Workflow recovery method and device

Abstract

The present disclosure relates to the technical field of computers, and disclosed are a workflow recovery method and apparatus. A specific implementation of the workflow recovery method comprises: in response to acquiring abortion information of a workflow, generating a corresponding prompt on the basis of the abortion information; inputting the prompt into a preset large model, and enabling the large model to generate guidance information of the workflow on the basis of the prompt; and displaying the guidance information on a preset front-end page, and enabling a user processing the workflow to return to, on the basis of the guidance information, a process node comprised in the workflow, so as to continue to process the workflow. According to the implementation, abortion information of a workflow is analyzed on the basis of a large model to obtain corresponding guidance information, and a user returns to a process node of the workflow on the basis of the guidance information, so as to continue to process the workflow, thereby improving the processing efficiency of the workflow, reducing the difficulty for the user to process the workflow, and enhancing user experience.

Description

Method and apparatus for workflow recovery

[0001] Cross-reference to related applications

[0002] This application claims priority to Chinese Patent Application No. 202410741119.8, filed on June 7, 2024, the entire contents of which are incorporated herein by reference. TECHNICAL FIELD

[0003] The present disclosure relates to the technical field of computer, and particularly relates to a method and apparatus for workflow recovery. BACKGROUND

[0004] In a business system, a workflow business scenario is very common, and a user often needs to process a workflow. For example, before purchasing an insurance service, the user needs to process a workflow including: entering basic information, consulting related questions, applying for an insurance company quotation and underwriting. The workflow is usually composed of multiple pages, and the user needs to submit information on the page, complete necessary interactive operations such as clicking, dragging, checking, and the like, and gradually process each page of the corresponding workflow until the entire workflow is completed. In the process of processing the workflow, if the user has doubts or the page is abnormal, the user can perform self-diagnosis through page prompts or consult a traditional robot customer service to obtain processing suggestions.

[0005] The inventors have found that the prior art at least has the following problems:

[0006] The language of the page prompts and the traditional robot customer service is often professional and has a certain use threshold. When the user has doubts or the page is abnormal, the user cannot be accurately helped, resulting in low efficiency of processing the workflow. SUMMARY

[0007] Therefore, the embodiments of the present disclosure provide a method and apparatus for workflow recovery, which can improve the processing efficiency of the workflow, reduce the difficulty of processing the workflow for the user, and improve the user experience.

[0008] To achieve the above object, according to a first aspect of the embodiments of the present disclosure, a method for workflow recovery is provided, comprising: in response to obtaining suspension information of a workflow, generating corresponding prompt words according to the suspension information; inputting the prompt words into a pre-set large model, so that the large model generates guide information of the workflow according to the prompt words; and displaying the guide information to a pre-set front-end page, so that a user processing the workflow returns to a process node included in the workflow according to the guide information, and continues to process the workflow.

[0009] Optionally, according to the suspension information, a corresponding prompt word is generated, including: determining a suspension scene of the workflow according to the suspension information; determining a guide scene corresponding to the suspension scene according to the pre-set business arrangement information; updating a prompt word template corresponding to the guide scene according to the suspension information, and taking the updated prompt word template as the prompt word.

[0010] Optionally, according to the pre-set business arrangement information, the guide scene corresponding to the suspension scene is determined, including: determining the business logic corresponding to the suspension scene according to the business arrangement information; executing the business logic to obtain an execution result of the business logic; and determining the guide scene corresponding to the suspension scene according to the execution result.

[0011] Optionally, before the guide information is displayed to the pre-set front-end page, intent information associated with the guide information is determined, and it is determined that the intent information does not meet the pre-set guide condition; in the case that the intent information meets the pre-set guide condition, the following steps are repeatedly executed to determine the guide information displayed to the front-end page:

[0012] Taking the intent information as the intent information of the first round, the user purpose information corresponding to the intent information of the current round is determined, and the new prompt word of the current round is generated according to the user purpose information; the large model of the current round is determined according to the large model, the new prompt word of the current round is input into the large model of the current round, and the new intent information and the new guide information output by the large model of the current round are obtained; in the case that the new intent information of the current round does not meet the guide condition, the cycle is terminated, and the new guide information of the current round is taken as the guide information displayed to the front-end page; in the case that the new intent information of the current round meets the guide condition, the new intent information of the current round is taken as the intent information of the next round.

[0013] Optionally, according to the large model, the large model of the current round is determined, including: in response to determining that the new prompt word of the current round matches the large model, taking the large model as the large model of the current round; in response to determining that the new prompt word does not match the large model, screening a target large model matching the new prompt word of the current round from the pre-set plurality of large models; and taking the target large model as the large model of the current round.

[0014] Optionally, before the new guide information and the new intent information output by the large model of the current round are obtained, historical guide information associated with the user is obtained; the historical guide information is input into the large model of the current round, so that the large model of the current round outputs the new guide information and the new intent information of the current round according to the new prompt word of the current round and the historical guide information.

[0015] Optionally, the obtaining the suspension information of the workflow comprises: in response to receiving the dialogue request, determining dialogue information of the user according to the dialogue request, and taking the dialogue information as the suspension information of the workflow; or in response to detecting that the target event occurs, determining whether the target event belongs to a pre-set suspension event set, and in a case where the target event belongs to the suspension event set, determining event information of the target event, and taking the event information as the suspension information of the workflow.

[0016] According to a second aspect of the embodiments of the present disclosure, a device for workflow recovery is provided, comprising:

[0017] a first generation module configured to, in response to obtaining the suspension information of the workflow, generate a corresponding prompt word according to the suspension information;

[0018] a second generation module configured to input the prompt word into a pre-set large model, so that the large model generates guidance information of the workflow according to the prompt word;

[0019] a guidance module configured to display the guidance information to a pre-set front-end page, so that a user processing the workflow returns to a process node included in the workflow according to the guidance information, and continues to process the workflow.

[0020] Optionally, the generating the corresponding prompt word according to the suspension information comprises: determining a suspension scenario of the workflow according to the suspension information; determining a guidance scenario corresponding to the suspension scenario according to pre-set business orchestration information; and updating a prompt word template corresponding to the guidance scenario according to the suspension information, and taking the updated prompt word template as the prompt word.

[0021] Optionally, the determining the guidance scenario corresponding to the suspension scenario according to the pre-set business orchestration information comprises: determining a business logic corresponding to the suspension scenario according to the business orchestration information; executing the business logic to obtain an execution result of the business logic; and determining the guidance scenario corresponding to the suspension scenario according to the execution result.

[0022] Optionally, the guidance module is further configured to: before displaying the guidance information to the pre-set front-end page, determine intent information associated with the guidance information, and determine that the intent information does not meet a pre-set guidance condition.

[0023] In a case where the intent information meets the pre-set guidance condition, the following steps are cyclically executed to determine the guidance information displayed to the front-end page:

[0024] The intention information is taken as the intention information of the first round, user purpose information corresponding to the intention information of the current round is determined, new prompt words of the current round are generated according to the user purpose information, a large model of the current round is determined according to the large model, the new prompt words of the current round are input into the large model of the current round, and new intention information and new guide information output by the large model of the current round are acquired; in a case where the new intention information of the current round does not meet the guide condition, the loop is terminated, and the new guide information of the current round is taken as the guide information displayed to the front-end page; in a case where the new intention information of the current round meets the guide condition, the new intention information of the current round is taken as the intention information of the next round.

[0025] Optionally, the guide module is further configured to: in response to determining that the new prompt words of the current round match the large model, taking the large model as the large model of the current round; in response to determining that the new prompt words do not match the large model, screening a target large model matching the new prompt words of the current round from a plurality of large models set in advance; and taking the target large model as the large model of the current round.

[0026] Optionally, the guide module is further configured to: acquire historical guide information associated with the user; and input the historical guide information into the large model of the current round, so that the large model of the current round outputs the new guide information and the new intention information of the current round according to the new prompt words of the current round and the historical guide information.

[0027] Optionally, the acquired suspension information of the workflow includes: in response to receiving a dialogue request, determining dialogue information of the user according to the dialogue request, and taking the dialogue information as the suspension information of the workflow; or in response to detecting occurrence of a target event, determining whether the target event belongs to a set of pre-set suspension events, and in a case where the target event belongs to the set of suspension events, determining event information of the target event, and taking the event information as the suspension information of the workflow.

[0028] According to a third aspect of the embodiments of the present disclosure, an electronic device is provided, including:

[0029] one or more processors;

[0030] a storage device configured to store one or more programs,

[0031] When the one or more programs are executed by the one or more processors, the one or more processors implement the method of any of the above embodiments.

[0032] According to a fourth aspect of the embodiments of the present disclosure, a computer readable medium is provided, which stores a computer program, and the program is executed by a processor to implement the method of any of the above embodiments.

[0033] According to a fifth aspect of the embodiments of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the method of any of the above-mentioned embodiments.

[0034] One of the embodiments in the above disclosure has the advantages or beneficial effects that: based on the large model, the suspension information of the workflow is parsed to obtain the corresponding guide information, the user returns to the process node of the workflow according to the guide information, and the processing efficiency of the workflow is improved, the difficulty of the user processing the workflow is reduced, and the user experience is improved; according to the business arrangement information, the guide scene corresponding to the suspension scene is determined, and the prompt word is generated according to the guide scene, which can flexibly and accurately generate the prompt word; the business logic corresponding to the suspension scene is executed, and the guide scene is determined according to the execution result of the business logic, which can flexibly and efficiently determine the guide scene; according to the intention information associated with the guide information, it is judged whether new guide information and intention information need to be generated again, and through multiple large model outputs, the guide information finally returned to the user is obtained, which can accurately determine the intention and purpose of the user, facilitate to provide accurate guide information for the user, and help the user quickly return to the process node of the workflow; in the case that the prompt word does not match the large model, the target large model is selected from multiple large models, which can improve the flexibility of the large model output and meet different analysis requirements of the large model; based on the historical guide information output to the user before, new guide information is generated, which can further improve the accuracy of the guide information and improve the processing efficiency of the workflow; the triggering time of the method includes that the user asks or requests a dialogue to the execution subject of the embodiments of the present disclosure, and it is detected that the workflow is abnormally exited or abnormally interrupted due to some operations of the user, which can reduce the difficulty of use of the user and improve the efficiency of the user processing the workflow.

[0035] The further effects of the above-mentioned non-conventional optional mode will be described in the following combined with the specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0036] The accompanying drawings serve to better understand the present disclosure and do not constitute an improper limitation thereof. Among them:

[0037] FIG. 1 is a schematic diagram of the main flow of the method for workflow recovery according to an embodiment of the present disclosure;

[0038] FIG. 2 is a schematic diagram of an application scenario of workflow recovery according to one referable embodiment of the present disclosure;

[0039] FIG. 3 is a schematic diagram of a large model dialogue chain according to one referable embodiment of the present disclosure;

[0040] FIG. 4 is a schematic diagram of the technical architecture of workflow recovery according to one referable embodiment of the present disclosure;

[0041] FIG. 5 is a schematic diagram of the main flow of workflow recovery in multi-module cooperation according to one embodiment of the present disclosure;

[0042] FIG. 6 is a schematic diagram of the main flow of a method of workflow recovery according to one embodiment of the present disclosure;

[0043] FIG. 7 is a schematic diagram of the main modules of an apparatus for workflow recovery according to an embodiment of the present disclosure;

[0044] FIG. 8 is an exemplary system architecture diagram to which embodiments of the present disclosure can be applied;

[0045] FIG. 9 is a structural schematic diagram of a computer system of a terminal device or a server suitable for use to implement embodiments of the present disclosure. DETAILED DESCRIPTION

[0046] Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, which include various details of the embodiments of the present disclosure to assist in understanding, and should be considered merely exemplary. Thus, those of ordinary skill in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, descriptions of known functions and constructions are omitted in the following description for clarity and conciseness.

[0047] It should be noted that in the technical solutions of the present disclosure, the collection, use, storage, sharing and transfer of user personal information involved in the technical solutions comply with relevant laws and regulations, and the user is informed and the user's consent or authorization is obtained, and when applicable, the user's personal information is subjected to technical processing of de-identification and / or anonymization and / or encryption.

[0048] In a business system, workflow business scenarios are very common, and users often need to process workflows. For example, before purchasing an insurance service, the workflows that the user needs to process include: entering basic information, consulting related questions, applying for insurance company quotes and underwriting. A workflow is usually composed of multiple pages, and the user needs to submit information on the page and complete necessary interactive operations, such as clicking, dragging, checking, etc., according to page prompts, to gradually process the workflow corresponding to each page until the entire workflow is completed. In the process of processing the workflow, if the user has doubts or the page is abnormal, self-diagnosis can be performed through page prompts, or traditional robot customer service can be consulted to obtain processing suggestions.

[0049] The language of the page prompts and the traditional robot customer service is often professional and has a certain usage threshold, and when the user has doubts or the page is abnormal, it cannot provide accurate help to the user, resulting in low workflow processing efficiency and poor user experience.

[0050] Therefore, according to a first aspect of the embodiments of the present disclosure, a method for workflow recovery is provided.

[0051] FIG. 1 is a schematic diagram of the main flow of the method for workflow recovery according to the embodiments of the present disclosure. As shown in FIG. 1, the method for workflow recovery according to the embodiments of the present disclosure mainly includes the following steps S101-S103.

[0052] In step S101, in response to obtaining the suspension information of the workflow, the corresponding prompt word is generated according to the suspension information.

[0053] The execution subject of the embodiments of the present disclosure monitors the process of the user processing the workflow, obtains the user behavior data, form collection data, page processing results and other information related to the workflow generated by the user in the process of processing the workflow, and extracts the suspension information from the above information.

[0054] Exemplarily, the workflow processed by the user includes multiple pages, and the execution subject of the embodiments of the present disclosure obtains the page processing result of each page, which includes whether the processing is successful, the information input by the user on the page form, the basic information displayed on the page, and the like. In the case of a "successful" page processing result, the next page of the workflow is displayed to the user, and in the case of an "unsuccessful" page processing result, it is determined that the user's workflow processing flow is suspended, the user behavior data such as the text input by the user on the page and the uploaded file, the number of times the user clicks the button, the number of times the user refreshes the page, and the basic information displayed on the page are obtained, and the above obtained information is taken as the suspension information.

[0055] After obtaining the suspension information, the suspension information is filled into the pre-set prompt word template, and the prompt word template after filling in the suspension information is taken as the prompt word. Exemplarily, the pre-set prompt word template is "You are a customer service personnel, the user encounters a problem when processing the workflow, which leads to the suspension of the workflow, the relevant information includes: the text input by the user on the page: [A1], the user behavior data: [A2], the basic information displayed on the page: [A3], the workflow processing result: [A4], please provide a solution for the user according to the above information, and guide the user to complete the processing of the workflow", and the corresponding suspension information is replaced with A1, A2, A3 and A4 in the above prompt word template. After replacement, the prompt word template becomes "You are a customer service personnel, the user encounters a problem when processing the workflow, which leads to the suspension of the workflow, the relevant information includes: the text input by the user on the page: user, 20,..., the user behavior data: contact the "confirm" button 5 times, the basic information displayed on the page: name, age,..., the workflow processing result: unsuccessful, part of the information processing is abnormal, please provide a solution for the user according to the above information, and guide the user to complete the processing of the workflow".

[0056] It should be noted that the prompt word template is configurable, and the execution subject of the embodiment of the present disclosure adds a new prompt word template, modifies or deletes an existing prompt word template according to the received prompt word update request.

[0057] The processing process of the workflow is monitored, the suspension information of the workflow is obtained in time, the suspension information is converted into a prompt word according to the prompt word template, the data basis for the large model is provided, the efficiency of the workflow processing is improved, and the user returns to the workflow processing process in time.

[0058] In step S102, the prompt word is input into the pre-set large model, so that the large model generates guidance information of the workflow according to the prompt word.

[0059] After generating the prompt word, the prompt word is input into the pre-set large model. The large model is a generative large model and has business knowledge of the business field to which the workflow belongs. Specifically, before the large model outputs the guidance information, the large model is pre-trained using the business knowledge, so that the large model is familiar with various processes in the business field and has existing business knowledge in the business field. Then the prompt word is input into the large model after pre-training.

[0060] The large model outputs corresponding guidance information according to the received prompt word. The guidance information is used to help the user continue to process the previously suspended workflow. The guidance information includes an explanation of the current workflow processing situation, a reason for the suspension of the workflow, and an operation (i.e., a method of returning to a previously suspended workflow process node) that the user can perform.

[0061] For example, the guidance information output by the large model is: the process node where the workflow suspension just occurred is node B1, the meaning of this process node is data B2, you need to provide information B3, if you have no other questions, you can click page link B4 to continue the process node that was suspended just now. It should be noted that the values of node B1, data B2, information B3 and page link B4 are different in different guidance information, which are generated according to the received suspension information.

[0062] The guidance information can answer the questions raised by the user during the processing of the workflow, so that the user can better understand the processing process of the workflow, and through the page link, the user can return to the previously suspended workflow in time, which helps the user to improve the processing efficiency and accuracy of the workflow and improve the user experience.

[0063] In step S103, the guidance information is displayed to the pre-set front-end page, so that the user processing the workflow returns to the process node included in the workflow according to the guidance information and continues to process the workflow.

[0064] After the generation of the guidance information, the guidance information is displayed to the front-end page. Specifically, the front-end page is a pre-set dialog box, and the execution subject of the embodiment of the present disclosure carries on a conversation with the user in the identity of a page intelligent robot, and the conversation content is the previously generated guidance information.

[0065] According to the guidance information, the user understands the meaning of the workflow, knows the information to be filled in and the operation to be performed, and then clicks the link in the guidance information to send a process node access request to the execution subject of the embodiment of the present disclosure. The execution subject of the embodiment of the present disclosure parses the received process node access request, determines the target process node to be accessed by the user, and displays the page corresponding to the target process node to the user, which is equivalent to the user returning to the previously suspended workflow. The user continues to process the unfinished workflow in the page corresponding to the target process node.

[0066] It should be noted that after receiving the guidance information, the user can continue to dialogue with the page intelligent robot (i.e. the execution subject of the embodiment of the present disclosure) in the dialog box, for example, inputting text, image or video and the like information in the dialog box. The execution subject of the embodiment of the present disclosure will generate new guidance information according to the dialogue content of the user and display it in the dialog box. Specifically, the dialogue content of the user is converted into new prompt words and input into the large model to obtain new guidance information. That is, the above steps S101 to S103 are repeatedly in the form of dialogue between the page intelligent robot and the user. The user sends dialogue content to the page intelligent robot, and the page intelligent robot takes the dialogue content of the user as suspension information, generates corresponding guidance information based on the large model, and displays the guidance information as reply content in the dialog box. The above steps are repeated until the user returns to the previously suspended workflow according to the link in the guidance information.

[0067] FIG. 2 is a schematic diagram of an application scenario of workflow recovery according to one embodiment of the present disclosure. As shown in FIG. 2, the conversation content between a user and an execution subject (i.e., AI) of an embodiment of the present disclosure is displayed in a conversation page 201. When the execution subject of the embodiment of the present disclosure detects that the user has suspended a workflow, it collects the suspension information, determines that the workflow is suspended at the "product usage description stage", generates a first guiding information (i.e., the first sentence sent by AI) based on a large model, and displays the guiding information in the conversation page 201 to have a conversation with the user. Based on the first guiding information, the user explains his / her question (i.e., the first sentence sent by the user). According to the question sent by the user, the execution subject of the embodiment of the present disclosure generates a second guiding information (i.e., the second sentence sent by AI) based on the large model again. The second guiding information includes a page link that can help the user return to the previously suspended workflow. If the user wants to continue the communication, he / she can input the content in the conversation box at the bottom of the conversation page 201 and send it to the execution subject of the embodiment of the present disclosure. The execution subject of the embodiment of the present disclosure will continue to output guiding information according to the content sent by the user to help the user answer questions.

[0068] Based on the large model, the suspension information of the workflow is analyzed to obtain the corresponding guiding information, so that the user can return to the process node of the workflow according to the guiding information and continue to process the workflow, which can improve the processing efficiency of the workflow, reduce the difficulty of the user in processing the workflow, and improve the user experience.

[0069] According to one embodiment of the present disclosure, when generating the corresponding prompt word according to the suspension information, the suspension scenario of the workflow is first determined according to the suspension information. Specifically, the suspension scenario includes the accessed page, whether the page is closed, whether there is user input in the page, whether the workflow corresponding to the page is successfully processed, and other page states. The suspension scenario is a collection of the above-mentioned page states, and the page states are combined into the suspension scenario according to the suspension information. For example, the suspension information includes that the user has accessed the page C, the page C is closed, and the workflow corresponding to the page C is not successfully processed. The above-mentioned suspension information is converted into the suspension scenario to obtain "the page C is closed and there is no user input".

[0070] After determining the suspension scenario, the guiding scenario corresponding to the suspension scenario is determined according to the pre-set business arrangement information. The business arrangement information includes the association relationship between the suspension scenario and the guiding scenario. Each suspension scenario corresponds to a guiding scenario. For example, in the case that the suspension scenario is "the page C1 is closed and there is no user input", the corresponding guiding scenario is "asking the user whether to access the C2 page" which is determined by querying the business arrangement information.

[0071] For each change (e.g., a user completes a page flow, a user raises a question, a user performs an abnormal operation, and the like) on the front-end page that causes the workflow to be aborted, a corresponding abort scenario and a corresponding guide scenario corresponding to the abort scenario are set in advance in the business orchestration information. It should be noted that the business orchestration information is configurable, and the execution subject of the embodiment of the present disclosure updates the business orchestration information according to the received update request, adds new abort scenarios and / or guide scenarios, and modifies the association between the abort scenarios and the guide scenarios.

[0072] Each guide scenario corresponds to a prompt word template, and the prompt word template is an incomplete prompt word. After determining the guide scenario, the prompt word template corresponding to the guide scenario is updated according to the abort information. Specifically, the abort information is filled into the prompt word template, and the prompt word template after filling in the abort information (i.e., after updating) is a complete prompt word. The updated prompt word template is used as a prompt word, and a large model is inputted to obtain corresponding guide information.

[0073] According to the abort information, the abort scenario is determined, the guide scenario associated with the abort scenario is determined according to the business orchestration information, and the prompt word template is determined according to the guide scenario. This can more flexibly and accurately determine the prompt word template, and facilitate the generation of accurate guide information. The business orchestration information is configurable, and the abort scenario and the guide scenario can be more flexibly associated together, meeting different business processing needs, efficiently and accurately returning the user to the workflow.

[0074] According to another reference embodiment of the present disclosure, when determining the guide scenario corresponding to the abort scenario according to the pre-set business orchestration information, the business logic corresponding to the abort scenario is first determined according to the business orchestration information. Specifically, the business orchestration information is used to manage the business logic under different abort scenarios, and the business orchestration information includes the business logic corresponding to each abort scenario, which guides the user to return to the workflow and continue processing the workflow.

[0075] The business logic corresponding to the abort scenario is executed, and the execution result of the business logic is obtained. The data type of the execution result can be a string, such as “yes”, “no”, “success”, “unsuccessful”, can be a numerical value, such as “10”, “20”, and can also be a Boolean type, such as “true”, “false”. After obtaining the execution result, the guide scenario corresponding to the abort scenario is determined according to the execution result. For example, in the case where the data type of the execution result is a numerical value, the execution result is compared with a plurality of numerical value ranges set in advance, each numerical value range corresponds to a guide scenario, the numerical value range to which the execution result belongs is taken as a target numerical value range, and the guide scenario corresponding to the target numerical value range is taken as the guide scenario corresponding to the abort scenario.

[0076] Exemplarily, the abort scenario is "page D appears an exception", the business logic corresponding to the abort scenario is a method "http: / / api / test", the business logic is called, the execution result of the business logic is obtained, in a case where the execution result is "yes", it is determined that the guide scenario of the abort scenario is the guide scenario E1, and in a case where the execution result is "no", it is determined that the guide scenario of the abort scenario is the guide scenario E2.

[0077] It should be noted that the business logic corresponding to the abort scenario can be implemented in the manner of a business interface or a custom script. The business logic corresponding to the abort scenario is configurable, and the execution subject of the embodiment of the present disclosure modifies the business interface or the custom script corresponding to the abort scenario according to the received update request, and further modifies the business logic corresponding to the abort scenario.

[0078] The execution of the business logic corresponding to the abort scenario, the determination of the guide scenario corresponding to the abort scenario according to the execution result of the business logic, the flexible and accurate determination of the guide scenario, and the improvement of the accuracy of the guide information.

[0079] According to another reference embodiment of the present disclosure, before the guide information is displayed to the pre-set front-end page, the intention information associated with the guide information is determined, and it is determined that the intention information does not meet the pre-set guide condition. The large model outputs the corresponding intention information while generating the guide information according to the prompt word, and the intention information is used to describe the user intention, which includes: purchase, consultation, reservation, return to the previous page, and the like.

[0080] The guide condition includes a set of intention information, in a case where the intention information output by the large model belongs to the set of intention information, it is determined that the intention information meets the guide condition, which indicates that the guide information output by the large model is inaccurate, and further processing needs to be performed to obtain more accurate guide information.

[0081] Specifically, in the case that the intention information meets the guidance condition, the following steps are executed in a loop to determine the guidance information displayed to the front-end page: taking the intention information as the intention information of the first round, determining the user purpose information corresponding to the intention information of the current round, generating a new prompt word of the current round according to the user purpose information; determining a large model of the current round according to the large model, inputting the new prompt word of the current round into the large model of the current round, and obtaining new intention information and new guidance information output by the large model of the current round; in the case that the new intention information of the current round does not meet the guidance condition, the loop is terminated, and the new guidance information of the current round is taken as the guidance information displayed to the front-end page; in the case that the new intention information of the current round meets the guidance condition, the new intention information of the current round is taken as the intention information of the next round. Specifically, the pre-set business arrangement information is queried to determine the purpose information associated with the intention information, for example, in the case that the intention information is "purchase", the corresponding purpose information is "obtain purchase discount consultation". According to the user purpose information, the corresponding prompt word template is filled in to generate the corresponding new prompt word, the new prompt word is input into the large model, and the new guidance information and the new intention information output by the large model are obtained. Then it is judged whether the new intention information meets the guidance condition.

[0082] FIG. 3 is a schematic diagram of a large model dialogue chain according to one embodiment of the present disclosure. As shown in FIG. 3, the user initiates a dialogue on the front-end page, triggering the workflow termination mechanism, and the execution subject of the embodiment of the present disclosure takes the text input by the user on the front-end page as the termination information, generates a prompt word 1 according to the termination information; then inputs the prompt word 1 into the large model to generate guidance information 1 and intention information 1; judges whether the intention information 1 is the target intention, in the case that the intention information 1 is not the target intention, the guidance information 1 is displayed on the front-end page and returned to the user; in the case that the intention information 1 is the target intention, the prompt word 2 is generated according to the termination information and the guidance information 1; then the prompt word 2 is input into the large model to generate guidance information 2 and intention information 2; then the previous judgment step is repeated until the intention information is not the target intention, and the corresponding guidance information is returned to the user. It should be noted that the input for generating the guidance information includes the text input by the user on the front-end page, the user behavior data, and the guidance information previously generated and returned to the user; the generation process of the guidance information is regarded as an input-output process, and the guidance information output by the previous process is the input information of the next process.

[0083] According to the intention information associated with the guidance information, it is judged whether new guidance information and intention information need to be generated again, and through multiple large model outputs, the final guidance information returned to the user is obtained, which can accurately determine the intention and purpose of the user, facilitate to provide accurate guidance information for the user, and help the user quickly return to the process node of the workflow.

[0084] According to an example of the present disclosure, when determining the large model of the current round according to the large model, it is determined whether the new prompt word of the current round matches the large model. For example, a prompt word template corresponding to the new prompt word is determined, the pre-set business arrangement information is queried, and it is determined whether the large model associated with the prompt word template is the large model to be input. If so, it is determined that the new prompt word matches the large model. For another example, the pre-set business arrangement information is queried to determine the prompt word condition corresponding to the large model. The prompt word condition includes: the maximum length of the prompt word, the keywords that the prompt word should (or should not) include, the absence of null values in the prompt word, and the like. In the case where the new prompt word meets the prompt word condition corresponding to the large model, it is determined that the new prompt word matches the large model. In response to determining that the new prompt word of the current round matches the large model, the large model is used as the large model of the current round, and the new prompt word is input into the large model.

[0085] In response to determining that the new prompt word of the current round does not match the large model, a target large model that matches the new prompt word of the current round is selected from the pre-set plurality of large models. Specifically, the new prompt word is compared with the prompt word conditions of the remaining large models, and the large model corresponding to the prompt word condition met by the new prompt word is used as the target large model, or the business arrangement information is queried to determine the prompt word template corresponding to the new prompt word, and the large model associated with the prompt word template is used as the target large model. Then the new prompt word is input into the target large model to obtain new guidance information and new intent information output by the target large model, and it is determined whether the new intent information meets the guidance condition.

[0086] It should be noted that there are corresponding workflows in the insurance, logistics, e-commerce, finance and other business fields. Business knowledge in multiple business fields is obtained, the same business knowledge in the business field is used to pre-train the large model, and different business fields correspond to the large model. Each large model has business knowledge of its own business field.

[0087] In the case where the prompt word does not match the large model, the target large model is selected from the plurality of large models, which can improve the flexibility of the large model output and meet different analysis requirements of the large model.

[0088] According to another embodiment of the present disclosure, before obtaining the new guidance information and the new intent information of the large model output of the current round, historical guidance information associated with the user is obtained. Specifically, the historical guidance information is guidance information previously generated and returned to the user, which helped the user return to the workflow and continue processing the workflow. The execution subject of the embodiment of the present disclosure has a short-term memory function, and the previously generated guidance information is stored. When it is necessary to generate new guidance information and new intent information, the execution subject of the embodiment of the present disclosure obtains the historical guidance information of the corresponding user, that is, the guidance information previously generated and returned to the user, combines the historical guidance information with the suspension information, inputs the combined information into the large model, and makes the large model output the new guidance information and the new intent information according to the new prompt word and the historical guidance information.

[0089] For example, when the user F processed the workflow last time, the execution subject of the embodiment of the present disclosure was actively asked “how to buy accident insurance”. At this time, the workflow was suspended, the user input was only text, and the execution subject of the embodiment of the present disclosure analyzed the guidance information and the intent information according to the user input text and other suspension information in combination with the large model, and stored the guidance information as historical guidance information. When the user F clicked the pop-up window to close this time, the suspension information included the action of the user closing the page, the page stopped before closing, and then the execution subject of the embodiment of the present disclosure obtained the historical guidance information corresponding to the user, that is, the user wanted to buy accident insurance last time, but the workflow of processing life insurance was abnormally closed this time, and it was determined that the user had the idea of buying life insurance, thereby generating the corresponding guidance information to guide the user to return to the workflow of life insurance.

[0090] Based on the historical guidance information previously output to the user, the new guidance information is generated, which can further improve the accuracy of the guidance information and improve the processing efficiency of the workflow.

[0091] According to another embodiment of the present disclosure, a plurality of mechanisms for triggering suspension of the workflow are pre-set, and each mechanism for triggering suspension of the workflow corresponds to a way of obtaining suspension information of the workflow. Specifically, the execution subject of the embodiment of the present disclosure generates a page component iframe and a dialog box in the front-end page, and then displays a sub-business page (i.e., a workflow page) in the iframe. The user processes the workflow in the sub-business page, and when the user encounters a problem, the user raises the problem in the dialog box, at which time the mechanism for triggering suspension of the workflow is triggered, and the front-end page sends a dialog request to the execution subject of the embodiment of the present disclosure through postMessage. Then, the execution subject of the embodiment of the present disclosure determines the dialog information of the user according to the received dialog request, and takes the dialog information as the suspension information of the workflow.

[0092] Alternatively, the execution subject of the embodiments of the present disclosure monitors events occurring on the front-end page, and in response to detecting the occurrence of a target event, determines whether the target event belongs to a pre-set set of abort events. If the target event belongs to the set of abort events, the execution subject determines event information of the target event, and uses information included in the event information as abort information of the workflow.

[0093] The user asks questions or requests a dialogue to the execution subject of the embodiments of the present disclosure, and detects that the workflow is abnormally exited or interrupted due to certain operations of the user, which can reduce the difficulty of use of the user and improve the efficiency of the user in processing the workflow.

[0094] FIG. 4 is a schematic diagram of a technical architecture of workflow recovery according to one referable embodiment of the present disclosure. As shown in FIG. 4, the execution subject 401 of the embodiments of the present disclosure includes a dialogue box front-end, a dialogue business back-end, a large model dialogue back-end, a business orchestration management, and a dialogue orchestration management, and the like. Specifically, a business component iframe is generated in the dialogue box front-end, and a sub-business front-end page is displayed in the iframe, each sub-business front-end page corresponding to a link in the workflow (i.e., a workflow page). The user executes a business process in the sub-business front-end page, and then asks a question in the dialogue box front-end page. The sub-business front-end synchronizes business information to the dialogue box front-end through postMessage. The dialogue box front-end arouses the corresponding sub-business front-end according to the user request, so as to make the user return to the previously aborted workflow. The sub-business front-end and the sub-business back-end internally interact business information, and the sub-business back-end synchronizes business information with the dialogue business back-end, such as abnormal information in the sub-business execution process, execution results of the workflow, and the like. The dialogue box front-end reports the front-end business state to the dialogue business back-end. The dialogue business back-end is used to determine an abort scenario, and the large model dialogue back-end is used to determine a guide scenario, and generate guide information according to the abort information, and then return the guide information to the user through the dialogue business back-end and the dialogue box front-end. The business orchestration management is used to manage the abort scenario, the business logic corresponding to the abort scenario, and the like. The dialogue orchestration management is used to manage the guide scenario corresponding to the abort scenario, generate a large model dialogue chain, and the like. The business orchestration management and the dialogue orchestration management dynamically configure the dialogue business back-end and the large model dialogue back-end.

[0095] FIG. 5 is a schematic diagram of the main flow of workflow recovery in multi-module cooperation according to one embodiment of the present disclosure. As shown in FIG. 5, the execution subject of the embodiment of the present disclosure includes a page dialogue receiving module, a page event listening module, a business result receiving module, a prompt word generating module, a dialogue business processing module, a dialogue assembling module, a page dialogue display module, and the like. Among them, the page dialogue receiving module is used to receive the dialogue request sent by the user, trigger the workflow suspension mechanism process, the page event listening module is used to listen to the page event, trigger the workflow suspension process, the business result receiving module is used to receive the execution result of the business logic corresponding to the workflow, in the case of abnormal information, trigger the workflow suspension process. The prompt word generating module is used to obtain the suspension information, convert the suspension scenario into a guide scenario, determine the prompt word template, generate the prompt word, the dialogue business processing module is used to generate the guide information according to the prompt word, the historical dialogue between the execution subject of the embodiment of the present disclosure and the user, the historical guide information, the suspension information, and the like, the dialogue assembling module is used to assemble the historical dialogue; the page dialogue display module is used to display the guide information, have a dialogue with the user, answer the user's questions, and help the user return to the workflow.

[0096] FIG. 6 is a schematic diagram of the main flow of the method of workflow recovery according to one embodiment of the present disclosure. As shown in FIG. 6, the method of workflow recovery can include:

[0097] Step S601, in response to receiving the dialogue request, determining the dialogue information of the user according to the dialogue request, and taking the dialogue information as the suspension information of the workflow;

[0098] Step S602, determining the suspension scenario of the workflow according to the suspension information;

[0099] Step S603, determining the business logic corresponding to the suspension scenario according to the pre-set business orchestration information;

[0100] Step S604, executing the business logic, obtaining the execution result of the business logic, and determining the guide scenario corresponding to the suspension scenario according to the execution result;

[0101] Step S605, updating the prompt word template corresponding to the guide scenario according to the suspension information, and taking the updated prompt word template as the prompt word;

[0102] Step S606, inputting the prompt word into the pre-set large model, and making the large model generate the guide information of the workflow according to the prompt word;

[0103] Step S607, displaying the guide information to the pre-set front-end page, and making the user processing the workflow return to the process node included in the workflow according to the guide information, and continue to process the workflow.

[0104] The specific implementation of the method for workflow recovery of one referable embodiment of the present disclosure has been described in detail above, and thus repeated description is not provided here.

[0105] According to a second aspect of the embodiments of the present disclosure, a device for workflow recovery is provided.

[0106] FIG. 7 is a schematic diagram of main modules of the device for workflow recovery according to the embodiments of the present disclosure. As shown in FIG. 7, the device 700 for workflow recovery mainly includes:

[0107] The first generation module 701 is configured to, in response to obtaining the suspension information of the workflow, generate a corresponding prompt word according to the suspension information.

[0108] The second generation module 702 is configured to input the prompt word into a pre-set large model, so that the large model generates guidance information of the workflow according to the prompt word.

[0109] The guidance module 703 is configured to display the guidance information to a pre-set front-end page, so that a user processing the workflow returns to a process node included in the workflow according to the guidance information and continues to process the workflow.

[0110] According to one referable embodiment of the present disclosure, generating the corresponding prompt word according to the suspension information includes: determining a suspension scenario of the workflow according to the suspension information; determining a guidance scenario corresponding to the suspension scenario according to pre-set business orchestration information; and updating a prompt word template corresponding to the guidance scenario according to the suspension information, taking the updated prompt word template as the prompt word.

[0111] According to another referable embodiment of the present disclosure, determining the guidance scenario corresponding to the suspension scenario according to the pre-set business orchestration information includes: determining a business logic corresponding to the suspension scenario according to the business orchestration information; executing the business logic to obtain an execution result of the business logic; and determining the guidance scenario corresponding to the suspension scenario according to the execution result.

[0112] According to still another referable embodiment of the present disclosure, the guidance module 703 is further configured to: before displaying the guidance information to the pre-set front-end page, determine intent information associated with the guidance information and determine that the intent information does not meet a pre-set guidance condition; and in a case where the intent information meets the pre-set guidance condition, cyclically execute the following steps to determine the guidance information displayed to the front-end page:

[0113] With the intention information as the intention information of the first round, the user purpose information corresponding to the intention information of the current round is determined, and the new prompt word of the current round is generated according to the user purpose information; the large model of the current round is determined according to the large model, the new prompt word of the current round is input into the large model of the current round, and the new intention information and the new guide information output by the large model of the current round are obtained; in the case that the new intention information of the current round does not meet the guide condition, the loop is terminated, and the new guide information of the current round is taken as the guide information displayed to the front-end page; in the case that the new intention information of the current round meets the guide condition, the new intention information of the current round is taken as the intention information of the next round.

[0114] According to another reference embodiment of the present disclosure, the guide module 703 is further configured to: in response to determining that the new prompt word of the current round matches the large model, taking the large model as the large model of the current round; in response to determining that the new prompt word does not match the large model, screening a target large model matching the new prompt word of the current round from the plurality of large models set in advance; and taking the target large model as the large model of the current round.

[0115] According to another reference embodiment of the present disclosure, the guide module 703 is further configured to: obtain historical guide information associated with the user; and input the historical guide information into the large model of the current round, so that the large model of the current round outputs the new guide information and the new intention information of the current round according to the new prompt word of the current round and the historical guide information.

[0116] According to an embodiment of the present disclosure, the obtained suspension information of the workflow includes: in response to receiving a dialogue request, determining dialogue information of the user according to the dialogue request, and taking the dialogue information as the suspension information of the workflow; or in response to detecting occurrence of a target event, determining whether the target event belongs to a set of pre-set suspension events, and in the case that the target event belongs to the set of suspension events, determining event information of the target event, and taking the event information as the suspension information of the workflow.

[0117] It should be noted that the specific implementation of the device for restoring the workflow in the embodiments of the present disclosure has been described in detail in the above method for restoring the workflow, and therefore the repeated content will not be described here.

[0118] According to the technical scheme of the embodiment of the present disclosure, the large model is used to analyze the suspension information of the workflow, and the corresponding guide information is obtained, so that the user returns to the process node of the workflow according to the guide information and continues to process the workflow, which can improve the processing efficiency of the workflow, reduce the difficulty of processing the workflow for the user, and improve the user experience; the guide scene corresponding to the suspension scene is determined according to the business arrangement information, and the prompt word is generated according to the guide scene, which can flexibly and accurately generate the prompt word; the business logic corresponding to the suspension scene is executed, and the guide scene is determined according to the execution result of the business logic, which can flexibly and efficiently determine the guide scene; the intention information associated with the guide information is used to determine whether new guide information and intention information need to be generated again, and the guide information finally returned to the user is obtained through multiple large model outputs, which can accurately determine the intention and purpose of the user, facilitate the accurate guide information provided for the user, and help the user quickly return to the process node of the workflow; in the case that the prompt word does not match the large model, the target large model is selected from multiple large models, which can improve the flexibility of the large model output and meet different analysis requirements of the large model; the new guide information is generated based on the historical guide information previously output to the user, which can further improve the accuracy of the guide information and improve the processing efficiency of the workflow; the triggering time of the method includes that the user asks or requests a dialogue to the execution subject of the embodiment of the present disclosure, and it is detected that the workflow is abnormally exited or interrupted due to some operations of the user, which can reduce the difficulty of use of the user and improve the efficiency of the user in processing the workflow.

[0119] According to a third aspect of the embodiment of the present disclosure, an electronic device is provided, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided by the first aspect of the embodiment of the present disclosure.

[0120] According to a fourth aspect of the embodiment of the present disclosure, a computer readable medium is provided, which stores a computer program, and the program is executed by a processor to implement the method provided by the first aspect of the embodiment of the present disclosure.

[0121] FIG. 8 shows an exemplary system architecture 800 to which the method of workflow recovery or the apparatus of workflow recovery can be applied.

[0122] As shown in FIG. 8, the system architecture 800 can include terminal devices 801, 802, 803, a network 804, and a server 805. The network 804 is used to provide a communication link medium between the terminal devices 801, 802, 803 and the server 805. The network 804 can include various connection types, such as wired, wireless communication links, or optical fiber cables, etc.

[0123] The user can use the terminal devices 801, 802, 803 to interact with the server 805 through the network 804 to receive or send messages and the like. Various communication client applications can be installed on the terminal devices 801, 802, 803, such as workflow recovery type applications, workflow query type applications, search type applications, instant messaging tools, mailbox clients, social platform software, and the like (only as examples).

[0124] The terminal devices 801, 802, 803 can be various electronic devices with display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0125] The server 805 can be a server providing various services, such as a background management server providing support for a request for workflow recovery sent by an upstream terminal device 801, 802, 803 (only as an example). The background management server can generate a corresponding prompt word in response to obtaining suspension information of a workflow, generate guidance information of the workflow according to the prompt word by inputting the prompt word into a pre-set large model, display the guidance information to a pre-set front-end page, make a user processing the workflow return to a process node included in the workflow according to the guidance information, continue processing the workflow, and feed back a workflow recovery situation (only as an example) to the terminal device.

[0126] It should be noted that the method for workflow recovery provided by the embodiments of the present disclosure is generally executed by the server 805, and accordingly, the apparatus for workflow recovery is generally arranged in the server 805. The method for workflow recovery provided by the embodiments of the present disclosure can also be executed by the terminal devices 801, 802, 803, and accordingly, the apparatus for workflow recovery can be arranged in the terminal devices 801, 802, 803.

[0127] It should be understood that the number of terminal devices, networks, and servers in FIG. 8 is only illustrative. Any number of terminal devices, networks, and servers can be provided according to implementation needs.

[0128] Reference is made below to FIG. 9, which shows a structural schematic diagram of a computer system 900 of a terminal device suitable for use to implement the embodiments of the present disclosure. The terminal device shown in FIG. 9 is only an example and should not bring any limitation to the functions and use range of the embodiments of the present disclosure.

[0129] As shown in FIG. 9, the computer system 900 includes a central processing unit (CPU) 901 which can perform various appropriate actions and processes in accordance with a program stored in a read only memory (ROM) 902 or a program loaded from a storage section 908 into a random access memory (RAM) 903. In the RAM 903, various programs and data required for the operation of the system 900 are also stored. The CPU 901, the ROM 902, and the RAM 903 are connected to each other through a bus 904. An input / output (I / O) interface 905 is also connected to the bus 904.

[0130] The following components are connected to the I / O interface 905: an input section 906 including a keyboard, a mouse, etc.; an output section 907 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 908 including a hard disk, etc.; and a communication section 909 including a network interface card such as a LAN card, a modem, etc. The communication section 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the I / O interface 905 as necessary. A removable recording medium 911 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 910 as necessary, so that a computer program read therefrom is installed into the storage section 908 as necessary.

[0131] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program in accordance with embodiments of the present disclosure. For example, embodiments of the present disclosure include a computer program including a computer program carried on a computer-readable medium, the computer program containing program codes for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 909, and / or installed from the removable recording medium 911. When the computer program is executed by the central processing unit (CPU) 901, the above-described functions defined in the system of embodiments of the present disclosure are executed.

[0132] It should be noted that the computer-readable medium in the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium may, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination thereof. More specific examples of the computer-readable storage medium can include, but are not limited to, an electrical connection having one or more wires, 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), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. 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 conjunction with an instruction execution system, device or apparatus. In the present disclosure, the computer-readable signal medium can include a data signal carried in a baseband or as a part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, which can send, transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, device or apparatus. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wireline, optical cable, RF, etc., or any suitable combination of the above.

[0133] The flowcharts and block diagrams in the drawings illustrate the possible architectures, functionality, and operations of systems, methods, and computer program products in accordance with various embodiments of the present disclosure. In this regard, each block in the flowcharts 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 functions. 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 noted in succession can in fact be executed substantially concurrently or in the reverse order, depending on the functionality involved. It will also be noted that each block in the block diagrams or flowcharts, and combinations of blocks in the block diagrams or flowcharts, 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.

[0134] The modules described in the embodiments of the present disclosure can be implemented in the form of software or in the form of hardware. The described modules can also be arranged in a processor, for example, a processor can be described as including a first generation module, a second generation module, and a guidance module, wherein the names of these modules do not constitute a limitation of the modules themselves in some cases, for example, the first generation module can also be described as a "module for generating a prompt word according to suspension information".

[0135] As another aspect, the embodiments of the present disclosure also provide a computer readable medium, which can be included in the device described in the above embodiments, or can exist separately without being assembled into the device. The computer readable medium carries one or more programs, and when the one or more programs are executed by the device, the device implements the following method: in response to obtaining suspension information of a workflow, generating a corresponding prompt word according to the suspension information; inputting the prompt word into a pre-set large model, so that the large model generates guidance information of the workflow according to the prompt word; and displaying the guidance information to a pre-set front-end page, so that a user processing the workflow returns to a process node included in the workflow according to the guidance information, and continues to process the workflow.

[0136] According to the technical solutions of the embodiments of the present disclosure, the suspension information of the workflow is analyzed based on the large model to obtain corresponding guidance information, so that the user returns to the process node of the workflow according to the guidance information and continues to process the workflow, which can improve the processing efficiency of the workflow, reduce the difficulty of the user processing the workflow, and improve the user experience; the prompt word is generated according to the guidance scene corresponding to the suspension scene determined according to the business arrangement information, which can flexibly and accurately generate the prompt word; the business logic corresponding to the suspension scene is executed, and the guidance scene is determined according to the execution result of the business logic, which can flexibly and efficiently determine the guidance scene; the intention information associated with the guidance information is judged to determine whether new guidance information and intention information need to be generated again, and the final guidance information returned to the user is obtained through multiple large model outputs, which can accurately determine the intention and purpose of the user, facilitate to provide accurate guidance information for the user, and help the user quickly return to the process node of the workflow; in the case that the prompt word does not match the large model, the target large model is selected from multiple large models, which can improve the flexibility of the large model output and meet different analysis requirements of the large model; based on the historical guidance information previously output to the user, new guidance information is generated, which can further improve the accuracy of the guidance information and improve the processing efficiency of the workflow; the triggering time of the method includes that the user asks or requests a dialogue to the execution subject of the embodiments of the present disclosure, and it is detected that the workflow is abnormally exited or abnormally interrupted due to some operations of the user, which can reduce the difficulty of the user and improve the efficiency of the user processing the workflow.

[0137] The specific implementation described above does not constitute a limitation on the protection scope of the embodiments of the present disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can occur depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the embodiments of the present disclosure should be included in the protection scope of the embodiments of the present disclosure.

Claims

1. A method for workflow recovery, comprising: in response to obtaining suspension information of a workflow, generating a corresponding prompt word according to the suspension information; inputting the prompt word into a pre-set large model, so that the large model generates guidance information of the workflow according to the prompt word; displaying the guidance information to a pre-set front-end page, so that a user processing the workflow returns to a process node included in the workflow according to the guidance information and continues processing the workflow.

2. The method of claim 1, wherein, generating a corresponding prompt word according to the suspension information, comprising: determining a suspension scenario of the workflow according to the suspension information; determining a guidance scenario corresponding to the suspension scenario according to pre-set business orchestration information; updating a prompt word template corresponding to the guidance scenario according to the suspension information, and taking the updated prompt word template as the prompt word.

3. The method of claim 2, wherein, determining a guidance scenario corresponding to the suspension scenario according to pre-set business orchestration information, comprising: determining a business logic corresponding to the suspension scenario according to the business orchestration information; executing the business logic to obtain an execution result of the business logic; determining a guidance scenario corresponding to the suspension scenario according to the execution result.

4. The method of claim 1, wherein, Before displaying the guidance information to the pre-set front-end page, the method further comprises: determining intent information associated with the guidance information and determining that the intent information does not meet a pre-set guidance condition; in a case where the intent information meets the pre-set guidance condition, cyclically executing the following steps to determine the guidance information displayed to the front-end page: taking the intent information as the intent information of a first round, determining user purpose information corresponding to the intent information of a current round according to the user purpose information, generating a new prompt word of the current round according to the user purpose information, determining a large model of the current round according to the large model, inputting the new prompt word of the current round into the large model of the current round, obtaining new intent information and new guidance information output by the large model of the current round, in a case where the new intent information of the current round does not meet the guidance condition, cyclically terminating, and taking the new guidance information of the current round as the guidance information displayed to the front-end page, and in a case where the new intent information of the current round meets the guidance condition, taking the new intent information of the current round as intent information of a next round.

5. The method of claim 4, wherein, determining a large model of a current round according to the large model, comprising: in response to determining that a new prompt word of the current round matches the large model, taking the large model as the large model of the current round; in response to determining that the new prompt word of the current round does not match the large model, screening a target large model matching the new prompt word of the current round from a plurality of pre-set large models, and taking the target large model as the large model of the current round.

6. The method of claim 5, wherein, Before obtaining new guidance information and new intent information output by the large model of the current round, the method further comprises: obtaining historical guidance information associated with the user; inputting the historical guidance information into the large model of the current round, so that the large model of the current round outputs the new guidance information and the new intent information of the current round according to the new prompt word of the current round and the historical guidance information.

7. The method of claim 1, wherein, The obtained suspension information of the workflow includes: In response to receiving the conversation request, determining conversation information of the user according to the conversation request, and taking the conversation information as the suspension information of the workflow; Or, in response to detecting occurrence of a target event, judging whether the target event belongs to a pre-set suspension event set, and in the case that the target event belongs to the suspension event set, determining event information of the target event, and taking the event information as the suspension information of the workflow.

8. An apparatus for workflow resumption, comprising: A first generation module configured to, in response to obtaining suspension information of a workflow, generate a corresponding prompt word according to the suspension information; A second generation module configured to input the prompt word into a pre-set large model, so that the large model generates guide information of the workflow according to the prompt word; A guide module configured to display the guide information to a pre-set front-end page, so that a user processing the workflow returns to a process node included in the workflow according to the guide information, and continues to process the workflow.

9. An electronic device, comprising: One or more processors; A storage device configured to store one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1-7.

10. A computer readable medium having stored thereon a computer program, the program being executed by a processor to implement the method of any one of claims 1-7.

11. A computer program product comprising a computer program, the computer program being executed by a processor to implement the method of any one of claims 1-7.

Images