node function information of each preset node, and determine at least one second target node whose node function information matches the workflow configuration requirement.

[0125] The dependency determination unit is configured to determine, according to the node dependency information of each second target node, a third target node having a dependency relationship with the second target node from the plurality of preset nodes.

[0126] The knowledge determination unit is configured to determine, according to the configuration knowledge information of each second target node and each third target node, a second target configuration knowledge information subset.

[0127] According to an embodiment of the present disclosure, the historical matching module 510 comprises a key extraction sub-module and a key matching sub-module.

[0128] The key extraction sub-module is configured to extract keywords from the workflow configuration requirement to obtain a scene keyword and a function keyword.

[0129] The key matching sub-module is configured to determine, according to the scene keyword and the function keyword, a target historical workflow matching the workflow configuration requirement from the plurality of historical workflows.

[0130] According to an embodiment of the present disclosure, the key matching sub-module comprises an information acquisition unit, a first matching unit, a second matching unit, and a target matching unit.

[0131] The information acquisition unit is configured to acquire scene information and function information of each historical workflow.

[0132] The first matching unit is configured to determine, according to the scene keyword and the scene information of each historical workflow, a first matching degree of the workflow configuration information and each historical workflow.

[0133] The second matching unit is configured to determine, according to the function keyword and the function information of each historical workflow, a second matching degree of the workflow configuration information and each historical workflow.

[0134] The target matching unit is configured to determine, according to the first matching degree and the second matching degree of the workflow configuration information and each historical workflow, the target historical workflow from the plurality of historical workflows.

[0135] According to an embodiment of the present disclosure, the configuration generation module 530 comprises a prompt sub-module, a file generation sub-module, a file verification sub-module, and a file determination sub-module.

[0136] The prompt sub-module is configured to input the workflow configuration requirement and the target configuration knowledge information set into a first prompt template to obtain first prompt information, the first prompt information being used to guide the pre-trained language model to generate a workflow configuration file according to the workflow configuration requirement and the target configuration knowledge information set.

[0137] The file generation sub-module is configured to input the first prompt information into the pre-trained language model to generate an initial workflow configuration file.

[0138] The file verification sub-module is configured to perform format verification on the initial workflow configuration file to obtain a verification result.

[0139] The file determination sub-module is configured to determine the initial workflow configuration file as the workflow configuration file in a case where the verification result represents that the initial workflow configuration file passes the verification.

[0140] According to an embodiment of the present disclosure, the workflow configuration apparatus 500 further comprises a prompt generation module and a prompt correction module.

[0141] The prompt generation module is configured to input the verification result and the initial workflow configuration file into a second prompt template to obtain second prompt information in a case where the verification result represents that the initial workflow configuration file fails the verification, the second prompt information being used to guide the pre-trained language model to correct the format of the initial workflow configuration file according to the verification result.

[0142] The prompt correction module is configured to input the second prompt information into the pre-trained language model to obtain the workflow configuration file.

[0143] According to an embodiment of the present disclosure, any of the historical matching module 510, the knowledge retrieving module 520 and the configuration generating module 530 can be implemented in one module, or any of them can be split into multiple modules. Alternatively, at least part of the function of one or more of these modules can be combined with at least part of the function of other modules, and implemented in one module. According to an embodiment of the present disclosure, at least one of the historical matching module 510, the knowledge retrieving module 520 and the configuration generating module 530 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on board, a system on package, an application specific integrated circuit (ASIC), or any other reasonable way of integrating or packaging a circuit, etc. or implemented in hardware or firmware, or implemented in any one of software, hardware and firmware or in a proper combination of any of them. Alternatively, at least one of the historical matching module 510, the knowledge retrieving module 520 and the configuration generating module 530 can be at least partially implemented as a computer program module which, when executed, can perform the corresponding function.

[0144] Figure 6 A block diagram of an electronic device suitable for implementing the workflow configuration method according to an embodiment of the present disclosure is schematically shown.

[0145] As shown in Figure 6 The electronic device 600 according to an embodiment of the present disclosure includes a processor 601 which can perform various appropriate actions and processes according to programs stored in a read only memory (ROM) 602 or loaded from a storage portion 608 into a random access memory (RAM) 603. The processor 601 can include, for example, a general purpose microprocessor (e.g. a CPU), an instruction set processor and / or a related chipset and / or a special purpose microprocessor (e.g. an application specific integrated circuit (ASIC)), etc. The processor 601 can also include an on-board memory for cache use. The processor 601 can include a single processing unit or multiple processing units for performing different actions of the method processes according to embodiments of the present disclosure.

[0146] In the RAM 603, various programs and data required for the operation of the electronic device 600 are stored. The processor 601, the ROM 602, and the RAM 603 are connected to each other via the bus 604. The processor 601 performs various operations of the method flow according to the embodiments of the present disclosure by executing the programs in the ROM 602 and / or the RAM 603. It should be noted that the programs can also be stored in one or more memories other than the ROM 602 and the RAM 603. The processor 601 can also perform various operations of the method flow according to the embodiments of the present disclosure by executing the programs stored in the one or more memories.

[0147] According to embodiments of the present disclosure, the electronic device 600 can further include an input / output (I / O) interface 605, which is also connected to the bus 604. The electronic device 600 can further include one or more of the following components connected to the input / output (I / O) interface 605: an input part 606 including a keyboard, a mouse, and the like; an output part 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), and the like, and a speaker, and the like; a storage part 608 including a hard disk, and the like; and a communication part 609 including a network interface card such as a LAN card, a modem, and the like. The communication part 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the input / output (I / O) interface 605 as necessary. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, and the like is mounted on the drive 610 as necessary, so that a computer program read therefrom is installed in the storage part 608 as necessary.

[0148] The present disclosure also provides a computer readable storage medium, which can be included in the device / apparatus / system described in the above embodiments; or can exist separately without being assembled into the device / apparatus / system. The above computer readable storage medium carries one or more programs, when the one or more programs are executed, the method according to the embodiments of the present disclosure is implemented.

[0149] According to an embodiment of the present disclosure, the computer readable storage medium can be a nonvolatile computer readable storage medium, for example, can include, but is not limited to, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any appropriate combination thereof. In the present disclosure, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present disclosure, the computer readable storage medium can include one or more memories, such as the ROM 602 and / or the RAM 603 described above, and / or one or more memory chips other than the ROM 602 and the RAM 603.

[0150] Embodiments of the present disclosure also include a computer program product, which includes a computer program containing program codes for executing the methods shown in the flowcharts. When the computer program product is run in a computer system, the program codes are used to make the computer system implement the workflow configuration method provided by the embodiments of the present disclosure.

[0151] The above-described functions defined in the system / apparatus / module / unit of the embodiments of the present disclosure are performed when the computer program is executed by the processor 601. According to an embodiment of the present disclosure, the system, apparatus, module, unit, etc. described above can be implemented by computer program modules.

[0152] In one embodiment, the computer program can rely on a tangible storage medium, such as an optical storage device, a magnetic storage device, etc. In another embodiment, the computer program can also be transmitted, distributed, and downloaded in the form of a signal on a network medium, and be downloaded and installed through the communication part 609, and / or be installed from the detachable medium 611. The program codes contained in the computer program can be transmitted by any appropriate network medium, including but not limited to wireless, wired, etc., or any appropriate combination thereof.

[0153] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 609, and / or be installed from the detachable medium 611. When the computer program is executed by the processor 601, the above-described functions defined in the system of the embodiments of the present disclosure are performed. According to an embodiment of the present disclosure, the system, device, apparatus, module, unit, etc. described above can be implemented by computer program modules.

[0154] According to embodiments of the present disclosure, program code of the computer programs provided by the embodiments of the present disclosure can be written in any combination of one or more programming languages, and specifically, these computer programs can be implemented using a high-level procedural and / or object-oriented programming language, and / or an assembly / machine language. The programming language includes, but is not limited to, a programming language such as Java, C++, Python, "C" language, or a similar programming language. The program code can be executed entirely on a user computing device, partially on a user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case involving a remote computing device, the remote computing device can be connected to the user computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, connected to the Internet through an Internet service provider).

[0155] The flow diagrams and the block diagrams in the drawings are illustrations of possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flow diagrams or block diagrams can represent a module, a segment, or a portion of code, which comprises one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flow diagrams, and combinations thereof, can be implemented by special purpose hardware-based systems that perform the specified functions or operations, or combinations of special purpose hardware and computer instructions.

[0156] Those skilled in the art will understand that features recited in the various embodiments of the present disclosure can be combined and / or integrated in various combinations and / or permutations, even if such combinations and / or permutations are not expressly noted in the present disclosure. In particular, features recited in the various embodiments of the present disclosure can be combined and / or integrated in various combinations and / or permutations without departing from the spirit and teachings of the present disclosure. All such combinations and / or integrations are within the scope of the present disclosure.

[0157] The embodiments of the present disclosure are described above. However, these embodiments are merely for illustrative purposes, and are not intended to limit the scope of the present disclosure. Although each embodiment is described above separately, this does not mean that the measures in each embodiment cannot be used advantageously in combination. Those skilled in the art can make various substitutions and modifications without departing from the scope of the present disclosure, and these substitutions and modifications should all fall within the scope of the present disclosure.

Claims

1. A workflow configuration method, characterized by, The method comprises: in response to receiving a workflow configuration requirement, determining a target historical workflow matching the workflow configuration requirement from a plurality of historical workflows; retrieving configuration knowledge information of each of a plurality of preset nodes stored in a configuration knowledge base according to the workflow configuration requirement and the target historical workflow to obtain a target configuration knowledge information set; processing the workflow configuration requirement and the target configuration knowledge information set by using a pre-trained language model to obtain a workflow configuration file, so as to configure a workflow based on the workflow configuration file.

2. The method of claim 1, wherein, The retrieving configuration knowledge information of each of a plurality of preset nodes stored in a configuration knowledge base according to the workflow configuration requirement and the target historical workflow to obtain a target configuration knowledge information set comprises: determining a plurality of first target nodes included in the target historical workflow; retrieving configuration knowledge information of each of the plurality of first target nodes from the configuration knowledge base to obtain a first target configuration knowledge information subset matching the target historical workflow; retrieving configuration knowledge information of each of the plurality of preset nodes based on the workflow configuration requirement to obtain a second target configuration knowledge information subset matching the workflow configuration requirement; obtaining the target configuration knowledge information set according to the first target configuration knowledge information subset and the second target configuration knowledge information subset.

3. The method of claim 2, wherein, The configuration knowledge information comprises node function information and node dependency information; the retrieving configuration knowledge information of each of the plurality of preset nodes based on the workflow configuration requirement to obtain a second target configuration knowledge information subset matching the workflow configuration requirement comprises: matching the workflow configuration requirement with node function information of each of the preset nodes to determine at least one second target node whose node function information matches the workflow configuration requirement; determining third target nodes having a dependency relationship with each of the second target nodes from the plurality of preset nodes according to node dependency information of each of the second target nodes; obtaining the second target configuration knowledge information subset according to configuration knowledge information of each of the second target nodes and each of the third target nodes.

4. The method of claim 1, wherein, The determining a target historical workflow matching the workflow configuration requirement from a plurality of historical workflows comprises: performing keyword extraction on the workflow configuration requirement to obtain a scene keyword and a function keyword; determining a target historical workflow matching the workflow configuration requirement from a plurality of historical workflows according to the scene keyword and the function keyword.

5. The method of claim 4, wherein, The determining a target historical workflow matching the workflow configuration requirement from a plurality of historical workflows according to the scene keyword and the function keyword comprises: obtaining scene information and function information of each of the historical workflows; determining a first matching degree between the workflow configuration information and each of the historical workflows according to the scene keyword and the scene information of each of the historical workflows; determine a second matching degree of the workflow configuration information and each of the historical workflows according to the function keyword and the function information of each of the historical workflows; determine a target historical workflow from the plurality of historical workflows according to the first matching degree and the second matching degree of the workflow configuration information and each of the historical workflows.

6. The method of claim 1, wherein, The processing of the workflow configuration requirement and the target configuration knowledge information set by the pre-trained language model includes: inputting the workflow configuration requirement and the target configuration knowledge information set into a first prompt template to obtain first prompt information, the first prompt information being used to guide the pre-trained language model to generate a workflow configuration file according to the workflow configuration requirement and the target configuration knowledge information set; inputting the first prompt information into the pre-trained language model to generate the initial workflow configuration file; performing format verification on the initial workflow configuration file to obtain a verification result; in a case where the verification result represents that the initial workflow configuration file passes the verification, determining that the initial workflow configuration file is the workflow configuration file.

7. The method of claim 6, wherein, The method further includes: in a case where the verification result represents that the initial workflow configuration file fails the verification, inputting the verification result and the initial workflow configuration file into a second prompt template to obtain second prompt information, the second prompt information being used to guide the pre-trained language model to correct the format of the initial workflow configuration file according to the verification result; inputting the second prompt information into the pre-trained language model to obtain the workflow configuration file.

8. A workflow configuration apparatus characterized by comprising: The apparatus includes: a historical matching module configured to, in response to receiving a workflow configuration requirement, determine a target historical workflow that matches the workflow configuration requirement from a plurality of historical workflows; a knowledge retrieval module configured to, according to the workflow configuration requirement and the target historical workflow, retrieve configuration knowledge information of each of a plurality of preset nodes stored in a configuration knowledge base to obtain a target configuration knowledge information set; a configuration generation module configured to process the workflow configuration requirement and the target configuration knowledge information set by a pre-trained language model to obtain a workflow configuration file, so as to configure a workflow based on the workflow configuration file. 9.An electronic device, comprising: one or more processors; a memory for storing one or more computer programs, characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program or instructions, characterized in that, The computer program or instructions are executed by the processor to implement the steps of the method according to any one of claims 1-7.

Images