Digital avatar message processing system

Abstract

The invention provides a message processing system for digital copies, relates to the technical field of artificial intelligence, and is used for still ensuring continuous promotion of businesses under the condition that employees are not online. The system comprises a communication platform used for sending a knowledge matching request to an intelligent agent platform through a first digital branch of a first user account under the condition that a service request message sent to the first user account is obtained and the first user account is in an offline state, the knowledge matching request is used for requesting to process the service content in the service request message. And the agent platform is used for calling a knowledge base of the first agent through the first agent associated with the first digital cloned body, processing the knowledge matching request and obtaining a knowledge matching result of the knowledge matching request. And the intelligent agent platform sends the knowledge matching result to the first digital cloned body. And the communication platform is also used for sending the knowledge matching result to a second user account through the first digital branch, and the second user account is an account sending the service request.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a message processing system for digital clones. Background Technology

[0002] In current collaborative office environments, when employees are offline due to meetings, leave, or focused work, the system typically responds only with preset automatic reply mechanisms, such as "I'm in a meeting, I'll reply later" or "Message received, will handle upon returning to work," and other generic responses. These replies lack understanding and knowledge of the specific business issues, failing to provide effective information addressing the actual needs of the collaborator. This results in collaboration remaining at the level of message confirmation rather than achieving genuine task progress.

[0003] Therefore, how to ensure business continuity even when employees are offline has become a pressing technical problem that needs to be solved. Summary of the Invention

[0004] This application provides a digital clone message processing system to solve the problem of low efficiency in business continuity when employees are offline.

[0005] To achieve the above objectives, this application adopts the following technical solution: Firstly, this application provides a message processing system for digital clones. The system includes an intelligent agent platform and a communication platform. The communication platform is used to, upon receiving a business request message sent to a first user account and when the first user account is offline, send a knowledge matching request to the intelligent agent platform through the first digital clone of the first user account. The knowledge matching request is used to request processing of the business content in the business request message. The intelligent agent platform is used to process the knowledge matching request by calling the knowledge base of the first intelligent agent associated with the first digital clone, and obtain the knowledge matching result of the knowledge matching request. The intelligent agent platform is also used to send the knowledge matching result to the first digital clone when the knowledge matching result meets preset business conditions. The communication platform is also used to send the knowledge matching result to a second user account through the first digital clone, where the second user account is the account that sent the business request.

[0006] Based on the above technical solution, by configuring a first digital avatar for the first user account in the communication platform and deeply binding it to the personal knowledge base and the first intelligent agent associated with the intelligent agent platform, it is possible to enable the digital avatar to receive and respond to business request messages from the second user account even when the first user account is offline. The communication platform initiates a knowledge matching request to the intelligent agent platform through the digital avatar. The intelligent agent platform uses the first intelligent agent associated with the digital avatar to call the exclusive knowledge base for content analysis and matching. When the matching result meets the preset business conditions, the platform accurately feeds back the knowledge to the digital avatar. Finally, the digital avatar automatically sends the processing result back to the second user account. This achieves highly accurate and personalized collaborative responses without human intervention, significantly improving the continuity of collaboration and information response efficiency during employee offline periods, and avoiding the problems of hollow responses and communication interruptions caused by traditional default reply mechanisms.

[0007] In one possible implementation, the communication platform is further configured to obtain the account information of the first intelligent agent and send an intelligent agent binding request to the intelligent agent platform, the intelligent agent binding request including the account information of the first intelligent agent. The intelligent agent platform is further configured to verify the account information of the first intelligent agent, and if the account information of the first intelligent agent passes verification, send a binding response message to the communication platform, the binding response message indicating that the first intelligent agent can be invoked. The communication platform is further configured to, in response to the binding response message, generate a first digital clone of the first user account.

[0008] In another possible implementation, the agent binding request also includes: the identifier of the communication platform and the first user account. The agent platform is also used to store permission management information, which indicates the platform or account with permission to invoke the agent. The agent platform is further used to verify the permissions of the communication platform identifier and the first user account based on the permission management information. If both the communication platform identifier and the first user account pass the permission verification, a binding response message is sent to the communication platform, indicating that the first agent can be invoked.

[0009] In another possible implementation, the intelligent agent platform is further configured to determine the knowledge matching degree based on the knowledge base and business content of the first intelligent agent. The knowledge matching degree is the degree of matching between the business content and the knowledge base of the first intelligent agent. The intelligent agent platform is also configured to, when the knowledge matching degree is greater than or equal to a preset matching degree threshold, process the knowledge matching request by calling the knowledge base of the first intelligent agent and obtain the knowledge matching result of the request.

[0010] In another possible implementation, the communication platform is also used to carry a first prompt message when sending the knowledge matching result to the second user account through the first digital clone. The first prompt message is used to indicate that the knowledge matching result was sent by the first digital clone.

[0011] In another possible implementation, the communication platform is further configured to issue a manual processing prompt message when it is determined that the first intelligent agent cannot obtain a processing result matching the business request message. The manual processing prompt message indicates that the digital clone does not have the capability to process the business request message. The communication platform is also configured to obtain the target business result input by the first user account and send the target business result and a second prompt message to the second user account. The second prompt message indicates that the target business result was sent by the first user account.

[0012] In another possible implementation, the intelligent agent platform is also used to send a failure response message to the communication platform when the knowledge matching degree is less than a preset matching degree threshold or the knowledge matching result does not meet the preset business conditions. The failure response message is used to indicate that the first intelligent agent cannot obtain a processing result that matches the business request message. The knowledge matching degree is the degree of matching between the business content and the knowledge base of the first intelligent agent.

[0013] In another possible implementation, the communication platform is also used to obtain a business failure message sent by the second user account to the first digital clone. The business failure message is used to indicate that the knowledge matching result does not match the business request message.

[0014] In another possible implementation, the communication platform is further configured to, upon receiving a business request message and finding the first user account to be offline, allocate the business request message to the first digital clone and store the business request message. The communication platform is also configured to, if it is determined that the first intelligent agent cannot obtain a processing result matching the business request message, allocate the business request message to the first user account.

[0015] In another possible implementation, the intelligent agent platform is also used to acquire the user's work record information and generate the user's employee knowledge base based on the work record information. The intelligent agent platform is also used to bind the user's employee knowledge base to the first intelligent agent. Attached Figure Description

[0016] Figure 1 This is a system architecture diagram of a message processing system for a digital clone provided in an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] The terms “first” and “second” in the specification and claims of this application are used to distinguish different objects, rather than to describe a specific order of objects.

[0019] Furthermore, the terms "comprising" and "having," and any variations thereof, used in the description of this application are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or modules is not limited to the steps or modules listed, but may optionally include other steps or modules not listed, or may optionally include other steps or modules inherent to such process, method, product, or device.

[0020] Furthermore, in the embodiments of this application, the words "exemplary" or "for example" are used to indicate that they are examples, illustrations, or descriptions. Any embodiment or design that is described as "exemplary" or "for example" in this application should not be construed as being more preferred or advantageous than other embodiments or design options. Specifically, the use of the words "exemplary" or "for example" is intended to present concepts in a concrete manner.

[0021] With the deep integration of artificial intelligence and office automation technologies, enterprise collaboration systems are gradually evolving from traditional person-to-person communication models towards intelligent and unattended operation. In knowledge-intensive industries such as civil aviation, finance, and technology, employees daily handle a large number of collaboration requests based on historical experience and proprietary documents, such as project reviews, customer communication process consultations, and technical parameter queries. These tasks heavily rely on individuals' accumulated knowledge assets. Meanwhile, communication platforms have become the mainstream entry point for office work, and AI technology is widely used in areas such as document understanding and intelligent question answering, providing a technological foundation for building virtual collaborative entities with cognitive capabilities.

[0022] However, in existing office collaboration systems, despite the presence of instant messaging tools, knowledge management platforms, and general AI assistants, these three components often operate independently, lacking effective integration. Employees' knowledge is scattered across local files, cloud drives, or document systems, making it difficult to access it promptly when collaboration occurs. AI-assisted functions typically generate responses based on public corpora or enterprise shared libraries, failing to reflect individual differences. Even when some systems support message collection or status reminders, they fail to provide automated responses to specific business issues. Overall, when employees are unable to be online in real-time due to meetings, business trips, or vacations, the efficiency of information flow within the organization significantly decreases, and there are noticeable delays and breaks in cross-personnel knowledge transfer, leading to insufficient collaboration continuity, prolonged response cycles, and impacting the overall operational efficiency of the team.

[0023] To address the aforementioned technical problems, this application provides a message processing system for digital clones. In this system, a communication platform, upon receiving a business request message sent to a first user account while the first user account is offline, sends a knowledge matching request to an intelligent agent platform through a first digital clone of the first user account. The knowledge matching request requests processing of the business content in the business request message. The intelligent agent platform, through a first intelligent agent associated with the first digital clone, calls the knowledge base of the first intelligent agent to process the knowledge matching request and obtain a knowledge matching result. The intelligent agent platform is also used to send the knowledge matching result to the first digital clone if the knowledge matching result meets preset business conditions. The communication platform is further used to send the knowledge matching result to a second user account through the first digital clone, where the second user account is the account that sent the business request. In this way, by configuring a first digital clone for the first user account in the communication platform and deeply binding it to the personal knowledge base and the first intelligent agent associated with the intelligent agent platform, it is possible to ensure that even when the first user account is offline, the digital clone can still represent it in receiving and responding to business request messages from the second user account. The communication platform initiates a knowledge matching request to the intelligent agent platform through a digital avatar. The intelligent agent platform uses the first intelligent agent associated with the digital avatar to call a dedicated knowledge base for content analysis and matching. When the matching result meets the preset business conditions, the platform feeds back the accurate knowledge to the digital avatar. Finally, the digital avatar automatically sends the processing result back to the second user account. This achieves highly accurate and personalized collaborative responses without human intervention, significantly improving the continuity of collaboration and information response efficiency when employees are offline, and avoiding the problems of empty replies and communication interruptions caused by traditional default reply mechanisms.

[0024] like Figure 1 As shown, a message processing system for a digital clone provided in an embodiment of this application is provided. The system includes: an intelligent agent platform 101 and a communication platform 102.

[0025] Among them, the intelligent agent platform 101 is responsible for the construction and management of the personal knowledge base. The intelligent agent platform 101 is also used to call intelligent agents to process data in the personal knowledge base and generate processing results.

[0026] For example, the intelligent agent platform 101 is deployed in an enterprise private cloud or a secure and controllable cloud environment to transform unstructured documents uploaded by employees into structured knowledge assets that can be invoked. Its output (i.e., structured knowledge) will serve as the knowledge source for subsequent digital avatar responses to business requests, directly affecting the accuracy of the response.

[0027] The intelligent agent platform 101 may include: a personal document import unit 111, a knowledge processing unit 112, and a personal knowledge base storage unit 113.

[0028] The Personal Document Import Unit 111 provides employees with a unified entry point, supporting the upload of various file formats, including but not limited to Word, PDF, TXT, PPT, and scanned images (such as JPG and PNG). Employees can directly drag and drop or select files to submit to the system via the web browser or client. All uploaded files are temporarily stored in encrypted form and bound to the employee's unique account identifier (such as employee ID) in the Intelligent Agent Platform 101, ensuring data ownership.

[0029] The raw document data received by the personal document import unit 111 will serve as the input source for the knowledge processing unit 112, and is the starting point of the entire knowledge transformation chain.

[0030] The knowledge processing unit 112 is used to perform multi-stage intelligent processing on the received original document. The knowledge processing unit 112 includes: an Optical Character Recognition (OCR) module, a Natural Language Processing (NLP) parsing module, and a knowledge structuring module.

[0031] The OCR module extracts text from image or scanned documents using optical character recognition technology. The NLP parsing module performs semantic segmentation, keyword extraction, and entity recognition (such as customer names, route numbers, and dates) based on natural language processing algorithms (e.g., BERT, Sentence-BERT models). The knowledge structuring module organizes the parsed key information into structured data formats such as "question-answer pairs" or "keyword-content fragments," facilitating efficient subsequent retrieval and matching. For example, "XX route fuel surcharge adjustment plan?" corresponds to a policy description text. The processed structured knowledge is timestamped and its source document path is marked, forming a preliminary knowledge index table.

[0032] The output of the knowledge processing unit 112 will be written into the personal knowledge base storage unit 113, forming the knowledge foundation of the digital clone; in addition, when the employee adds new documents through the employee interaction terminal, this unit will be called again for incremental processing.

[0033] Personal knowledge base storage unit 113 is used to store the user's personal knowledge base.

[0034] For example, the personal knowledge base storage unit 113 adopts a distributed database architecture (such as Elasticsearch + MySQL hybrid storage), using the employee's unique account ID as the index key, to centrally store all structured knowledge data generated by the knowledge processing unit 112. The system supports two retrieval modes: keyword matching retrieval: suitable for explicit terminology queries (such as "cost analysis template"). Semantic similarity retrieval: based on vector embedding technology, it calculates the semantic distance between the user's question and the knowledge item, returning the most relevant results.

[0035] For example, the system defaults to setting the matching threshold to 80%. If the value is lower than this, it is considered "unable to answer effectively" and a manual intervention process is required.

[0036] The personal knowledge base storage unit 113 is the final carrier of knowledge service capabilities in the entire system. Its responsiveness will directly affect whether the digital clone response unit can successfully generate a high-quality response. At the same time, it is also the data landing point when employees supplement knowledge after human intervention.

[0037] Communication platform 102 is used to support the collaborative interaction and response execution of digital clones. Communication platform 102 is the user-facing front-end collaboration entry point in this application, integrated into the enterprise's internal office system. It is responsible for receiving business request messages sent by collaborators and scheduling digital clones to complete automatic responses or trigger manual intervention processes.

[0038] In this embodiment, the communication platform 102 includes: a digital clone registration and binding unit 121, a collaborative message interaction unit 122, a digital clone response unit 123, a manual intervention triggering unit 124, and an employee interaction terminal 125.

[0039] The digital avatar registration and binding unit 121 provides employees with a configuration interface, allowing them to input login credentials (such as account password or API key token) from the intelligent agent platform 101 to initiate a registration application. After receiving the request, the system sends an identity verification request to the intelligent agent platform 101 through a cross-platform data interaction interface to verify whether the account is legitimate and whether the knowledge base initialization has been completed.

[0040] If the verification passes, the system generates a digital avatar that corresponds one-to-one with the employee's account and establishes a mapping relationship between the employee account, the digital avatar, and the personal knowledge base in the background. Afterward, this digital avatar can receive and process collaborative messages on behalf of the employee.

[0041] The collaborative message interaction unit 122 is used to monitor the message stream sent to employee accounts that have been bound to digital clones. When a new collaborative message is detected (such as the text message "Please provide the client contact person for last year's XX project"), the system first determines whether the target employee is currently online (based on heartbeat signals or client activity status).

[0042] If the employee is offline, the system will not push the message to their terminal device, but will forward it to the digital clone response unit 123 for processing. If the employee is online, the message will be pushed according to the normal process to avoid duplicate responses.

[0043] The collaborative message interaction unit 122 can determine whether a message has entered the automated processing flow and provide raw input to the digital clone response unit 123.

[0044] The digital clone response unit 123 is used to receive message streams, call the intelligent agent platform 101 to process the message streams, and respond to the message streams.

[0045] For example, the digital avatar response unit 123 receives a business request message from the collaborative message interaction unit 122 and extracts key semantic information from the message (such as "last year", "XX project", "customer contact"). Then, through a cross-platform data interaction interface, it initiates a knowledge matching request to the personal knowledge base storage unit 113 of the intelligent agent platform 101. The digital avatar response unit 123 can then receive the matching result and provide feedback to the collaborators.

[0046] Optionally, the digital clone response unit 123 can assess whether the matching result meets preset business conditions. If it does, the digital clone response unit 123 feeds back the matching result to the collaborator. If it does not meet the conditions, the digital clone response unit 123 can request manual intervention.

[0047] The manual intervention triggering unit 124 is used to issue a manual processing prompt message when it is determined that the first intelligent agent cannot obtain a processing result that matches the business request message.

[0048] For example, upon receiving a knowledge matching failure notification from the digital clone response unit 123, it is determined that the first intelligent agent cannot obtain a processing result matching the business request message. Alternatively, upon receiving a business failure message sent by a collaborator account, it is determined that the first intelligent agent cannot obtain a processing result matching the business request message.

[0049] The manual intervention trigger unit 124 can generate a pending collaboration reminder, containing complete contextual information (original message, AI attempted reply, collaborator ID, and time of occurrence), and push it to the first user account via the employee interaction terminal 125. Simultaneously, the request is marked as having triggered manual intervention in the message history to prevent duplicate reminders.

[0050] The employee interaction terminal 125 is used by employees to view and process pending collaborative tasks, and is integrated into their daily communication client. Employee interaction terminal 125 can be used to view reminder details, access context information temporarily stored in the manual intervention trigger unit 124, and gain a comprehensive understanding of the background. Employee interaction terminal 125 can also be used to send manual replies, compose responses, and the system automatically marks them as "employee manual reply" and sends them to collaborators. Employee interaction terminal 125 can also be used to supplement and update the knowledge base. If a problem is found to be caused by a lack of knowledge, a new document can be directly uploaded. The system sends it to the personal document import unit 111 of the intelligent agent platform 101 through a cross-platform data interaction interface, triggering a new round of knowledge processing.

[0051] The cross-platform data interaction interface is used to connect the intelligent agent platform 101 and the communication platform 102.

[0052] For example, cross-platform data interaction interfaces can be deployed on a separate security service layer, using a hybrid communication protocol of RESTful API and WebSocket to ensure stability and real-time performance under high concurrency.

[0053] Optionally, requests transmitted through cross-platform data interaction interfaces may carry a valid token.

[0054] Optionally, requests transmitted via cross-platform data interaction interfaces can be encrypted using HTTPS.

[0055] Optionally, a maximum request limit can be set per unit time through the cross-platform data interaction interface.

[0056] The embodiments of this application are described in detail below.

[0057] In some embodiments, users can create personal employee knowledge bases on the agent platform.

[0058] In this embodiment of the application, the intelligent agent platform is also used to obtain the user's work record information and generate the user's employee knowledge base based on the work record information.

[0059] For example, work record information includes: email correspondence (such as a notification about the adjustment of the delivery time of the XX project), meeting minutes (such as the 2025 Q1 customer liaison meeting minutes), approval process records (such as travel expense reimbursement form - Zhang San), and task notes or chat records in the collaboration platform (such as @Li Gong, the interface parameters discussed last time have been updated, please check).

[0060] For example, the intelligent agent platform extracts key content from the above emails and automatically generates structured knowledge entries. For instance, the question might be: Has the delivery date for Project XX changed? The answer is: Originally scheduled for May 10th, it has been adjusted to May 20th. See the email "Notification Regarding the Adjustment of the Delivery Date for Project XX" (March 15, 2025) for details.

[0061] The intelligent agent platform is also used to bind the user's employee knowledge base to the first intelligent agent.

[0062] In this way, when the first agent is invoked later, it can automatically access the employee's knowledge base to obtain the knowledge answer that matches the question.

[0063] In some embodiments, a digital clone can be created for a user account on the communication platform.

[0064] In one possible implementation, a communication platform is used to obtain the account information of the first intelligent agent and send an intelligent agent binding request to the intelligent agent platform, the intelligent agent binding request including the account information of the first intelligent agent.

[0065] It should be noted that the account information of the first intelligent agent can be the user's account information for logging into the intelligent agent platform, or it can be the user account or account identifier of the first intelligent agent. This application embodiment does not limit this.

[0066] For example, in the settings interface of the communication platform, an employee accesses the "Digital Avatar" function module and enters their account information (such as username, password, or API key) registered on the intelligent agent platform. This account corresponds to their exclusive first intelligent agent. The communication platform receives the above account information and generates an intelligent agent binding request.

[0067] The intelligent agent platform is also used to verify the account information of the first intelligent agent. If the account information of the first intelligent agent passes the verification, a binding response message is sent to the communication platform. The binding response message is used to indicate that the first intelligent agent can be invoked.

[0068] In one possible implementation, the agent platform verifies the legitimacy of the account information, confirming that the account has been created and its personal knowledge base has been constructed. After successful verification, the agent platform and the communication platform synchronize permissions, establishing a mapping relationship between the first user account and the first agent.

[0069] The communication platform is also used to respond to the binding response message and generate the first digital clone of the first user account.

[0070] The communication platform can also generate digital clones.

[0071] In some embodiments, business processes can be completed through a digital clone when the first user account is offline.

[0072] In one possible implementation, the communication platform is used to send a knowledge matching request to the intelligent agent platform through the first digital clone of the first user account when it receives a business request message sent to the first user account and the first user account is offline. The knowledge matching request is used to request the processing of the business content in the business request message.

[0073] The knowledge matching request includes: the identifier of the first digital clone.

[0074] For example, the identifier of the first digital clone can be the first user account, or the identifier of the first digital clone can be an ID generated based on the first user account.

[0075] In one possible implementation, the communication platform can receive a business request message sent from a second user account to a first user account. Then, the communication platform can obtain the account status of the first user account to determine if it is offline. If the first user account is offline, a knowledge matching request is sent to the intelligent agent platform through the first digital clone of the first user account.

[0076] For example, account status may include: online status and offline status. Offline status includes: busy status, offline status, away status, etc. This application embodiment does not limit this.

[0077] For example, a business request message could be, "Mr. Zhang, hello. What stage is our 'customer data integration' project with your company currently at? What were the conclusions of the last review meeting?" The business content could include: your company, customer data integration, and review meeting conclusions. Alternatively, a business request message could be, "Mr. Liu, the new intern doesn't know how to upload e-invoices to the system for expense reimbursement. Do you have a standard operating procedure document?" The business content could include: e-invoice reimbursement, uploading to the system, and a standard operating procedure document.

[0078] The intelligent agent platform is used to call the knowledge base of the first intelligent agent associated with the first digital clone, process knowledge matching requests, and obtain the knowledge matching results of the knowledge matching requests.

[0079] In one possible implementation, the agent platform stores multiple knowledge bases, each bound to an agent. Furthermore, the platform stores a first association relationship, which is the association between the digital avatar and the agent. The agent platform can determine the first agent based on the first association relationship and the identifier of the first digital avatar. Then, the platform can input a knowledge matching request into the first agent, which will then access its own knowledge base to process the request and obtain the matching result.

[0080] In some embodiments, the intelligent agent platform can verify whether the first digital clone has the authority to invoke the first intelligent agent.

[0081] In one possible implementation, the agent binding request also includes: the identifier of the communication platform and the first user account. The agent platform is also used to store permission management information, which indicates the platform or account with permission to invoke the agent. The agent platform is further used to verify the permissions of the communication platform identifier and the first user account based on the permission management information. If both the communication platform identifier and the first user account pass the permission verification, a binding response message is sent to the communication platform, indicating that the first agent can be invoked.

[0082] For example, suppose the intelligent agent platform stores information about platforms a, b, and c that can call intelligent agents, and the first intelligent agent is called by user a. If the identifier of the first user account is the same as the identifier of user a, and the communication platform of the first user account is platform a, then the first user account can call the first intelligent agent.

[0083] In some embodiments, before the agent platform processes the knowledge matching request through the first agent, it can determine whether the agent has the capability to process the knowledge matching request.

[0084] In one possible implementation, the intelligent agent platform is further configured to determine a knowledge matching degree based on the knowledge base and business content of the first intelligent agent, wherein the knowledge matching degree is the degree of matching between the business content and the knowledge base of the first intelligent agent. The intelligent agent platform is also configured to, when the knowledge matching degree is greater than or equal to a preset matching degree threshold, process the knowledge matching request by calling the knowledge base of the first intelligent agent, and obtain the knowledge matching result of the knowledge matching request.

[0085] For example, the intelligent agent platform performs semantic retrieval in its personal knowledge base storage unit, calculating the matching degree of each candidate item. If there is a valid result with a knowledge matching degree greater than or equal to a preset matching degree threshold, and it complies with the permission and security policies, the first intelligent agent calls its own knowledge base to process the knowledge matching request and obtain the knowledge matching result. For example, the matching degree threshold can be 80%.

[0086] The intelligent agent platform is also used to send the knowledge matching results to the first digital clone when the knowledge matching results meet the preset business conditions.

[0087] In one possible implementation, the intelligent agent platform can check and process the knowledge matching results to determine whether the knowledge matching results meet the preset business conditions.

[0088] The preset business conditions include at least one of the following: the knowledge matching result contains information from the business request message, the knowledge matching result does not contain sensitive information, and the validity period of the knowledge matching result is within a preset time.

[0089] For example, content integrity checks determine whether the returned content contains key information elements (such as time, subject, action, etc.) to avoid fragmented and invalid responses. Sensitive information filtering, combined with a pre-built strategy library, identifies whether content involves customer privacy, financial data, undisclosed decisions, etc., and blocks it if a match is found. Knowledge timeliness verification compares the document creation / update time with the current date; knowledge that has exceeded its validity period (i.e., a preset time, such as 180 days) and is not specially marked is automatically downgraded or excluded.

[0090] In this embodiment of the application, after the intelligent agent platform completes knowledge matching and returns a knowledge matching result that meets the preset business conditions, the knowledge matching result is transmitted to the first digital clone of the communication platform via the cross-platform data interaction interface.

[0091] The communication platform is also used to send knowledge matching results to a second user account via a first digital avatar. The second user account is the account that sends the business request.

[0092] In this embodiment of the application, the digital clone response unit in the communication platform receives the knowledge matching result and generates a structured reply message. Through the collaborative message interaction unit, the structured reply message is pushed to the client interface of the second user account.

[0093] The structured response message includes: knowledge matching results, the source basis of the knowledge matching results (such as document name, timestamp), and feedback entry point (such as the "Not satisfied? Click to transfer to human agent" button), ensuring that the response is traceable and interactive.

[0094] In some embodiments, the communication platform is further configured to carry a first prompt message when sending the knowledge matching result to the second user account through the first digital clone, the first prompt message being used to indicate that the knowledge matching result was sent by the first digital clone.

[0095] For example, a structured reply message includes a reply identifier of the first digital clone.

[0096] When the communication platform sends knowledge matching results to the second user account through the first digital avatar, it includes a primary notification (such as a "digital avatar auto-reply" label) to clearly indicate that the response was generated by an AI agent rather than by the employee themselves. This enhances the transparency of collaboration, allowing collaborators to clearly identify the source of the message and avoiding misinterpretations of a human response that could lead to trust bias.

[0097] Based on the above technical solution, the communication platform, when receiving a business request message sent to a first user account and the first user account is offline, sends a knowledge matching request to the intelligent agent platform through a first digital clone of the first user account. The knowledge matching request requests the processing of business content in the business request message. The intelligent agent platform, through a first intelligent agent associated with the first digital clone, calls the knowledge base of the first intelligent agent to process the knowledge matching request and obtain the knowledge matching result. The intelligent agent platform is also used to send the knowledge matching result to the first digital clone if the knowledge matching result meets preset business conditions. The communication platform is also used to send the knowledge matching result to a second user account through the first digital clone, where the second user account is the account that sent the business request. In this way, by configuring a first digital clone for the first user account in the communication platform and deeply binding it to the personal knowledge base and the first intelligent agent associated with it in the intelligent agent platform, it is possible to ensure that even when the first user account is offline, the digital clone can still represent it to receive and respond to business request messages from the second user account. The communication platform initiates a knowledge matching request to the intelligent agent platform through a digital avatar. The intelligent agent platform uses the first intelligent agent associated with the digital avatar to call a dedicated knowledge base for content analysis and matching. When the matching result meets the preset business conditions, the platform feeds back the accurate knowledge to the digital avatar. Finally, the digital avatar automatically sends the processing result back to the second user account. This achieves highly accurate and personalized collaborative responses without human intervention, significantly improving the continuity of collaboration and information response efficiency when employees are offline, and avoiding the problems of empty replies and communication interruptions caused by traditional default reply mechanisms.

[0098] In some embodiments, the communication platform is further configured to issue a manual processing prompt message when it is determined that the first intelligent agent cannot obtain a processing result matching the service request message. The manual processing prompt message is used to indicate that the digital clone does not have the ability to process the service request message. The communication platform is also configured to obtain the target service result input by the first user account and send the target service result and a second prompt message to the second user account. The second prompt message is used to indicate that the target service result was sent by the first user account.

[0099] For example, after the digital avatar response unit determines that the first intelligent agent cannot handle the request, it triggers a manual intervention mechanism. The manual intervention triggering unit generates a manual processing prompt message and pushes it to the first user account through the employee interaction terminal. For example, the manual processing prompt message includes: "You have a collaboration request that requires manual processing" and the context of the business request message (such as sender, time, and question content). After logging into the system, the first user account views the reminder on its client and replies with the target business result. The employee interaction terminal sends the target business result to the second user account through the collaboration message interaction unit, along with a second prompt message (such as "Employee has replied" or "Manual reply") to clarify that the message source is a person and not AI.

[0100] In one possible implementation, the communication platform can determine whether the first intelligent agent can obtain a processing result that matches the business request message based on the knowledge matching result and the business request message.

[0101] It should be noted that the process by which the communication platform verifies the knowledge matching results can refer to the implementation method of the intelligent agent platform in verifying the knowledge matching results, which will not be elaborated here.

[0102] In another possible implementation, the intelligent agent platform is also used to send a failure response message to the communication platform when the knowledge matching degree is less than a preset matching degree threshold or the knowledge matching result does not meet the preset business conditions. The failure response message is used to indicate that the first intelligent agent cannot obtain a processing result that matches the business request message.

[0103] In other words, if the knowledge matching degree is less than a preset matching degree threshold, the intelligent agent platform cannot generate a knowledge matching result, and therefore can notify the communication platform that it cannot process the business request message. Alternatively, if the intelligent agent platform generates a knowledge matching result, but the result fails to meet the business conditions, it can also notify the communication platform that it cannot process the business request message.

[0104] In another possible implementation, the communication platform is also used to obtain a business failure message sent by the second user account to the first digital clone. The business failure message is used to indicate that the knowledge matching result does not match the business request message.

[0105] It should be understood that when the first digital clone generates an automatic reply based on the knowledge matching results returned by the intelligent agent platform, if the second user account (i.e., the collaborator) believes that the reply fails to accurately meet its business needs (such as irrelevant content, missing information, incorrect answer, etc.), it can send a business failure message to the first digital clone through interactive controls provided by the communication platform (such as "unsatisfied" or "invalid reply" buttons) or by manually entering feedback information.

[0106] In one possible implementation, after receiving the automatic response from the digital clone, the second user account determines whether the knowledge matching result is valid. If it is deemed invalid, the user initiates a service failure message by clicking a feedback button or sending a specific text command (such as "Unable to resolve"). This service failure message is received by the collaborative message interaction unit of the communication platform and identified as negative feedback to the digital clone. The collaborative message interaction unit forwards the service failure message to the manual intervention triggering unit. The communication platform associates the service request message, the knowledge matching result, and the current feedback from the second user account to form a complete context record.

[0107] In some embodiments, after the communication platform obtains the target business result, it can update the employee knowledge base of the first user account based on the target business result and the business request message.

[0108] In some embodiments, the communication platform is further configured to, upon receiving a service request message and finding that the first user account is offline, assign the service request message to the first digital clone and store the service request message. The communication platform is also configured to, upon determining that the first intelligent agent cannot obtain a processing result matching the service request message, assign the service request message to the first user account.

[0109] For example, after receiving a business request message from a second user account, the collaborative message interaction unit checks the account status of the first user account through the employee status management module. If it is determined to be offline, the message is not directly pushed to the employee client. The communication platform routes the business request message to the first digital clone bound to the first user account, and the collaborative message interaction unit temporarily stores the message content and context to ensure subsequent traceability and recall.

[0110] If the intelligent agent platform returns a valid knowledge matching result, an automatic reply is generated and sent to the second user account. If no valid result is returned, the system determines that the first intelligent agent cannot obtain a matching processing result. The communication platform can redistribute the business request message and its context (including collaborator information, timestamp, and knowledge matching result) to the employee interaction terminal of the first user account. The communication platform generates a to-do reminder, marks it as "requires manual processing," and synchronizes it to the message queue after the first user account comes online.

[0111] The foregoing primarily describes the solutions provided in the embodiments of this application from a system perspective. It is understood that, in order to achieve the aforementioned functions, the digital clone message processing system includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that the digital clone message processing systems described in conjunction with the embodiments disclosed in this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0112] This application embodiment can divide the message processing system for digital clones into functional modules or functional units based on the above system example. For example, each function can be divided into its own functional modules or functional units, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module or functional unit. The module or unit division in this application embodiment is illustrative and represents only one logical functional division; other division methods may be used in actual implementation.

[0113] Through the above description of the implementation methods, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above.

[0114] In the embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between devices or units, and may be electrical, mechanical, or other forms.

[0115] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0116] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0117] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, in essence, or the parts that contribute to the prior art, or all or part of the technical solutions, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0118] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A message processing system for digital clones, characterized in that, The system includes: an intelligent agent platform and a communication platform; The communication platform is used to send a knowledge matching request to the intelligent agent platform through the first digital clone of the first user account when it receives a business request message sent to the first user account and the first user account is offline. The knowledge matching request is used to request the processing of the business content in the business request message. The intelligent agent platform is used to call the knowledge base of the first intelligent agent associated with the first digital clone, process the knowledge matching request, and obtain the knowledge matching result of the knowledge matching request. The intelligent agent platform is also used to send the knowledge matching result to the first digital clone when the knowledge matching result meets the preset business conditions; The communication platform is also used to send the knowledge matching result to the second user account through the first digital clone, where the second user account is the account that sent the business request.

2. The system according to claim 1, characterized in that, The communication platform is also used to obtain the account information of the first intelligent agent and send an intelligent agent binding request to the intelligent agent platform, wherein the intelligent agent binding request includes the account information of the first intelligent agent; The intelligent agent platform is also used to verify the account information of the first intelligent agent. If the account information of the first intelligent agent passes the verification, it sends a binding response message to the communication platform. The binding response message is used to indicate that the first intelligent agent can be invoked. The communication platform is also used to generate the first digital clone of the first user account in response to the binding response message.

3. The system according to claim 2, characterized in that, The smart agent binding request also includes: the identifier of the communication platform and the first user account; The intelligent agent platform is also used to store permission management information, which is used to indicate the platform or account with permission to call the intelligent agent; The intelligent agent platform is further configured to perform permission verification on the identifier of the communication platform and the first user account based on the permission management information. If both the identifier of the communication platform and the first user account pass the permission verification, the platform sends a binding response message to the communication platform. The binding response message indicates that the first intelligent agent can be invoked.

4. The system according to claim 1, characterized in that, The intelligent agent platform is further configured to determine the knowledge matching degree based on the knowledge base of the first intelligent agent and the business content, wherein the knowledge matching degree is the degree of matching between the business content and the knowledge base of the first intelligent agent; The intelligent agent platform is further configured to, when the knowledge matching degree is greater than or equal to a preset matching degree threshold, call the knowledge base of the first intelligent agent to process the knowledge matching request and obtain the knowledge matching result of the knowledge matching request.

5. The system according to claim 1, characterized in that, The communication platform is also used to carry a first prompt message when sending the knowledge matching result to the second user account through the first digital clone. The first prompt message is used to indicate that the knowledge matching result was sent by the first digital clone.

6. The system according to any one of claims 1-5, characterized in that, The communication platform is also used to issue a manual processing prompt message when it is determined that the first intelligent agent cannot obtain a processing result that matches the service request message. The manual processing prompt message is used to indicate that the digital clone does not have the ability to process the service request message. The communication platform is also used to obtain the target service result input by the first user account, and send the target service result and a second prompt message to the second user account, wherein the second prompt message is used to indicate that the target service result was sent by the first user account.

7. The system according to claim 6, characterized in that, The intelligent agent platform is further configured to send a failure response message to the communication platform when the knowledge matching degree is less than the preset matching degree threshold, or when the knowledge matching result does not meet the preset business conditions. The failure response message is used to indicate that the first intelligent agent cannot obtain a processing result that matches the business request message. The knowledge matching degree is the degree of matching between the business content and the knowledge base of the first intelligent agent.

8. The system according to claim 6, characterized in that, The communication platform is also used to obtain a service failure message sent by the second user account to the first digital clone, the service failure message being used to indicate that the knowledge matching result does not match the service request message.

9. The system according to claim 6, characterized in that, The communication platform is also used to, when it receives the service request message and the first user account is offline, allocate the service request message to the first digital clone and store the service request message. The communication platform is further configured to allocate the service request message to the first user account if it is determined that the first intelligent agent cannot obtain a processing result that matches the service request message.

10. The system according to any one of claims 1-5, characterized in that, The intelligent agent platform is also used to acquire the user's work record information and generate the user's employee knowledge base based on the work record information; The intelligent agent platform is also used to bind the user's employee knowledge base to the first intelligent agent.

Images