Workflow creation method, device, equipment and storage medium
Through the combination of language chat robots and preset process knowledge bases, automated generation and intelligent filling of workflows have been solved, and the problems of low efficiency and human error in workflow creation in the existing technology have been achieved, efficient and accurate workflow management has been achieved, and the workflow management has been reduced. Operational cost and user operation difficulty.
Patent Information
- Application Number
- CN202410102089.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-25
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-01-25
AI Technical Summary
The existing workflow system requires manual operation when creating processes and filling out forms, resulting in inefficiency and human errors, difficulty in user operations, and increased operational costs and communication costs.
Using a language chat robot combined with preset process knowledge base and knowledge graph, it obtains user needs through text segmentation, vectorization and structured processing, enters the business process question-and-answer model to generate response information, automatically creates workflows, and implements the process through the workflow creation engine Automatic generation and intelligent filling.
It significantly improves the efficiency and accuracy of workflow creation, reduces human errors, reduces operational costs and user operation burden, and improves user experience.
Smart Images

Figure CN117634865B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of workflow management, and in particular to a workflow creation method, device, equipment and storage medium. Background Art
[0002] For workflow systems, whether using open source engines such as Activiti, Flowable or Camunda, business processes such as creating processes and filling out forms usually require manual operations, and when the process design changes, it is also necessary to rely on system administrators for manual processing. The administrator may also need to draw flowcharts and configure forms, etc., resulting in inefficient process design. Manual processing may also result in incorrect filling out of operations.
[0003] At the same time, the workflow system has a lot of complicated processes, and most users are not clear about how to choose a process. Therefore, developers need to spend a lot of time training and writing operation manuals to guide users to submit bills, resulting in low efficiency and greatly increased operating costs. There are also cases where users are unclear about the meaning of the fields when submitting bills, which makes it impossible to fill them out. Therefore, it takes a lot of time to communicate and fill out the forms, and the cost of answering questions is high. Although there are various unstructured related documents such as FAQ, due to the passage of time and personnel changes, the related documents lack iterative updates, and the user's access effect is not good. When users encounter problems, they need to communicate with corresponding professional operation and maintenance colleagues, which increases the communication cost. Summary of the invention
[0004] Based on this, it is necessary to provide a workflow creation method, device, equipment and storage medium to address the above-mentioned technical problems, aiming to realize the automatic generation and intelligent filling of workflows, thereby significantly improving work efficiency and accuracy, while reducing human errors, effectively reducing operating costs, reducing user operation burden, and improving user experience.
[0005] A workflow creation method, which is applied to a voice chat robot and includes:
[0006] Obtain the workflow creation information corresponding to the target user;
[0007] Determine first response information and a workflow type based on the workflow creation information, a preset process knowledge base, and a preset process knowledge graph;
[0008] Inputting the first response information into a business process question and answer model to obtain second response information output by the business process question and answer model;
[0009] The second response information and the workflow type are sent to a workflow creation engine, so that the workflow creation engine creates a workflow corresponding to the workflow creation information.
[0010] According to a workflow creation method provided by the present invention, determining first response information and a workflow type based on the workflow creation information, a preset process knowledge base, and a preset process knowledge graph includes:
[0011] Performing text segmentation on the workflow creation information, and vectorizing the segmented workflow creation information to obtain a workflow creation vector;
[0012] Query and match the workflow creation vector with a preset process knowledge base to obtain the workflow type and initial response information;
[0013] The initial response information is structured, and the structured initial response information is matched with a preset process knowledge graph to obtain the first response information.
[0014] According to a workflow creation method provided by the present invention, before determining the first response information and the workflow type based on the workflow creation information, the preset process knowledge base and the preset process knowledge graph, the method further includes:
[0015] Acquire business process knowledge;
[0016] Extracting information and integrating knowledge from the business process knowledge to obtain workflow structured knowledge, and performing quality assessment on the workflow structured knowledge;
[0017] If the quality assessment passes, the workflow structured knowledge is stored in the preset process knowledge graph.
[0018] According to a workflow creation method provided by the present invention, before inputting the first response information into a business process question and answer model to obtain the second response information output by the business process question and answer model, the method further includes:
[0019] Obtaining several groups of process requirement samples, and dividing each of the process requirement samples into a model training set, a model verification set, and a model test set;
[0020] Based on the model training set, the model verification set and the model test set, the initial business process question and answer model is iteratively trained and fine-tuned to obtain the business process question and answer model.
[0021] According to a workflow creation method provided by the present invention, the step of obtaining workflow creation information corresponding to a target user includes:
[0022] Obtain workflow creation requirements input by target users;
[0023] Based on the workflow creation requirement, determine whether it matches the historical creation information;
[0024] If so, the matched historical creation information and the workflow creation requirement are associated and combined to generate the workflow creation information;
[0025] If not, the workflow creation information is generated based on the workflow creation requirement.
[0026] A workflow creation method, which is applied to a workflow creation engine and includes:
[0027] Receiving the second response information and the workflow type sent by the voice chat robot;
[0028] Based on the workflow type, determining a workflow creation module corresponding to the workflow;
[0029] Parsing the second response information to obtain workflow entities, entity attributes, and entity associations;
[0030] The workflow entity, the entity attributes and the entity association status are input into the workflow creation module so that the workflow creation module can create the workflow.
[0031] According to a workflow creation method provided by the present invention, after inputting the workflow entity, the entity attribute and the entity association status into the workflow creation module so that the workflow creation module can create the workflow, the method further comprises:
[0032] Obtain the successful creation time and viewing method of the workflow;
[0033] Generate creation success information based on the workflow type, the creation success time and the viewing method;
[0034] The creation success information is pushed to the target user to remind the target user to check.
[0035] A workflow creation device, comprising:
[0036] The voice chat robot is used to obtain workflow creation information input by a user; determine first response information and a workflow type based on the workflow creation information and a preset process knowledge graph; input the first response information into a business process question and answer model to obtain second response information output by the business process question and answer model;
[0037] A workflow creation engine is used to receive a second response message and a workflow type sent by the voice chat robot; based on the workflow type, determine a workflow creation module corresponding to the workflow; parse the second response message to obtain workflow entities, entity attributes, and entity associations; input the workflow entities, entity attributes, and entity associations into the workflow creation module so that the workflow creation module can create the workflow.
[0038] An electronic device comprises a memory, a processor and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the above-mentioned workflow creation method when executing the computer-readable instructions.
[0039] One or more readable storage media storing computer-readable instructions, when the computer-readable instructions are executed by one or more processors, enable the one or more processors to execute the workflow creation method as described above.
[0040] The above-mentioned workflow creation method, device, equipment and storage medium obtain the workflow creation information corresponding to the target user, and determine the first response information and the workflow type based on the workflow creation information, the preset process knowledge base and the preset process knowledge graph, and then input the first response information into the business process question and answer model to obtain the second response information output by the business process question and answer model, and then send the second response information and the workflow type to the workflow creation engine, so that the workflow creation engine can create a workflow corresponding to the workflow creation information, thereby realizing the automatic generation and intelligent filling of workflows, etc., thereby significantly improving work efficiency and accuracy, while reducing human errors, effectively reducing operating costs, reducing user operation burden, and improving user experience. BRIEF DESCRIPTION OF THE DRAWINGS
[0041] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0042] Figure 1 This is one of the flowcharts of the workflow creation method provided by the present invention;
[0043] Figure 2 This is the second flowchart of the workflow creation method provided by the present invention;
[0044] Figure 3 It is a structural schematic diagram of a workflow creation device provided by the present invention;
[0045] Figure 4 It is a partial construction flow chart of the preset process knowledge base provided by the present invention;
[0046] Figure 5 It is a partial functional implementation diagram of the workflow creation system provided by the present invention;
[0047] Figure 6 It is a partial construction flow chart of the preset process knowledge graph provided by the present invention;
[0048] Figure 7 is an example diagram of information structure provided by the present invention;
[0049] Figure 8 This is an example diagram of the function implementation of the work order generation provided by the present invention;
[0050] Fig. 9 This is an example diagram of the implementation of the form filling function provided by the present invention;
[0051] Fig.10 This is an example diagram of the functional implementation of the process design change provided by the present invention;
[0052] Fig.11 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION
[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are 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 creative work are within the scope of protection of the present invention.
[0054] The terms used in one or more embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present invention. The singular forms of "a", "said" and "the" used in one or more embodiments of the present invention are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present invention refers to and includes any or all possible combinations of one or more associated listed items.
[0055] It should be understood that, although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present invention, these information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of the present invention, the first may also be referred to as the second, and similarly, the second may also be referred to as the first. Depending on the context, the word "if" as used herein may be interpreted as "when..." or "when...".
[0056] Figure 1 Schematic diagram of the workflow creation method provided by the present invention. Figure 1 As shown, the workflow creation method is applied to a voice chat robot, including:
[0057] Step S11, obtaining workflow creation information corresponding to the target user;
[0058] It should be noted that the target user refers to the user who needs to create a workflow, and the workflow creation information refers to the creation information required by the target user's specific requirements and expectations in the process of creating the workflow, including business process descriptions (such as workflow steps, stages, involved workflow entities, entity attributes, and entity associations, etc.), user interaction requirements (such as form elements and user authority management, etc.), historical creation information, custom rules (such as approval conditions, process branches, and automated decisions, etc.), process monitoring and reporting, business knowledge Q&A and other structured data and unstructured data, without limitation here.
[0059] Specifically, the workflow creation requirement input by the target user is obtained, and then based on the workflow creation requirement, it is determined whether the historical creation information is matched, and if so, the matched historical creation information and the workflow creation requirement are associated and combined to generate the workflow creation information.
[0060] Step S12, determining first response information and a workflow type based on the workflow creation information, a preset process knowledge base, and a preset process knowledge graph;
[0061] It should be noted that the preset process knowledge base is formed by data such as BPM (Business Process Management) related FAQ documents (Frequently Asked Questions), system inventory processes, user information, and historical creation information. Data access can be achieved through SQL, Excel, Word and other methods to enrich the data source. Figure 4 , Figure 4This is a partial construction flow chart of the preset process knowledge base provided by the present invention, in which the LangChain document loader can be used to perform file loading and other operations. In addition, the preset process knowledge base can be continuously updated as data is accessed to improve the ability to generate more accurate responses.
[0062] It should be further explained that the preset process knowledge graph refers to a pre-set knowledge graph that stores various types of structured data such as basic attribute knowledge, association knowledge, event knowledge, time series knowledge, and resource knowledge, wherein the preset process knowledge graph can be automatically iterated and updated to obtain more precise knowledge and generate more accurate and detailed responses.
[0063] Additionally, the first response information is response information generated by the preset process knowledge graph and the workflow creation information, etc. The workflow types include process design fine-tuning, work order generation, and form filling, etc., which are not limited here.
[0064] Specifically, the workflow creation information is subjected to text segmentation, and the segmented workflow creation information is vectorized to obtain a workflow creation vector, and then the workflow creation vector is queried and matched with a preset process knowledge base to obtain the workflow type and initial response information, thereby structuring the initial response information, and matching the structured initial response information with a preset process knowledge graph to obtain the first response information.
[0065] Step S13, inputting the first response information into the business process question and answer model to obtain second response information output by the business process question and answer model;
[0066] It should be noted that the business process question and answer model is obtained by iterative training of process requirement samples such as model training set, model verification set and model test set. In addition, the second response information refers to the response information generated by the business process question and answer model for the first response information. When the target user makes a business inquiry through the voice chat robot, the second response information can be returned to the target user to achieve user interaction and enhance the system response capability.
[0067] Step S14: Send the second response information and the workflow type to a workflow creation engine, so that the workflow creation engine can create a workflow corresponding to the workflow creation information.
[0068] It should be noted that the workflow creation engine refers to a logic solution engine that transmits routing, content level, etc. according to different decision information in the workflow creation information, such as workflow entities, entity attributes, and entity associations. It can be an Activiti process engine, etc., and is not limited here. For example, a workflow for generating an employee's leave application. When the department head approves the leave, the workflow creation engine generates a workflow according to the business logic to send the leave application to the company's upper-level leaders for confirmation and approval. Among them, the workflow creation engine includes various workflow type modules, such as a work order generation module, a form filling module, and a process design fine-tuning module, etc., which are not limited here.
[0069] It should be further explained that the workflow is a systematic and orderly business process used to organize, manage and execute specific work tasks or business activities. The workflow can cover business processes within various organizations. From simple daily tasks to complex business processes, workflows can be used for planning and management. It can be set according to actual conditions and will not be described in detail here.
[0070] Specifically, the second response information and the workflow type and other information are sent to the workflow creation engine, so that the workflow creation engine can create a workflow corresponding to the workflow creation information after receiving the second response information and the workflow type and other information. Figure 5 , Figure 5 It is a partial functional implementation diagram of the workflow creation system provided by the present invention.
[0071] The embodiment of the present invention obtains the workflow creation information corresponding to the target user, and determines the first response information and the workflow type based on the workflow creation information, the preset process knowledge base and the preset process knowledge graph, and then inputs the first response information into the business process question and answer model to obtain the second response information output by the business process question and answer model, and then sends the second response information and the workflow type to the workflow creation engine, so that the workflow creation engine can create a workflow corresponding to the workflow creation information, thereby realizing the automatic generation and intelligent filling of the workflow, thereby significantly improving work efficiency and accuracy, while reducing human errors, effectively reducing operating costs, reducing user operation burden, and improving user experience.
[0072] In one embodiment of the present invention, the determining the first response information and the workflow type based on the workflow creation information, the preset process knowledge base and the preset process knowledge graph includes:
[0073] The workflow creation information is subjected to text segmentation, and the segmented workflow creation information is vectorized to obtain a workflow creation vector; the workflow creation vector is queried and matched with a preset process knowledge base to obtain the workflow type and initial response information; the initial response information is structured, and the structured initial response information is matched with a preset process knowledge graph to obtain the first response information.
[0074] It should be noted that the workflow creation vector refers to the vector value obtained in the process of converting the workflow creation information into a mathematical vector, so as to allow computer systems, etc. to better understand and process data such as text. In addition, the initial response information refers to the response information generated according to the preset process knowledge base.
[0075] Specifically, the workflow creation information is subjected to text segmentation, such as segmenting the text into words or vocabulary units and removing stop words, wherein the stop words are words that frequently appear in the text but usually do not carry specific information, such as "and", "the", etc. Then, the segmented workflow creation information is vectorized to obtain a workflow creation vector, which can be vectorized using a bag of words model (BoW), TF-IDF (Term Frequency-Inverse Document Frequency), or a word embedding model (such as Word2Vec, GloVe, FastText), etc., without limitation here.
[0076] Furthermore, the workflow creation vector is queried and matched with a preset process knowledge base, wherein the faiss (Facebook AI Similarity Search) search method can be used to determine the workflow type and the initial response information according to the matching result, and then the initial response information is structured, that is, the initial response information is converted into structured information, which can be referred to Figure 7 , Figure 7 It is an information structuring example diagram provided by the present invention, and matches the structured initial response information with the preset process knowledge graph, that is, matches the relevant entity nodes corresponding to the initial response information, thereby filling the structured initial response information into the relevant entity nodes to obtain the first response information.
[0077] The embodiment of the present invention performs text segmentation on the workflow creation information and vectorizes the segmented workflow creation information to obtain a workflow creation vector, and then queries and matches the workflow creation vector with a preset process knowledge base to obtain the workflow type and initial response information, thereby structuring the initial response information and matching the structured initial response information with a preset process knowledge graph to obtain the first response information, and then through the preset process knowledge base and the preset process knowledge graph, efficient and accurate identification of user intentions is achieved, and accurate and practical response information is generated according to the intentions, thereby effectively improving the response ability of the voice chat robot, improving user experience and work efficiency, and thereby reducing operating costs and alleviating the user's operating burden.
[0078] In one embodiment of the present invention, before determining the first response information and the workflow type based on the workflow creation information, the preset process knowledge base and the preset process knowledge graph, the method further includes:
[0079] Acquire business process knowledge; perform information extraction and knowledge fusion on the business process knowledge to obtain workflow structured knowledge, and perform quality assessment on the workflow structured knowledge; if the quality assessment passes, store the workflow structured knowledge in the preset process knowledge graph.
[0080] It should be noted that the traditional vector-based search process is: load file - read text - text segmentation - text vectorization - question vectorization - match the top k most similar to the question vector in the text vector - add the matched text as context and question to the prompt (answer) - submit to LLM (Large Language Model) to generate answers. This method is very effective when vector search can produce relevant text blocks. However, when LLM needs information from multiple documents or even multiple blocks to generate answers, simple vector similarity search may not be accurate enough. Moreover, this method can easily miss interrelated document blocks, resulting in incomplete information retrieval. At the same time, therefore, in order to reduce the inaccuracy caused by embedding-based semantic search, it is necessary to build a preset process knowledge graph based on the BPM business process field.
[0081] It should be further explained that the business process knowledge refers to various types of knowledge related to BPM, including basic attribute knowledge, association knowledge, event knowledge, timing knowledge, resource knowledge, etc. The workflow structured knowledge is obtained by extracting information from the business process knowledge and integrating knowledge.
[0082] Specifically, business process knowledge is acquired, and then information extraction and knowledge fusion are performed on the business process knowledge to obtain workflow structured knowledge, wherein information extraction refers to extracting workflow entities, entity attributes, and entity associations from the business process knowledge, such as the two entities of department head and company upper management in the leave process, and the logical relationship between the two, and on this basis, an ontological knowledge expression is formed. For example, the leave application process is transferred to the company's upper management only after the department head approves the leave; knowledge fusion refers to the integration of the business process knowledge to eliminate contradictions and ambiguities. For example, some entities may have multiple expressions, and a specific title can also correspond to multiple different entities and the association between synonyms.
[0083] Furthermore, the quality of the workflow structured knowledge is evaluated, wherein the quality evaluation can be performed by the system or by manual screening. If the quality evaluation passes, the workflow structured knowledge is stored in the preset process knowledge graph. Figure 6 , Figure 6 It is a partial construction flow chart of the preset process knowledge graph provided by the present invention. In addition, in order to update the knowledge graph and pursue more accurate response generation capabilities, building the preset process knowledge graph is an iterative update process, and each round of iteration includes four stages: information extraction, knowledge fusion, knowledge calculation (quality assessment), and knowledge storage.
[0084] In addition, the preset process knowledge graph can also realize knowledge modeling, that is, a target user-centered knowledge graph is established for all workflow creation information of the target user, so as to generate more efficient and personalized accurate response information and improve the work efficiency of the target user.
[0085] The embodiment of the present invention obtains business process knowledge, extracts information and fuses knowledge of the business process knowledge to obtain workflow structured knowledge, and performs quality assessment on the workflow structured knowledge. If the quality assessment passes, the workflow structured knowledge is stored in the preset process knowledge graph, thereby reducing the inaccuracy caused by embedded semantic search, improving the accuracy of responses, and providing users with better service quality.
[0086] In one embodiment of the present invention, before inputting the first response information into the business process question and answer model to obtain the second response information output by the business process question and answer model, the method further includes:
[0087] Acquire several groups of process requirement samples, and divide each of the process requirement samples into a model training set, a model verification set, and a model test set; based on the model training set, the model verification set, and the model test set, iteratively train and fine-tune the initial business process question and answer model to obtain the business process question and answer model.
[0088] It should be noted that if the large model corpus of language models such as the ChatGLM6B model is directly used, since there is no corresponding training set for the BPM workflow, the response results generated by the ChatGLM6B model cannot achieve the target accuracy. Therefore, retraining is required to obtain the business process question and answer model, among which the P-Tuning v2 method can be used to fine-tune the model to better serve the target users and bring users a better service experience.
[0089] It should be further explained that the process requirement sample refers to the relevant question and answer data generated by the target users during the system operation and maintenance process, such as user questions and answers such as "What is a configuration class?", and each process requirement sample carries a corresponding sample label, such as "What is a configuration class?" The corresponding sample label is "The requirements usually raised do not involve development, such as non-development requirements such as data import and permission application. The currently supported configuration requirements are: business data import, permission configuration, process node adjustment, etc." In addition, the model training set is used in the model training process, and the model validation set is used to evaluate the performance of the validation set after each training iteration, so as to fine-tune the model according to the evaluation results; the model test set is used to evaluate the performance and accuracy of the model.
[0090] Specifically, in one embodiment, several groups of process requirement samples are obtained, and each of the process requirement samples is divided into a model training set, a model verification set and a model test set, and then data preprocessing is performed on the model training set, the model verification set and the model test set, such as word segmentation of the input text, removal of stop words, and conversion into the input format required by the model.
[0091] Furthermore, a pre-trained initial business process question and answer model is loaded, including its vocabulary and weight parameters, etc., wherein the initial business process question and answer model may be a ChatGLM6B model, etc., and additional trainable parameters are added to the input layer of the initial business process question and answer model to represent the Prompt parameter, so as to construct an initial business process question and answer model based on the P-Tuning v2 method.
[0092] Furthermore, the model training set is input into the initial business process question and answer model to obtain the predicted value output by the initial business process question and answer model, and then based on the predicted value and the sample label corresponding to the process requirement sample, the model loss value is calculated using the entropy loss function. In this embodiment, the loss function can be set according to actual needs and is not specifically limited here. After the model loss value is calculated, the training process ends, and the error back propagation algorithm and SGD optimizer are used to update the model parameters in the initial business process question and answer model before the next training is carried out. After each training iteration, the performance of the model verification set is evaluated, and the model is adjusted as needed.
[0093] Furthermore, during the model adjustment process, it is determined whether the updated initial business process question and answer model meets the preset training end conditions based on the model test set. If so, the updated initial business process question and answer model is used as the business process question and answer model. If not, the model training continues, wherein the preset training end conditions include loss convergence and reaching the maximum number of iterations threshold, etc.
[0094] Furthermore, after the model training is completed, the training test set is used to evaluate the performance and accuracy of the business process question answering model.
[0095] The embodiment of the present invention obtains several groups of process requirement samples, and divides each of the process requirement samples into a model training set, a model verification set and a model test set, and then iteratively trains the initial business process question and answer model based on the model training set, the model verification set and the model test set to obtain the business process question and answer model, thereby increasing the robustness of the model and generating more accurate and professional professional field responses, thereby improving quality services for users and improving user experience.
[0096] In one embodiment of the present invention, the step of obtaining workflow creation information corresponding to the target user includes:
[0097] Obtaining a workflow creation requirement input by a target user; based on the workflow creation requirement, determining whether it matches historical creation information; if so, associating and combining the matched historical creation information and the workflow creation requirement to generate the workflow creation information; if not, generating the workflow creation information based on the workflow creation requirement.
[0098] It should be noted that the workflow creation requirement refers to the requirement information for the target user to create a workflow, which can be obtained through multiple rounds of conversations between the voice chat robot and the target user. For example, when the target user wants to generate a work order, the voice chat robot needs to have multiple rounds of conversations with the target user to determine whether to generate a work order based on historical creation records and whether to fill in default values according to the form, etc., so as to obtain the information required to generate the work order from the target user. When the target user only needs to generate a general notification, the voice chat robot can also understand the target user's intention directly through one round of conversation. Therefore, the method of obtaining the workflow creation requirement depends on the needs of the target user. There is no restriction here and it can be set according to actual conditions.
[0099] It should be further explained that in order to further improve the efficiency of workflow creation, the voice chat robot can clarify the real needs of the target user by extracting the target user's historical creation information, where the historical creation information refers to the record information of the target user's previous creation of the workflow, such as form generation history information, work order extraction history information, and process design modification history information, etc. It also includes structured and unstructured data such as form data and process design nodes, which can be set according to the actual needs of the target user and are not restricted here.
[0100] Specifically, the workflow creation requirement input by the target user is obtained, and then based on the workflow creation requirement, whether the historical creation information is matched is determined, such as by performing semantic analysis on the workflow creation requirement, and by directly inquiring the target user (such as: whether to make a bill of lading based on the historical bill of lading record of January 8, 2024?), etc., to determine whether the historical creation information is matched. Further, if so, the matched historical creation information and the workflow creation requirement and other information are associated and combined to generate the workflow creation information.
[0101] In addition, if there is no need to match historical creation information, the workflow creation information can be generated directly according to the workflow creation requirements. For example, the data information required to extract the form can be directly obtained through multiple rounds of conversations between the voice chat robot and the target user, and then the data information is analyzed and integrated to generate the workflow creation information. It can be set according to actual conditions and is not limited here.
[0102] The embodiment of the present invention obtains the workflow creation requirements input by the target user, and then determines whether the historical creation information is matched based on the workflow creation requirements. If so, the matched historical creation information and the workflow creation requirements are associated and combined to generate the workflow creation information. Additionally, if not, the workflow creation information is generated based on the workflow creation requirements, and then a conversation is conducted with the target user through a voice chat robot to clarify the real needs of the target user, thereby effectively improving the workflow creation efficiency, providing users with efficient and portable services, and then effectively reducing operating costs and alleviating the user's operating burden.
[0103] Figure 2 Schematic diagram of the workflow creation method provided by the present invention. Figure 2 As shown, the workflow creation method is applied to a workflow creation engine, including:
[0104] Step S21, receiving the second response information and workflow type sent by the voice chat robot;
[0105] It should be noted that the voice chat robot is an AI voice chat robot, which can generate corresponding response information for the target user's questions and return it to the target user to achieve user interaction and clarify user intentions.
[0106] Step S22, determining a workflow creation module corresponding to the workflow based on the workflow type;
[0107] It should be noted that the workflow creation module refers to the creation module corresponding to the workflow type, including the work order generation module, the form filling module, and the process design fine-tuning module, etc., and there is no limitation here.
[0108] Specifically, the workflow type is compared and matched with a preset workflow creation module library to determine the workflow creation module corresponding to the workflow.
[0109] Step S23, parsing the second response information to obtain workflow entities, entity attributes and entity associations;
[0110] It should be noted that the workflow entity includes the name, executor, and filled data of the workflow, the entity attributes include the identity of the executor, and the entity association includes the execution order. Figure 7 .
[0111] Specifically, it can be parsed according to the corresponding relevant entity nodes (ie, knowledge graph) in the second response information, and can also be parsed by semantic segmentation and other methods to obtain information such as workflow entities, entity attributes, and entity associations, without limitation here.
[0112] Step S24: input the workflow entity, the entity attributes and the entity association status into the workflow creation module, so that the workflow creation module can create the workflow.
[0113] It should be noted that the creation method of the workflow creation module depends on the workflow type and is not consistent. Each module has a corresponding creation method, which can be referred to Figure 8 , Fig. 9 as well as Fig.10 There is no restriction on the creation method, which can be set according to the actual situation. At the same time, setting up a separate creation module is conducive to improving work efficiency, and multiple workflows can be created at the same time.
[0114] The embodiment of the present invention receives the second response information and the workflow type sent by the voice chat robot, and based on the workflow type, determines the workflow creation module corresponding to the workflow, and then parses the second response information to obtain the workflow entity, entity attributes and entity association status, and then inputs the workflow entity, the entity attributes and the entity association status into the workflow creation module for the workflow creation module to create the workflow, and then realizes the automatic generation and intelligent filling of the workflow through the workflow creation engine, thereby significantly improving work efficiency and accuracy, while reducing human errors, thereby effectively reducing operating costs, alleviating user operation burdens, and improving user experience.
[0115] In one embodiment of the present invention, after inputting the workflow entity, the entity attribute and the entity association status into the workflow creation module so that the workflow creation module can create the workflow, the method further includes:
[0116] The creation success time and viewing method of the workflow are obtained; based on the workflow type, the creation success time and the viewing method, creation success information is generated; and the creation success information is pushed to the target user to remind the target user to view it.
[0117] It should be noted that the creation success information refers to the reminder information generated when the workflow is successfully created, including the workflow type, creation success time, workflow screenshots and viewing methods, etc., which are not limited here.
[0118] Specifically, the successful creation time and viewing method of the workflow are obtained, and then the workflow type, the successful creation time, the viewing method and other information are associated and combined to generate successful creation information, which is then pushed to the target user to remind the target user to view it. The push can be performed through SMS, APP notification, mini-program reminder, and email, etc., which is not limited here and can be set according to actual conditions.
[0119] The embodiment of the present invention obtains the successful creation time and viewing method of the workflow, and generates successful creation information based on the workflow type, the successful creation time and the viewing method, and then pushes the successful creation information to the target user to remind the target user to view it, thereby improving work efficiency, enabling the target user to view the successfully created workflow in a timely and efficient manner, and improving user experience.
[0120] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.
[0121] In one embodiment, a workflow creation device is provided, which corresponds one-to-one to the workflow creation method in the above embodiment. Figure 3 As shown, the workflow creation device includes a voice chat robot 31 and a workflow creation engine 32. Each functional module is described in detail as follows:
[0122] The voice chat robot 31 is used to obtain workflow creation information input by the user; determine first response information and workflow type based on the workflow creation information and the preset process knowledge graph; input the first response information into the business process question and answer model to obtain second response information output by the business process question and answer model;
[0123] The workflow creation engine 32 is used to receive the second response information and the workflow type sent by the voice chat robot; based on the workflow type, determine the workflow creation module corresponding to the workflow; parse the second response information to obtain the workflow entity, entity attributes and entity association status; input the workflow entity, the entity attributes and the entity association status into the workflow creation module so that the workflow creation module can create the workflow.
[0124] The workflow creation device is also used to:
[0125] Performing text segmentation on the workflow creation information, and vectorizing the segmented workflow creation information to obtain a workflow creation vector;
[0126] Query and match the workflow creation vector with a preset process knowledge base to obtain the workflow type and initial response information;
[0127] The initial response information is structured, and the structured initial response information is matched with a preset process knowledge graph to obtain the first response information.
[0128] The workflow creation device is also used to:
[0129] Acquire business process knowledge;
[0130] Extracting information and integrating knowledge from the business process knowledge to obtain workflow structured knowledge, and performing quality assessment on the workflow structured knowledge;
[0131] If the quality assessment passes, the workflow structured knowledge is stored in the preset process knowledge graph.
[0132] The workflow creation device is also used to:
[0133] Obtaining several groups of process requirement samples, and dividing each of the process requirement samples into a model training set, a model verification set, and a model test set;
[0134] Based on the model training set, the model verification set and the model test set, the initial business process question and answer model is iteratively trained and fine-tuned to obtain the business process question and answer model.
[0135] The workflow creation device is also used to:
[0136] Obtain workflow creation requirements input by target users;
[0137] Based on the workflow creation requirement, determine whether it matches the historical creation information;
[0138] If so, the matched historical creation information and the workflow creation requirement are associated and combined to generate the workflow creation information;
[0139] If not, the workflow creation information is generated based on the workflow creation requirement.
[0140] The workflow creation device is also used to:
[0141] Obtain the successful creation time and viewing method of the workflow;
[0142] Generate creation success information based on the workflow type, the creation success time and the viewing method;
[0143] The creation success information is pushed to the target user to remind the target user to check.
[0144] For the specific definition of the workflow creation device, please refer to the definition of the workflow creation method above, which will not be repeated here. Each module in the above-mentioned workflow creation device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the electronic device in the form of hardware, or can be stored in the memory of the electronic device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0145] In one embodiment, an electronic device is provided, which may be a server. The electronic device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a readable storage medium and an internal memory. The readable storage medium stores an operating system, computer-readable instructions and a database. The internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The database of the electronic device is used to store data involved in a workflow creation method. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer-readable instructions are executed by the processor, a workflow creation method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.
[0146] In one embodiment, an electronic device is provided. The electronic device may be a terminal device, and its internal structure diagram may be as follows: Fig.11 As shown. The electronic device includes a processor, a memory, and a network interface connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a readable storage medium. The readable storage medium stores computer-readable instructions. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer-readable instructions are executed by the processor, a workflow creation method is implemented. The readable storage medium provided in this embodiment includes a non-volatile readable storage medium and a volatile readable storage medium.
[0147] In one embodiment, an electronic device is provided, including a memory, a processor, and computer-readable instructions stored in the memory and executable on the processor, wherein the processor implements the steps of the workflow creation method described above when executing the computer-readable instructions.
[0148] In one embodiment, a readable storage medium is provided, the readable storage medium stores computer-readable instructions, and the computer-readable instructions are executed by the processor to implement the steps of the workflow creation method as described above. A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through computer-readable instructions, and the computer-readable instructions can be stored in a non-volatile readable storage medium or a volatile readable storage medium. When the computer-readable instructions are executed, they may include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.
[0149] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0150] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.
Claims
1. A workflow creation method, characterized in that: The workflow creation method is applied to a voice chat robot, and includes: Obtain the workflow creation information corresponding to the target user; Determine first response information and a workflow type based on the workflow creation information, a preset process knowledge base, and a preset process knowledge graph; Inputting the first response information into a business process question and answer model to obtain second response information output by the business process question and answer model; Sending the second response information and the workflow type to a workflow creation engine, so that the workflow creation engine can create a workflow corresponding to the workflow creation information; Performing text segmentation on the workflow creation information, and vectorizing the segmented workflow creation information to obtain a workflow creation vector; Query and match the workflow creation vector with a preset process knowledge base to obtain the workflow type and initial response information; Structuring the initial response information, and matching the structured initial response information with a preset process knowledge graph to obtain relevant entity nodes; Fill the initial response information into the relevant entity node to obtain the first response information.
2. The workflow creation method according to claim 1, characterized in that: Before determining the first response information and the workflow type based on the workflow creation information, the preset process knowledge base and the preset process knowledge graph, the method further includes: Acquire business process knowledge; Extracting information and integrating knowledge from the business process knowledge to obtain workflow structured knowledge, and performing quality assessment on the workflow structured knowledge; If the quality assessment passes, the workflow structured knowledge is stored in the preset process knowledge graph.
3. The workflow creation method according to claim 1, characterized in that: Before inputting the first response information into the business process question and answer model to obtain the second response information output by the business process question and answer model, the method further includes: Obtaining several groups of process requirement samples, and dividing each of the process requirement samples into a model training set, a model verification set, and a model test set; Based on the model training set, the model verification set and the model test set, the initial business process question and answer model is iteratively trained and fine-tuned to obtain the business process question and answer model.
4. The workflow creation method according to claim 1, characterized in that: The step of obtaining workflow creation information corresponding to the target user includes: Obtain workflow creation requirements input by target users; Based on the workflow creation requirement, determine whether it matches the historical creation information; If so, the matched historical creation information and the workflow creation requirement are associated and combined to generate the workflow creation information; If not, the workflow creation information is generated based on the workflow creation requirement.
5. The workflow creation method according to claim 1, characterized in that: The workflow creation method is applied to a workflow creation engine, comprising: Receiving the second response information and the workflow type sent by the voice chat robot; Based on the workflow type, determining a workflow creation module corresponding to the workflow; Parsing the corresponding related entity nodes in the second response information to obtain workflow entities, entity attributes and entity associations; The workflow entity, the entity attributes and the entity association status are input into the workflow creation module so that the workflow creation module can create the workflow.
6. The workflow creation method according to claim 5, characterized in that: After the workflow entity, the entity attribute and the entity association status are input into the workflow creation module so that the workflow creation module can create the workflow, the method further includes: Obtain the successful creation time and viewing method of the workflow; Generate creation success information based on the workflow type, the creation success time and the viewing method; The creation success information is pushed to the target user to remind the target user to check.
7. A workflow creation device, characterized in that: include: The voice chat robot is used to obtain the workflow creation information corresponding to the target user; Determine first response information and a workflow type based on the workflow creation information, a preset process knowledge base, and a preset process knowledge graph; Inputting the first response information into a business process question and answer model to obtain second response information output by the business process question and answer model; Sending the second response information and the workflow type to a workflow creation engine, so that the workflow creation engine can create a workflow corresponding to the workflow creation information; Performing text segmentation on the workflow creation information, and vectorizing the segmented workflow creation information to obtain a workflow creation vector; performing query matching on the workflow creation vector and a preset process knowledge base to obtain the workflow type and initial response information; Structuring the initial response information, and matching the structured initial response information with a preset process knowledge graph to obtain relevant entity nodes; filling the initial response information into the relevant entity nodes to obtain the first response information; The workflow creation engine is used to receive the second response information and the workflow type sent by the voice chat robot; Based on the workflow type, determining a workflow creation module corresponding to the workflow; The corresponding related entity nodes in the second response information are parsed to obtain workflow entities, entity attributes and entity associations; the workflow entities, entity attributes and entity associations are input into the workflow creation module so that the workflow creation module can create the workflow.
8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the program, the workflow creation method according to any one of claims 1 to 6 is implemented.
9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the workflow creation method according to any one of claims 1 to 6 is implemented.
Citation Information
Patent Citations
Method used for intelligently linking work orders with knowledge of knowledge base and suitable for IT operation and maintenance system
CN103942220A
Customer service robot question and answer data creation method and device and computer equipment
CN111753075A
Construction method of intelligent fishery question-answering robot based on knowledge graph
CN111966787A
Workflow creation method and device
CN112036736A
Implementation method for completing intelligent work order filling function based on AI text robot
CN112507092A
