An intelligent office method, device, equipment and program product

Abstract

This application relates to the field of artificial intelligence technology, and more particularly to an intelligent office method, device, equipment, and program product. The method includes: collecting office information; using a first extraction layer of a pre-set extraction model to extract at least one task flow executed by a target user during task execution from the structured information; using a second extraction layer of the extraction model to extract the target user's habitual characteristics during task execution from unstructured information; constructing a target task processing knowledge base for the target user using the at least one task flow and the habitual characteristics; and executing the task through a user avatar of the target user, based on the target task processing knowledge base, according to the task flow matching the task instructions and processing logic conforming to the habitual characteristics. Through the above method, automated task and workflow processing, or the execution of specific capabilities, based on user-personalized behavior and habitual characteristics can be achieved.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to an intelligent office method, device, equipment, and program product. Background Technology

[0002] Enterprise robots are intelligent interactive programs deployed in enterprise systems that can engage in dialogue with users, help users execute workflows, and assist users with tasks such as office work.

[0003] However, traditional enterprise robots, after receiving user instructions, usually match the corresponding task process from a pre-built general knowledge base and then execute the task. In other words, a single robot is used to handle all the work tasks of various departments in the enterprise, such as taking leave, expense reimbursement, sending and receiving emails, and other work tasks that are involved in different departments.

[0004] These types of robots use the same knowledge base across the entire enterprise, making it difficult to adapt to the different job functions and personal business processing habits of different employees, and thus unable to represent users in office collaboration. Summary of the Invention

[0005] This application discloses an intelligent office method, device, equipment, and program product, which can realize the processing of automated tasks and workflows or the execution of specific capabilities based on users' personalized behaviors and habits.

[0006] In a first aspect, embodiments of this application provide an intelligent office method, the method comprising: Collect office information, which includes structured information and unstructured information. The structured information is process-type data generated by the target user when performing tasks, and the unstructured information is habit-type data generated by the target user when performing tasks. Using the first extraction layer of a pre-set extraction model, at least one task process executed by the target user during task execution is extracted from the structured information; Using the second extraction layer of the extraction model, the habitual features of the target user when performing tasks are extracted from the unstructured information; A target task processing knowledge base for the target user is constructed using the at least one task flow and the habitual features. The user avatar of the target user executes the task according to the target task processing knowledge base, following the task flow that matches the task instructions and the processing logic that conforms to the habitual features.

[0007] In one possible implementation, constructing a target task processing knowledge base for the target user using the at least one task flow and the habitual features includes: The at least one task flow and habit feature are classified according to a preset task type, and task flows and habit features belonging to the same task type are paired to obtain at least one set of paired data. Based on the task type, generate a task identifier for each pair of data; Each pair of data and the corresponding task identifier are stored to obtain the target task processing knowledge base.

[0008] In one possible implementation, the pairing data includes first pairing data and second pairing data. The first pairing data is the pairing result of the overall task flow with the corresponding global habitual features, and the second pairing data is the pairing result of each execution node in the task flow with the corresponding node habitual features.

[0009] In one possible implementation, after constructing the target task processing knowledge base for the target user using the at least one task flow and the habitual features, the method further includes: In response to the generation command triggered by the target user, at least one user clone corresponding to the target user is generated according to the reference configuration information; The reference configuration information includes the task identifiers of the tasks that each user clone needs to handle.

[0010] In one possible implementation, generating at least one user clone corresponding to the target user based on reference configuration information includes at least one of the following steps: Based on a preset time period, the frequency with which the target user executes the task flow corresponding to each task identifier, the target task identifier is determined, and initial reference configuration information is generated based on the target task identifier. Initial reference configuration information is provided to the target user through a user interface. Modified reference configuration information is obtained based on the target user's modification of the initial reference configuration information. The at least one user clone is generated using the modified reference configuration information.

[0011] In one possible implementation, the step of executing the task through the user avatar of the target user, based on the target task processing knowledge base and following a task flow matching the task instructions and processing logic conforming to habitual characteristics, includes: The user clone responds to the task command triggered by the target user, parses the task command to obtain the task identifier of the task to be executed, queries the target task processing knowledge base to find matching data that matches the task identifier, and executes the task to be processed according to the task flow in the matching data and the processing logic that conforms to the habitual characteristics in the matching data.

[0012] In one possible implementation, the method further includes: When it is determined that the user clone is unable to complete the task to be processed, a prompt message is sent to the target user through the user interface; The target task processing knowledge base is updated based on the process data and habit data generated by the target user when processing the task to be processed.

[0013] Secondly, embodiments of this application provide an intelligent office device, the device comprising: The data collection module is used to collect office information, which includes structured information and unstructured information. The structured information is process-type data generated by the target user when performing tasks, and the unstructured information is habit-type data generated by the target user when performing tasks. The first extraction module is used to extract at least one task process executed by the target user during the execution of a task from the structured information using the first extraction layer of a pre-set extraction model. The second extraction module is used to extract the habitual features of the target user when performing tasks from the unstructured information using the second extraction layer of the extraction model. A building module is used to construct a target task processing knowledge base for the target user using the at least one task flow and the habitual features; The execution module is used to execute tasks by the user clone of the target user, based on the target task processing knowledge base, and in accordance with the task flow matching the task instructions and the processing logic conforming to habitual characteristics.

[0014] Thirdly, embodiments of this application provide an electronic device, the electronic device comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the method of the first aspect described above.

[0015] Fourthly, embodiments of this application provide a computer storage medium storing a computer program for causing a computer to perform the method described in the first aspect above.

[0016] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the method described in the first aspect above.

[0017] This application provides an intelligent office method that collects process-related data and habit-related data generated by users during their work to construct a personalized task processing knowledge base for users. This personalized task processing knowledge base enables users to process tasks according to their habit characteristics, achieving automated task and workflow processing or execution of specific capabilities based on users' personalized behaviors and habit characteristics. Moreover, in the process of constructing the personalized task processing knowledge base, a pre-trained extraction model is used to analyze the collected office information, which can accurately and quickly identify the user's task flow and habit characteristics during their work. Attached Figure Description

[0018] Figure 1 This is a schematic diagram illustrating an application scenario of an intelligent office method according to an exemplary embodiment of the present invention; Figure 2 This is a schematic diagram illustrating an application scenario of another intelligent office method according to an exemplary embodiment of the present invention; Figure 3 This is a schematic diagram illustrating a smart office method according to an exemplary embodiment of the present invention; Figure 4 A schematic diagram of a user interface as an example of an exemplary embodiment of the present invention; Figure 5 This is a schematic diagram of another user interface as an example of an exemplary embodiment of the present invention; Figure 6 This is a schematic diagram illustrating the overall process of an intelligent office method according to an exemplary embodiment of the present invention; Figure 7 A schematic diagram of an intelligent office device according to an exemplary embodiment of the present invention; Figure 8 This is a schematic diagram of an electronic device as an example of an exemplary embodiment of the present invention. Detailed Implementation

[0019] The principles and spirit of this application will now be described with reference to several exemplary embodiments. It should be understood that these embodiments are provided merely to enable those skilled in the art to better understand and implement this application, and are not intended to limit the scope of this application in any way. Rather, these embodiments are provided to make this disclosure more thorough and complete, and to fully convey the scope of this disclosure to those skilled in the art.

[0020] Those skilled in the art will recognize that embodiments of this application can be implemented as a system, apparatus, method, or computer program product. Therefore, this disclosure can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0021] In this article, it is important to understand that any number of elements in the accompanying figures is for illustrative purposes and not for limitation, and any naming is for distinction only and has no limiting meaning.

[0022] The following describes some of the concepts involved in the embodiments of this application.

[0023] User clone: ​​In this embodiment of the application, the user clone is used to replace or assist the user in handling tasks, and the task is handled using processing logic that conforms to the user's habit characteristics.

[0024] Virtual Assistant: In this embodiment of the application, the virtual assistant can be an intelligent robot with information collection function, which can collect the user's office information and assist the user in office work. For example, it can have a conversation with the user, understand the user's intention, manage the user's user clone, and provide feedback to the user on the operation status of the user clone. This embodiment of the application does not specifically limit the purpose of the virtual assistant.

[0025] Structured information: In this embodiment, structured information can also be referred to as hard features, possessing quantifiable and measurable attributes. It can be process-like data generated by a user when performing a task. For example, when a user processes task A, it needs to be executed according to the process of node a—node b—node c.

[0026] Unstructured information: In this embodiment, unstructured information can be soft features, which are habitual data generated by users when performing tasks. For example, when users handle work tasks such as scheduling and sending / receiving emails, their preferences, expression styles, decision-making habits, and capability boundaries.

[0027] The following description, in conjunction with the accompanying drawings, illustrates an intelligent office method according to an embodiment of this application.

[0028] like Figure 1 The image shown is an intelligent office system provided in an embodiment of this application. The system includes: a task processing knowledge base corresponding to each user, such as... Figure 1 The task processing knowledge base 1, task processing knowledge base 2, ..., task processing knowledge base 3 are shown in the figure; each user has at least one user clone, such as Figure 1 As shown, Task Processing Knowledge Base 1 corresponds to User Avatar 11, User Avatar 12, ..., User Avatar 13; Task Processing Knowledge Base 2 corresponds to User Avatar 21, User Avatar 22, ..., User Avatar 23; and Task Processing Knowledge Base 3 corresponds to User Avatar 31, User Avatar 32, ..., User Avatar 33.

[0029] Among them, the user clone is used to replace or assist the user in handling work-related tasks, and the task knowledge base is used to provide the user clone with knowledge when handling work tasks.

[0030] Figure 1 The task processing knowledge base and user avatars in the intelligent office system described herein can be deployed in one place, such as on a single server, or they can be deployed separately, such as... Figure 2 As shown.

[0031] Figure 2 The intelligent office system shown includes a server 201 and a client 202. The task processing knowledge base corresponding to each user is deployed on the server 201, and at least one user avatar corresponding to each user is deployed on the client 202. The server 201 and client 202 can communicate with each other. The client 202 also provides a user interface for displaying the task processing status of the user avatar, providing prompts, etc. This embodiment does not specifically limit the content displayed on the user interface. It should be noted that... Figure 2 The client 202 shown can also be used as a terminal device for users when they are working; Figure 1 The smart office system shown also includes a user interface.

[0032] Traditional user cloning lacks a mechanism for segmenting users by job function, project domain, communication partner, or skill type. This makes it difficult to achieve permission isolation, tool capability differentiation, and decision threshold differentiation across different work scenarios, and results in misalignment between output style and user task processing methods. Therefore, this application provides an intelligent office method, such as... Figure 3 As shown, the method includes: S301: Collect office information.

[0033] In one possible implementation, the aforementioned office information includes structured information and unstructured information. The structured information is process-type data generated by the user when performing a task, which can be multiple process nodes in a fixed order that the user goes through when performing a certain task. For example, when the user performs task A, the process follows the flow of node a—node b—node c.

[0034] The unstructured information mentioned above is habitual data generated by users when performing tasks. For example, when sending emails, users are used to bolding important content in the email.

[0035] In one possible implementation, an information collection tool can be deployed on the user's terminal device (such as a computer or mobile phone). This tool can collect user operation information from external devices like mice and keyboards, information about the terminal device's processor, memory, processes, and network, log information from when the user logs into the system, and user behavior information on the terminal device's user interface. This application does not limit the source of the collected information. The aforementioned structured and unstructured information originates from information collected by the information collection tool.

[0036] In one possible implementation, the aforementioned information collection tool can be a virtual assistant, which can be deployed in the form of an intelligent robot in the intelligent office system provided in this application embodiment. While collecting user office information, it can also assist users in their work.

[0037] by Figure 2 Taking the illustrated intelligent office system as an example, users may use multiple terminal devices (i.e., client 202) to collaborate while working, such as editing documents on a computer and communicating with clients using a mobile terminal. To more conveniently collect the soft features (i.e., unstructured information) generated during the user's work process, a virtual assistant can be deployed on the mobile terminal. This virtual assistant can remain online while the user is working and can collect the user's expression style and decision-making habits when communicating with other people in real time. Due to the portability of mobile terminals, deploying a virtual assistant on a mobile terminal can also collect unstructured information (such as preferences, expression style, and capability boundaries) of the user's tasks during non-office hours, thereby uncovering more user habitual characteristics and making the generated user avatar more closely resemble the user's task-handling style. When the user only uses a computer for work, the virtual assistant can also be deployed on the computer; this embodiment of the application does not specifically limit this.

[0038] In one possible implementation, the aforementioned office information can be collected after the user logs into the preset office software, or after entering the user interface of the client 202. As long as the monitoring permissions are granted, structured and unstructured information generated when the user uses non-office software can also be collected. This application does not specifically limit this aspect.

[0039] S302: Using the first extraction layer of the pre-set extraction model, extract at least one task process executed by the target user during task execution from the structured information; using the second extraction layer of the extraction model, extract the habitual features of the target user when executing tasks from the unstructured information.

[0040] In order to accurately analyze the collected office information and build a task processing knowledge base, this application embodiment can use a pre-set extraction model to extract the required features from structured and unstructured information.

[0041] The extraction model described above comprises multiple extraction layers. The first extraction layer extracts at least one task flow executed by the user during task execution from structured information. The second extraction layer extracts the user's habitual features during task execution from unstructured information. This extraction model is trained based on sample structured information, the task flows contained within the sample structured information, sample unstructured information, and the habitual features contained within the sample unstructured information.

[0042] In one possible approach, a first extraction model and a second extraction model can be trained separately. The first extraction model is used to extract task flow from structured information and is trained based on sample structured information and the task flow contained in the sample structured information. The second extraction model is used to extract user habitual features when performing tasks from unstructured information and is trained based on habitual features contained in sample unstructured information.

[0043] S303: Construct a target task processing knowledge base for the target user using the at least one task flow and the habitual features, and execute the task according to the target user's user avatar based on the target task processing knowledge base, following the task flow that matches the task instructions and the processing logic that conforms to the habitual features.

[0044] To enable user clones to handle tasks according to user habits, embodiments of this application analyze user habit characteristics and build a personalized task processing knowledge base for users. The implementation method for building a personalized task processing knowledge base is as follows: The at least one task flow and habit feature are classified according to a preset task type, and task flows and habit features belonging to the same task type are paired to obtain at least one set of paired data. Based on the task type, generate a task identifier for each pair of data; Each pair of data and the corresponding task identifier are stored to obtain the target task processing knowledge base.

[0045] The aforementioned target task processing knowledge base corresponds to the target user, that is, it is a task processing knowledge base that belongs exclusively to the target user, including the task process and habitual characteristics of the target user when processing tasks.

[0046] Different tasks have different task flows, and users execute these flows using their own processing logic when handling a task. Therefore, each task corresponds to a specific task flow and habitual characteristics. For example, for task A, the task flow is node a—node b—node c, and user 1's habitual characteristic for executing task A is: to handle it after logging into the intelligent office system every day.

[0047] The task flow and habit features are classified according to task type. Then, the task flow and habit features of the same task type are paired to form multiple sets of paired data. The paired data are then identified according to task type. When storing, the paired data and its corresponding task identifier are stored together. The storage format is shown in Table 1 below.

[0048] Table 1

[0049] The task identifier mentioned above can be the name of the task type, such as a schedule creation task, expense reimbursement task, email sending and receiving task, invoice issuance task, etc., or it can be the number corresponding to the task type, such as the task identifier of a schedule creation task being 001; the task identifier of an expense reimbursement task being 002; or it can be composed of both the task name and the number, such as the task identifier of a schedule creation task being: schedule 001. In this embodiment of the application, no specific limitation is made.

[0050] The aforementioned pairing data includes first pairing data and second pairing data. The first pairing data is the pairing result of the overall task flow and the corresponding global habitual feature. For example, pairing data 1 includes: flow a (flow a includes node 1—node 2—node 3) + global habitual feature; the second pairing data is the pairing result of each execution node in the task flow and the corresponding node habitual feature. For example, pairing data 2 includes: node 1 + node 1 habitual feature.

[0051] In one possible implementation, for the second pairing data, when the execution nodes in different second pairing data belong to the same task flow, their task type names and task type numbers may be the same. For example, node 1 in pairing data 2 and node 2 in pairing data 3 both belong to flow a, and the task type name corresponding to flow a is "reimbursement task". In this case, when setting the task identifier, the same task identifier can be set for pairing data 2 and pairing data 3, such as both using "reimbursement task" as the task identifier; they can also be distinguished, such as setting the task identifier for pairing data 2 as "reimbursement task_a" and the task identifier for pairing data 3 as "reimbursement task_b"; or the task identifier can be set according to the execution order. For example, for flow a, node 1 (belonging to pairing data 2) is executed first, and then node 2 (belonging to pairing data 3) is executed. When setting the task identifier, the task identifier for pairing data 2 is set as "reimbursement task_1", and the task identifier for pairing data 3 is set as "reimbursement task_2", where "_1" and "_2" can indicate the execution order of the execution nodes, that is, the node corresponding to "_1" is executed first, and then the node corresponding to "_2" is executed.

[0052] Furthermore, the task processing database can also store the frequency and execution time of the task flow corresponding to each task identifier being executed by the user, so as to provide a basis for the subsequent generation of user clones.

[0053] It should be noted that for the same task, different users may perform the same task flow, but their habits may be completely different. For example, for task A, user 2's habit might be to process it before leaving get off work each day. Therefore, in this embodiment, a personalized task processing knowledge base with its own style is set up for each different user.

[0054] The aforementioned personalized task processing knowledge base can provide a basis for users to handle various tasks. Specifically, the user clone can respond to the task instructions triggered by the user, parse the task instructions, obtain the task identifier of the task to be executed, and then query the matching data that matches the task identifier from the user's corresponding task processing knowledge base. The user clone will then execute the task to be processed according to the task flow in the matching data and the processing logic that conforms to the habitual characteristics in the matching data.

[0055] In one possible implementation, this application also provides a method for generating clones, as described below.

[0056] In response to the generation command triggered by the target user, at least one user clone corresponding to the target user is generated according to reference configuration information, wherein the reference configuration information includes the task identifier of the task that each virtual assistant needs to handle.

[0057] Users can send generation commands to the virtual assistant deployed on client 202 of the smart office system via dialogue, such as... Figure 4 As shown, the user can enter a generation command in the dialog box. The server 201 can respond to the generation command obtained from the virtual assistant, determine the target task identifier based on the frequency of the task flow corresponding to each task identifier being executed by the user according to the preset time period, and generate initial reference configuration information based on the target task identifier.

[0058] For example, if the task process corresponding to task identifier A was executed by the user 10 times in a week, and the task process corresponding to task identifier B was executed by the user 5 times in a week, the high execution frequency of the task process corresponding to task identifier A indicates that the task process corresponding to task identifier A is of higher importance. Server 101 will use task identifier A as the initial reference configuration information to reply to the user through the virtual assistant.

[0059] like Figure 5In the dialog box shown, server 201 generates initial reference configuration information such as "A user clone will be generated for you, which will be used to execute the task flow corresponding to task identifier A" and displays it to the user through the display interface of client 201 (that is, replying to the user through a virtual assistant). The user can modify the initial reference configuration information and obtain the modified reference configuration information, such as the user can... Figure 5 If you reply "Generate a clone of task identifier A and task identifier B" on the interface shown, the virtual assistant will submit the user's instructions to server 201, and server 201 will generate the user clone according to the modified reference configuration information.

[0060] Specifically, server 201 can generate one identifier for task identifier A and one for task identifier B respectively; it can also generate only one user clone that can execute the task flow corresponding to task identifier A and the task flow corresponding to task identifier B. The specific generation can be based on user needs and is not specifically limited here.

[0061] In one possible implementation, in addition to providing reference configuration information based on execution frequency, it can also be based on the frequency (or duration) of user pauses or delays when executing a task flow corresponding to a certain task identifier; the manual takeover rate of the task, i.e., the frequency at which user intervention is required when a user fails to execute a task; and the frequency of errors made by the user when executing a task flow corresponding to a certain task identifier. This application does not specifically limit these aspects.

[0062] Once at least one virtual assistant is created, it can assist the user in handling various tasks. If the virtual assistant is unable to complete a task, a notification message can be sent to the target user via the computer user interface or the virtual assistant on the mobile terminal. This notification message informs the user that the virtual assistant cannot complete the task and requires manual intervention. Furthermore, the notification message can also include a work report on the virtual assistant's task completion, allowing the user to assess the virtual assistant's performance and better manage the task. It should be noted that a work report needs to be generated when the virtual assistant is unable to complete a task; a work report can also be generated when the virtual assistant successfully completes a task, helping the user optimize workflows in subsequent tasks.

[0063] After a user takes over a task that their clone hasn't finished, the information collection tool will further collect process-related and habit-related data from when the user was handling the task. It will then use the newly collected process-related and habit-related data to update the user's task handling knowledge base, providing a reference for subsequent clones to handle the task.

[0064] The following is based on Figure 6The process of an intelligent office method provided in the embodiments of this application will be described in its entirety.

[0065] In one possible office scenario, the target user collaborates using both a mobile app and a computer with a virtual assistant deployed. Both the mobile app and the computer app are client-side (201).

[0066] The virtual assistant can collect information and communicate with the target user continuously in natural language to better obtain data on the user's task-handling habits, such as user preferences, expression style, decision-making habits, and capability boundaries. Furthermore, if the target user executes a task on a mobile device, the virtual assistant can also collect process-related data. An information collection tool is deployed on the computer to collect the target user's work information, including process-related and habit-related data. Specifically, this includes the target user's behavior with external connected devices such as the mouse and keyboard, communication records with others, and summarized work documents. Because process-related and habit-related data are intertwined in the work information generated by the target user during task processing, the server-side 201, after obtaining the work information, needs to use a pre-built extraction model to extract at least one task process and habit feature.

[0067] After obtaining at least one task flow and habitual feature, the task processing knowledge base of the target user is constructed using the obtained task flow and habitual feature. The specific construction method is as described in the implementation method in S303 above, and will not be repeated here.

[0068] The task processing knowledge base can guide the creation of user avatars and guide them in handling various tasks. Specifically, server 201 can display reference configuration information to the user through a virtual assistant based on the frequency of each task performed by the target user recorded in the task processing knowledge base. The reference configuration information includes task identifiers of the tasks that the user avatar can handle. The target user can modify this reference configuration information to generate a user avatar according to their own needs.

[0069] The target user can generate user clones with different capabilities according to their needs. Each user clone can perform the tasks required by the user on the client side (201). For example... Figure 6 In the process, user clone 1 needs to work with colleague 1. User clone 1 can work with colleague 1 according to the target task processing knowledge base, the task process that matches the task instructions and the processing logic that conforms to habitual characteristics.

[0070] When user clone 1 is unable to respond to colleague 1's question, server 201 can generate a work report. This work report corresponds to the task "Coordinating work with colleague 1" and can include specific work content, urgent matters, record of difficulties, review and updates, etc. This work report can also be recorded in the task processing knowledge base and associated with the task "Coordinating work with colleague 1" to provide a reference for user clones who subsequently perform this task.

[0071] Furthermore, the virtual assistant can monitor the work status of each user's clone, report to the target user in a timely manner, and communicate with the user to obtain more information about the target user's work regarding the task of "coordinating with colleague 1". In this case, if the target user can provide timely feedback, the virtual assistant and the server 201 can collaboratively analyze the feedback and continue to guide user clone 1 to continue coordinating with colleague 1 based on the user's feedback. The target user does not need to completely take over the task of "coordinating with colleague 1" for the time being. At the same time, the task processing knowledge base needs to be updated using the user's feedback information to provide reference for future work.

[0072] If the target user cannot provide timely feedback, the virtual assistant can guide user clone 1 to reply to colleague 1 with a message such as "The target user will handle this later." At the same time, the virtual assistant can provide a work report to the target user so that the target user is clear about the status of user clone 1 in completing the task.

[0073] This application provides an intelligent office method that collects process data and habit data generated by users during their work to build a personalized task processing knowledge base for users. The personalized task processing knowledge base enables users to handle tasks according to their habit characteristics. Moreover, in the process of building the personalized task processing knowledge base, a pre-trained extraction model is used to analyze the collected office information, which can accurately and quickly identify the user's task flow and habit characteristics during their work.

[0074] The user avatar provided in this application embodiment can assist users in their work by utilizing a personalized task processing knowledge base (user avatars can also be generated based on the task processing knowledge base). This allows the user avatar to understand and characterize the user's work style, communication and work methods, preferences, personality, and work insights. This personalized task processing knowledge base provides a stable basis for the generation and task processing of the avatar. The output style and processing method of the user avatar in this application embodiment are aligned with the user, distinguishing it from ordinary robots. This can improve enterprise collaboration efficiency and consistency, and realize the automated processing of tasks and workflows based on the user's personalized behavior and habit characteristics, or the execution of specific capabilities. Furthermore, when generating user avatars, different user avatars can be generated according to user needs (e.g., users can generate different user avatars based on job processes / project domains / communication objects / capability types, etc.), and each user avatar can handle different tasks, improving permission isolation, capability matching, and risk control capabilities, while also supporting continuous editing and optimization by users.

[0075] This application embodiment also deploys a virtual assistant on the mobile device. The mobile virtual assistant remains on standby and, through natural language communication, more easily acquires soft features such as the user's expression habits, emotions, and speaking style. These features exist in the form of a virtual avatar (such as large eyes), making it easier to arouse the user's desire to communicate. Users will be willing to communicate with it about work and even life, making feature collection more comprehensive. The virtual assistant can also serve as an information collection and feedback entry point, clarifying issues from work report reviews to the user in a natural way, collecting the user's processing preferences and response methods, and updating and optimizing the clone capability in reverse. During the clone generation process, the virtual assistant can also communicate with the user in a timely manner to clarify any problems encountered, quickly closing the loop on clone construction matters and improving the efficiency and fit of clone generation. During the user clone generation process, the user can directly request the generation of a clone, modify clone settings, and provide verbal feedback on clone issues through verbal communication with the virtual assistant, making clone generation and optimization more convenient and flexible.

[0076] Furthermore, the embodiments of this application, through designs such as processing reports, instructing users to take over tasks urgently, and reviewing and updating problem records, can continuously align with users' real-world work methodologies under an explainable, interventionable, and reviewable mechanism, thereby promoting long-term optimization of cloning capabilities and strategies.

[0077] Based on the same inventive concept, embodiments of this application provide a smart office device, such as... Figure 7 As shown, the device includes: The data collection module 701 is used to collect office information, which includes structured information and unstructured information. The structured information is process data generated by the target user when performing tasks, and the unstructured information is habit data generated by the target user when performing tasks. The first extraction module 702 is used to extract at least one task process executed by the target user during the execution of a task from the structured information using the first extraction layer of a pre-set extraction model. The second extraction module 703 is used to extract the habitual features of the target user when performing tasks from the unstructured information using the second extraction layer of the extraction model. Module 704 is configured to construct a target task processing knowledge base for the target user using the at least one task flow and the habitual features. The execution module 705 is used to execute the task by the user clone of the target user according to the target task processing knowledge base, following the task flow that matches the task instructions and the processing logic that conforms to habitual characteristics.

[0078] In one possible implementation, the building module 704 is used for: The at least one task flow and habit feature are classified according to a preset task type, and task flows and habit features belonging to the same task type are paired to obtain at least one set of paired data. Based on the task type, generate a task identifier for each pair of data; Each pair of data and the corresponding task identifier are stored to obtain the target task processing knowledge base.

[0079] In one possible implementation, the construction module 704 is used to determine: The pairing data includes first pairing data and second pairing data. The first pairing data is the pairing result of the overall task flow with the corresponding global habit features, and the second pairing data is the pairing result of each execution node in the task flow with the corresponding node habit features.

[0080] In one possible implementation, the device further includes a generation module for: In response to the generation command triggered by the target user, at least one user clone corresponding to the target user is generated according to the reference configuration information; The reference configuration information includes the task identifiers of the tasks that each user clone needs to handle.

[0081] In one possible implementation, the generation module is used to perform at least one of the following steps: Based on a preset time period, the frequency with which the target user executes the task flow corresponding to each task identifier, the target task identifier is determined, and initial reference configuration information is generated based on the target task identifier. Initial reference configuration information is provided to the target user through a user interface. Modified reference configuration information is obtained based on the target user's modification of the initial reference configuration information. The at least one user clone is generated using the modified reference configuration information.

[0082] In one possible implementation, the execution module 705 is used to: The user clone responds to the task command triggered by the target user, parses the task command to obtain the task identifier of the task to be executed, queries the target task processing knowledge base to find matching data that matches the task identifier, and executes the task to be processed according to the task flow in the matching data and the processing logic that conforms to the habitual characteristics in the matching data.

[0083] In one possible implementation, the device further includes an update module for: When it is determined that the user clone is unable to complete the task to be processed, a prompt message is sent to the target user through the user interface; The target task processing knowledge base is updated based on the process data and habit data generated by the target user when processing the task to be processed.

[0084] Based on the same inventive concept, this application provides an electronic device, the device comprising: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform an intelligent office method provided in the embodiments of this application.

[0085] The following reference Figure 8 To describe an electronic device 80 according to this embodiment of the present application. Figure 8 The electronic device 80 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0086] like Figure 8 As shown, the electronic device 80 is presented in the form of a general-purpose electronic device. The components of the electronic device 80 may include, but are not limited to: at least one processor 81, at least one memory 82, and a bus 83 connecting different system components (including memory 82 and processor 81).

[0087] The processor 81 is used to read and execute instructions from the memory 82, so that the at least one processor can execute an intelligent office method provided in the above embodiments.

[0088] Bus 83 represents one or more of several bus structures, including a memory bus or memory controller, peripheral bus, processor, or local bus using any of the various bus structures.

[0089] The memory 82 may include a readable medium in the form of volatile memory, such as random access memory (RAM) 821 and / or cache memory 822, and may further include read-only memory (ROM) 823.

[0090] The memory 82 may also include a program / utility 825 having a set (at least one) of program modules 824, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0091] Electronic device 80 can also communicate with one or more external devices 84 (e.g., keyboard, pointing device, etc.), and with one or more devices that enable a user to interact with electronic device 80, and / or with any device that enables electronic device 80 to communicate with one or more other electronic devices (e.g., router, modem, etc.). This communication can be performed via input / output (I / O) interface 85. Furthermore, electronic device 80 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 88. As shown, network adapter 86 communicates with other modules used in electronic device 80 via bus 83. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with electronic device 80, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0092] In some possible implementations, various aspects of the intelligent office method provided in this application can also be implemented in the form of a program product, which includes program code. When the program product is run on a computer device, the program code is used to cause the computer device to perform the steps of the intelligent office method according to the various exemplary embodiments of this application described above.

[0093] In addition, this application also provides a computer-readable storage medium storing a computer program for causing a computer to perform the method described in any of the above embodiments.

[0094] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0095] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0096] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0097] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope and intent of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application is also intended to include such modifications and variations.

Claims

1. A smart office method, characterized in that, The method includes: Collect office information, which includes structured information and unstructured information. The structured information is process-type data generated by the target user when performing tasks, and the unstructured information is habit-type data generated by the target user when performing tasks. Using the first extraction layer of a pre-set extraction model, at least one task process executed by the target user during task execution is extracted from the structured information; Using the second extraction layer of the extraction model, the habitual features of the target user when performing tasks are extracted from the unstructured information; A target task processing knowledge base for the target user is constructed using the at least one task flow and the habitual features. The user avatar of the target user executes the task according to the target task processing knowledge base, following the task flow that matches the task instructions and the processing logic that conforms to the habitual features.

2. The method according to claim 1, characterized in that, The construction of a target task processing knowledge base for the target user using the at least one task flow and the habitual features includes: The at least one task flow and habit feature are classified according to a preset task type, and task flows and habit features belonging to the same task type are paired to obtain at least one set of paired data. Based on the task type, generate a task identifier for each pair of data; Each pair of data and the corresponding task identifier are stored to obtain the target task processing knowledge base.

3. The method according to claim 2, characterized in that, The pairing data includes first pairing data and second pairing data. The first pairing data is the pairing result of the overall task flow with the corresponding global habit features, and the second pairing data is the pairing result of each execution node in the task flow with the corresponding node habit features.

4. The method according to claim 2, characterized in that, The method further includes: In response to the generation command triggered by the target user, at least one user clone corresponding to the target user is generated according to the reference configuration information; The reference configuration information includes the task identifiers of the tasks that each user clone needs to handle.

5. The method according to claim 4, characterized in that, The step of generating at least one user clone corresponding to the target user based on the reference configuration information includes at least one of the following steps: Based on a preset time period, the frequency with which the target user executes the task flow corresponding to each task identifier, the target task identifier is determined, and initial reference configuration information is generated based on the target task identifier. Initial reference configuration information is provided to the target user through a user interface. Modified reference configuration information is obtained based on the target user's modification of the initial reference configuration information. The at least one user clone is generated using the modified reference configuration information.

6. The method according to claim 2, characterized in that, The step of executing a task through the user avatar of the target user, based on the target task processing knowledge base and following a task flow matching the task instructions and processing logic conforming to habitual characteristics, includes: The user clone responds to the task command triggered by the target user, parses the task command to obtain the task identifier of the task to be executed, queries the target task processing knowledge base to find matching data that matches the task identifier, and executes the task to be processed according to the task flow in the matching data and the processing logic that conforms to the habitual characteristics in the matching data.

7. The method according to claim 6, characterized in that, The method further includes: When it is determined that the user clone is unable to complete the task to be processed, a prompt message is sent to the target user through the user interface; The target task processing knowledge base is updated based on the process data and habit data generated by the target user when processing the task to be processed.

8. An intelligent office device, characterized in that, The device includes: The data collection module is used to collect office information, which includes structured information and unstructured information. The structured information is process-type data generated by the target user when performing tasks, and the unstructured information is habit-type data generated by the target user when performing tasks. The first extraction module is used to extract at least one task process executed by the target user during the execution of a task from the structured information using the first extraction layer of a pre-set extraction model. The second extraction module is used to extract the habitual features of the target user when performing tasks from the unstructured information using the second extraction layer of the extraction model. A building module is used to construct a target task processing knowledge base for the target user using the at least one task flow and the habitual features; The execution module is used to execute tasks by the user clone of the target user, based on the target task processing knowledge base, and in accordance with the task flow matching the task instructions and the processing logic conforming to habitual characteristics.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the method as described in any one of claims 1-7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-7.

Images