Aggregation system for multi-account message processing

Abstract

This application relates to the field of instant messaging technology and provides an aggregation system for multi-account message processing, which can stably and efficiently push message data. The aggregation system's AI processing device calls an AI service to process a portion of the message data in the message dataset, obtaining a preliminary message dataset. The server aggregates the preliminary message dataset to obtain a session set. Based on its perception of the current view of the instant messaging application's client, the server determines the minimum message data set required for user device interface rendering within the session set. A data sharing device receives the minimum message data sets uploaded by each server, aggregates the minimum message data sets of the same user, obtains the minimum dataset for multiple accounts of the same user, and sends the minimum dataset for multiple accounts to a storage unit. The server retrieves the minimum dataset for multiple accounts from the storage unit and sends it to the client, enabling the user device to perform interface rendering based on the minimum dataset for multiple accounts.

Description

Technical Field

[0001] This application relates to the field of instant messaging technology, and in particular to an aggregation system for multi-account message processing. Background Technology

[0002] In insurance marketing scenarios, sales personnel can communicate with customers through instant messaging applications. In some situations, such as during peak periods for group insurance business, sales personnel may use multiple accounts on instant messaging applications to serve hundreds or even thousands of customers.

[0003] Relying solely on the native capabilities of instant messaging applications for service delivery presents certain challenges. By leveraging an aggregation system, message data from multiple accounts of sales personnel can be distributed and pushed, enabling them to communicate with customers from multiple accounts on a single interface and provide services.

[0004] In this mode, the amount of message data that the aggregation system needs to push is enormous. How to enable the aggregation system to push message data stably and efficiently is a problem that needs to be solved. Summary of the Invention

[0005] Therefore, it is necessary to provide an aggregation system for multi-account message processing to address the aforementioned technical issues.

[0006] This application provides an aggregation system for multi-account message processing, the aggregation system including an AI processing device, a server, a data sharing device, and a storage unit;

[0007] The AI ​​processing device is used to receive the corresponding message dataset pushed by the server of the instant messaging application for the account hosted on this system.

[0008] The AI ​​processing device is used to call AI services to process message data in the message dataset that can be automatically processed by AI, and obtain a preliminary message dataset;

[0009] The server is used to receive the initial screening message dataset, aggregate the initial screening message dataset, and obtain a session set;

[0010] The server is configured to determine, based on its perception of the current view of the client of the instant messaging application, the minimum set of message data required for the user device to render the interface from the set of sessions.

[0011] The data sharing device is used to receive the minimum message data set uploaded by each of the servers, and to aggregate the minimum message data set of the same user to obtain the minimum dataset of multiple accounts of the same user, and to send the minimum dataset of multiple accounts to the storage unit.

[0012] The server is used to obtain the minimum dataset of multiple accounts from the storage unit and send it to the corresponding client so that the user device can perform interface rendering based on the minimum dataset of multiple accounts.

[0013] In one embodiment, when the AI ​​processing device invokes an AI service to process message data in the message dataset that can be automatically processed by AI to obtain a preliminary message dataset, the specific steps include:

[0014] The AI ​​processing device is used to call AI services to perform semantic recognition on message data in the message dataset;

[0015] The AI ​​processing device is also used to determine, based on the semantic recognition results, whether the preset question-and-answer knowledge base includes matching response content;

[0016] The AI ​​processing device is also used to respond based on the response content if the preset question-and-answer knowledge base includes matching response content;

[0017] The AI ​​processing device is also used to add the message data to the initial screening message dataset if the preset question-and-answer knowledge base does not contain matching response content.

[0018] In one embodiment, when the AI ​​processing device responds based on the response content, it specifically includes:

[0019] The AI ​​processing device is also used to invoke the PRA device to simulate a client and reply with the response content.

[0020] In one embodiment, when the server aggregates the initial screening message dataset to obtain a session set, the specific steps include:

[0021] The server is used to aggregate the initial message dataset based on the sender and receiver of the messages to obtain a session set.

[0022] In one embodiment, when the server determines the minimum set of message data required for user device interface rendering from the session set based on its perception of the current view of the instant messaging application's client, it specifically includes:

[0023] The server is used to receive operation data sent by the user device for the client of the instant messaging application;

[0024] The server is also used to structure the operation data to obtain view information of the view currently in which the client is located;

[0025] The server is further configured to determine, based on the session identifier carried in the view information, the session in the session set that matches the session identifier, so as to obtain the minimum message data set required by the user device for interface rendering.

[0026] In one embodiment, when the data sharing device sends the multi-account minimum dataset to the storage unit, it specifically includes:

[0027] The data sharing device is used to divide the minimum dataset of multiple accounts of multiple users into different storage units, so that the minimum dataset of multiple accounts of the same user is sent to the same storage unit and the amount of data received by different storage units is basically the same.

[0028] In one embodiment, the storage unit is further configured to determine, based on the server's idle time, a server that meets the requirements from among several servers belonging to the local end.

[0029] The storage unit is also used to send the minimum dataset of multiple accounts stored in this device to a server that meets the requirements.

[0030] In one embodiment,

[0031] The server is also used to determine the storage unit to which the local terminal belongs after the local terminal goes online, and to register services with the storage unit.

[0032] The storage unit is also used to determine several servers belonging to this device based on the servers that register services with this device.

[0033] In one embodiment, when the storage unit determines a server that meets the requirements from among several servers belonging to the local end based on the server's idle time, the specific steps include:

[0034] The storage unit is also used to obtain the idle time of several servers belonging to this end;

[0035] The storage unit is also used to sort the servers in ascending order of their free time, and select the servers that are ranked first as the servers that meet the requirements.

[0036] Alternatively, the storage unit may also be used to sort the servers in descending order of their idle time, and select the servers ranked last as the servers that meet the requirements.

[0037] In one embodiment, the server's idle time includes at least one of the server's processor idle time and memory idle time.

[0038] The aforementioned multi-account message processing aggregation system includes an AI processing device, a server, a data sharing device, and a storage unit. The AI ​​processing device receives message datasets pushed by the server of the instant messaging application hosted on this system, targeting the accounts of the instant messaging application. The AI ​​processing device also invokes AI services to process message data that can be automatically processed by AI within the message dataset, obtaining a preliminary message dataset. The server receives the preliminary message dataset, aggregates it, and obtains a session set. Based on its perception of the current view of the instant messaging application's client, the server determines the minimum message data set required for user device interface rendering within the session set. The data sharing device receives the minimum message data sets uploaded by each server, summarizes the minimum message data sets for the same user, obtains the minimum dataset for multiple accounts of the same user, and sends the minimum dataset for multiple accounts to the storage unit. Finally, the server retrieves the minimum dataset for multiple accounts from the storage unit and sends it to the corresponding client, enabling user devices to perform interface rendering based on the minimum dataset for multiple accounts. The aggregation system provided in this application uses AI-assisted filtering, and then leverages view awareness and message data structure optimization to form the minimum message data set required for user device interface rendering. Finally, the minimum message data set of the same user is aggregated and pushed through a data sharing device, reducing the amount of message data that needs to be pushed to the client of the instant messaging application, and achieving stable and efficient push. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating the process involved in an aggregation system for multi-account message processing in one embodiment.

[0041] Figure 2 This is the view currently viewed by the client in one embodiment;

[0042] Figure 3 This is a flowchart illustrating the AI-assisted filtering process in one embodiment;

[0043] Figure 4 This is a flowchart illustrating the view awareness and message data structure optimization process in one embodiment;

[0044] Figure 5 This is a schematic diagram illustrating the process of data aggregation and push from multiple accounts in one embodiment;

[0045] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] In insurance marketing scenarios, during peak business periods, sales personnel in various regions can use an aggregation system to view and reply to customer messages. A single salesperson may have dozens of accounts on instant messaging applications, and each account is associated with hundreds to thousands of customers. In this scenario, the message data push pressure on the aggregation system is extremely high. The multi-account message processing aggregation system provided in this application can stably and efficiently push message data to sales personnel.

[0048] Specifically, refer to Figure 1 The aggregation system provided in this application uses AI (Artificial Intelligence) to assist in filtering, and then uses view perception and message data structure optimization to form the minimum message data set required for user device interface rendering. Finally, the minimum message data set of the same user is aggregated and pushed through the data sharing device, which reduces the amount of message data that needs to be pushed to the client of the instant messaging application and can achieve stable and efficient push.

[0049] in, Figure 1 The users shown can be sales personnel.

[0050] The aggregation system provided in this application includes AI processing equipment, servers, data sharing equipment, and storage units.

[0051] AI processing equipment is used to receive corresponding message datasets pushed by the server of the instant messaging application for accounts hosted on this system.

[0052] AI processing equipment is used to call AI services to process message data that can be automatically processed by AI in the message dataset, and obtain a preliminary message dataset.

[0053] User_1 can have N accounts associated with a certain instant messaging application. User_1 can entrust some or all of these accounts to the aggregation system. For example, if user_1 entrusts all N accounts to the aggregation system, the server of the instant messaging application, after receiving the message data sent to these N accounts, can send the message dataset containing this message data to the AI ​​processing device in the aggregation system.

[0054] Of the message datasets received by the AI ​​processing device, some can be processed automatically by the AI ​​service without user intervention. Therefore, this portion of the message data can be automatically processed by the AI ​​service, reducing the amount of message data that needs to be pushed. Another portion of the message datasets received by the AI ​​processing device is difficult to process automatically by the AI ​​service and requires user intervention. Based on this message data, a preliminary message dataset can be formed. If the amount of message data requiring user intervention is large, the AI ​​processing device can generate multiple preliminary message datasets based on this data and send each preliminary message dataset to a different server. The server can then optimize based on view awareness and message data structure to obtain the minimum set of message data required for interface rendering.

[0055] The server is used to receive the initial screening message dataset, aggregate the initial screening message dataset, and obtain a session set;

[0056] A server is used to determine the minimum set of message data required for user device interface rendering from a set of sessions, based on an awareness of the current view of the instant messaging application's client.

[0057] Let's take server_1 as an example:

[0058] In the initial message dataset received by server_1, each message identifies the sender and receiver. Messages with the same sender and receiver are aggregated together, and the aggregated message data can be recorded as a session. This method optimizes the structure of the message data and allows subsequent data to be processed in session format.

[0059] Server_1 aggregates the initial message dataset to obtain several sessions, which can be called a session set. Then, based on its awareness of the current view of the instant messaging application's client, server_1 determines the minimum set of message data required for the user device to render the interface from the session set.

[0060] Taking user_1 as an example, user_1 includes 10 accounts, all of which are hosted in the aggregation system. When user_1 accesses these 10 accounts simultaneously through the instant messaging application client, the client's current view includes the individual views of each of the 10 accounts. The view of each account can include the chat window between that account and a specific client, as well as the chat list.

[0061] Taking userID_1 as an example, Figure 2As shown, the view of account userID_1 includes a conversation window between user_1 and customer "Xiaoming". The conversation list displays conversations between user_1 and other customers, such as customers "Xiaohong", "Xiaobai", "Xiaolan" and "Xiaowang".

[0062] Based on its awareness of the client's current view, server_1 may be aware of the views of some accounts. For example, server_1 may be aware of the view of userID_1.

[0063] If server_1 perceives the view of userID_1, server_1 can know for sure that: the view of userID_1 includes the session window between user_1 and client "Xiaoming", and the session list included in the view of userID_1 contains the sessions between user_1 and "Xiaohong", "Xiaobai", "Xiaolan" and "Xiaowang".

[0064] Since the view of userID_1 includes the session window between user_1 and customer "Xiaoming", all the message data in the session between user_1 and customer "Xiaoming" is the message data required by user_1's user device for interface rendering. Therefore, server_1 can retain all the message data of the session and include it in the minimum message data set.

[0065] In the conversation between user_1 and customer "Xiao Hong", the latest message is the message data required for interface rendering; therefore, it can be retained and included in the minimum message data set. Similarly, in the conversation between user_1 and customer "Xiao Bai", the latest message is the message data required for interface rendering; therefore, it can be retained and included in the minimum message data set. In the conversation between user_1 and customer "Xiao Lan", the latest message is the message data required for interface rendering; therefore, it can be retained and included in the minimum message data set. Finally, in the conversation between user_1 and customer "Xiao Wang", the latest message is the message data required for interface rendering; therefore, it can be retained and included in the minimum message data set.

[0066] After server_1 becomes aware of the view, it can generate a minimum set of message data. This minimum set of message data includes the minimum message data required for rendering the view for userID_1. The minimum message data required for rendering the views for other accounts of user_1 may be generated by other servers.

[0067] Therefore, after each server generates a minimum message data set, it can upload the minimum message data set to the data sharing device. The server can also associate the minimum message data set with the user identifier to which that minimum message data set belongs before uploading it to the data sharing device.

[0068] The data sharing device is used to receive the minimum message data set uploaded by each server, summarize the minimum message data set of the same user, obtain the minimum dataset of multiple accounts of the same user, and send the minimum dataset of multiple accounts to the storage unit.

[0069] After receiving the minimum message data sets uploaded by each server, the data sharing device can determine which minimum message data sets belong to the same user, since each minimum dataset is associated with a user identifier. By aggregating the minimum message data sets of the same user, the data sharing device can obtain the minimum datasets for multiple accounts of the same user. The data sharing device can then send the minimum datasets for each of the multiple accounts to the storage unit.

[0070] The server is used to obtain the minimum dataset of multiple accounts from the storage unit and send it to the corresponding client so that the user device can render the interface based on the minimum dataset of multiple accounts.

[0071] The aggregation system provided in this application uses AI-assisted filtering, and then leverages view awareness and message data structure optimization to form the minimum message data set required for user device interface rendering. Finally, the minimum message data set of the same user is aggregated and pushed through a data sharing device, reducing the amount of message data that needs to be pushed to the client of the instant messaging application, and can achieve stable and efficient push.

[0072] In one embodiment, when the AI ​​processing device invokes an AI service to process message data in the message dataset that can be automatically processed by AI to obtain a preliminary message dataset, the specific steps include:

[0073] AI processing equipment is used to call AI services and perform semantic recognition on message data in the message dataset.

[0074] AI processing equipment is also used to determine whether a pre-set question-and-answer knowledge base includes matching response content based on semantic recognition results;

[0075] The AI ​​processing device is also used to respond based on the content of the pre-built question-and-answer knowledge base if it includes matching response content.

[0076] The AI ​​processing device is also used to add message data to the initial screening message dataset if the pre-set question-and-answer knowledge base does not contain matching answers.

[0077] In this embodiment, by leveraging AI services and a question-and-answer knowledge base, message data that can be processed automatically by AI without user intervention is automatically replied to, thereby achieving preliminary filtering of the message dataset and forming a preliminary message dataset with reduced message data volume.

[0078] When some instant messaging applications do not provide API calls related to message replies, a PRA device can be invoked to simulate a client and reply. Based on this, this application provides an embodiment in which an AI processing device, when replying based on the response content, specifically includes: an AI processing device further configured to invoke a PRA device to simulate a client and reply with the response content.

[0079] PRA stands for Process Robot Automation, which is a technology that uses software robots to simulate human operations and automate business processes.

[0080] The aggregation system provided in this application uses AI-assisted filtering to perform initial screening of message datasets. Figure 3 A timeline diagram of AI-assisted filtering is shown. (Refer to...) Figure 3 After receiving a message dataset from the server of an instant messaging application, the AI ​​processing device can invoke AI services to perform semantic recognition on the message data in the dataset. Then, based on the semantic recognition results, the AI ​​processing device can determine whether a pre-set question-and-answer knowledge base includes matching response content. If the pre-set question-and-answer knowledge base includes matching response content, the AI ​​processing device can invoke a PRA device to simulate a client and reply with the response content. If the pre-set question-and-answer knowledge base does not include matching response content, the AI ​​processing device can add the message data to the initial screening message dataset.

[0081] In one embodiment, when the server aggregates the initial screening message dataset to obtain a session set, the specific steps include:

[0082] The server is used to aggregate the initially screened message dataset by the sender and receiver of the messages, and obtain a session set.

[0083] Specifically, both the sender and receiver of a message have corresponding accounts in the instant messaging application. Therefore, the initial message dataset can be aggregated based on the dimension of "message sender's account - message receiver's account," grouping together message datasets corresponding to the same "message sender's account - message receiver's account." The aggregated message data is called a session. A session can be the smallest unit for the program to perform allocation, flow, and other processing.

[0084] In one embodiment, when the server determines the minimum set of message data required for user device interface rendering from the session set based on its awareness of the current view of the instant messaging application's client, it specifically includes:

[0085] The server is used to receive operation data sent by user devices to clients of instant messaging applications;

[0086] The server is also used to structure the operational data in order to obtain view information of the view currently being viewed by the client.

[0087] The server is also used to determine the sessions in the session set that match the session identifier based on the session identifier carried in the view information, so as to obtain the minimum set of message data required by the user device for interface rendering.

[0088] Specifically, when a user interacts with an instant messaging application client on their user device, the user device can actively collect interaction data. This data may include the button the user has clicked, the page they are currently viewing, the open chat window, and so on. The user device then sends the collected interaction data to at least one server.

[0089] After receiving the operation data, the server can structure the data, group it into multi-level groups, and obtain the view information of the client's current view. Once the server has obtained the view information, it can upload it to the data sharing device for caching. The view information received by the data sharing device can be integrated, split into multiple parts, and sent to various servers for further processing.

[0090] In insurance marketing scenarios, the specific structure of view information is as follows:

[0091] key:userViewMap

[0092] Field:userNo / / User ID

[0093] value:{

[0094] orgCode:123123 / / User organization code

[0095] groupID:123 / / User group

[0096] view:assign / / The view currently viewed by the client

[0097] userID:123213 / / Account ID for the instant messaging application

[0098] sessionID:123321 / / Session ID, uniquely identifying a session}

[0099] Among these, the user ID serves as the user identifier, uniquely identifying each user within a user organization. The user organization code uniquely identifies the user organization. The user group is used to determine the group to which a user belongs within the user organization.

[0100] Some message data is sent to all users within a user organization; therefore, the view information includes the user organization code. Some message data is sent to all users within a specific group; therefore, the view information includes the user group.

[0101] The server can obtain view information, which reflects the state of the instant messaging application client on the user's device. Therefore, the server can determine the session in the session set that matches the session identifier based on the session identifier carried in the view information. If the session is the currently open session window of the client, all the message data of the session can be included in the minimum message data set required for interface rendering. If the session is a session displayed in the session list, the latest message data in the session can be included in the minimum message data set required for interface rendering.

[0102] The aggregation system provided in this application further filters and aggregates the initial message data set through view awareness and message data structure optimization, further optimizing the data structure. Under the premise of ensuring that the message data is timely and complete, it obtains the minimum message data set required for user device to perform interface rendering, further reducing the amount and frequency of message data to be pushed. Figure 4 The process of view awareness and message data structure optimization is illustrated.

[0103] Reference Figure 4 When a user interacts with an instant messaging application client on their device, the device can actively collect operation data. This data can include the button the user clicked, the page they are currently viewing, the open chat window, and so on. The user device then sends the collected operation data to one of the servers.

[0104] After receiving the operation data, the server can structure the data, group it into multi-level groups, and assemble the view information of the client's current view. Once the server receives the view information, it can upload it to the data sharing device for caching. The view information received by the data sharing device can then be split into multiple parts and sent to various servers for further processing.

[0105] The server aggregates the initial message dataset obtained from AI-assisted filtering to obtain a session set.

[0106] The server receives view information, which reflects the state of the instant messaging application client on the user's device. Therefore, the server can determine the session in the session set that matches the session identifier based on the session identifier carried in the view information. If the session is the currently open session window of the client, all message data of the session can be included in the minimum message data set required for interface rendering. If the session is a session displayed in the session list, the latest message data in the session can be included in the minimum message data set required for interface rendering.

[0107] In one embodiment, when the data sharing device sends a minimum dataset of multiple accounts to the storage unit, it specifically includes:

[0108] Data sharing devices are used to divide the minimum datasets of multiple users' accounts into different storage units, so that the minimum datasets of multiple users' accounts are sent to the same storage unit and the amount of data received by different storage units is basically the same.

[0109] The minimum set of message data uploaded by multiple servers can be uploaded to a cache stream on a data-sharing device, which is shared by all servers. Using a cache stream structure allows for lightweight aggregation of the minimum set of message data for the same user, thereby achieving the effect of simultaneous push of message data from multiple accounts and facilitating subsequent unified scheduling.

[0110] After obtaining the cache stream containing the minimum datasets for each user's multiple accounts, the data sharing device can structurally partition the cache stream to obtain multiple storage units, each containing the minimum datasets for multiple accounts. The partitioning principle prioritizes grouping the minimum datasets for the same user's multiple accounts into the same storage unit, while secondaryly, different storage units receiving data of roughly the same size.

[0111] The minimum datasets of multiple accounts in different storage units can be processed separately, thereby increasing the parallelism of processing and improving the ability to push message data.

[0112] The aforementioned cache stream can be a stream, which is a data structure primarily used in real-time data processing scenarios. When the cache stream is a stream, logically decomposing the stream structure yields storage units called segments.

[0113] In one embodiment, the storage unit is further configured to determine, based on the server's idle time, a server that meets the requirements from among several servers belonging to the local end; the storage unit is further configured to send the minimum dataset of multiple accounts stored in the device to the server that meets the requirements.

[0114] The server's idle time includes at least one of the server's processor idle time and memory idle time. The processor can be a CPU (central processing unit).

[0115] Each storage unit contains several servers. Among these servers, the storage unit can determine the server with the most idle processor and memory based on their processor and memory availability, and select that server as the one that meets the requirements. The storage unit can then trigger the server that meets the requirements to read the stored minimum dataset for multiple accounts.

[0116] After obtaining the minimum dataset for multiple accounts, the server can establish a long-lived WebSocket connection with the client containing the minimum dataset and send the dataset to the client. The user device on the client can then render the interface based on this minimum dataset. WebSocket is a protocol for full-duplex communication over a single TCP connection.

[0117] In one embodiment, the server is further configured to, after the local end goes online, determine the storage unit to which the local end belongs and register the service with the storage unit; the storage unit is further configured to, based on the servers that have registered the service with the device, determine a number of servers belonging to the local end.

[0118] Once any server comes online, a hash algorithm can be used to determine its associated storage unit, and the server can then register its services with that storage unit. This allows the storage unit to determine which servers belong to its service provider based on the server's service registration.

[0119] In one embodiment, when the storage unit determines a server that meets the requirements from among several servers belonging to the local end based on the server's idle time, the specific steps include:

[0120] The storage unit is also used to obtain the idle time of several servers belonging to this end;

[0121] The storage unit is also used to sort the servers in ascending order of their free time, and select the server that ranks first as the server that meets the requirements.

[0122] Alternatively, the storage unit can also be used to sort the servers in descending order of their free time, and select the servers ranked last as the servers that meet the requirements.

[0123] For example, after sorting the servers according to their idle time from smallest to largest, the top N servers can be selected as the servers that meet the requirements. N is a positive integer greater than or equal to 1.

[0124] As another example, after sorting the servers according to their idle time from highest to lowest, the last M servers can be selected as the servers that meet the requirements. M is a positive integer greater than or equal to 1.

[0125] The aggregation system provided in this application summarizes the minimum set of message data for the same user through a data sharing device, thereby aggregating data from multiple accounts of the same user. During push notifications, unified load scheduling and batch pushes are implemented to achieve stable and efficient push functionality. Figure 5 The process of data sharing and distribution push is shown.

[0126] Reference Figure 5 An aggregation system can deploy multiple servers, such as Figure 5 The diagram shows servers 1, 2, and 3. After any server generates a minimum message data set, it can upload this minimum message data set to the stream of the data sharing device. The stream is shared by all servers. Using the stream structure allows for lightweight aggregation of the minimum message data set of the same user, achieving the effect of simultaneous push of message data from multiple accounts and facilitating subsequent unified scheduling.

[0127] Data sharing devices can structurally divide a stream into multiple segments, such as... Figure 5 The diagram shows segment1 and segment2. The partitioning principle is primarily to group the smallest datasets from multiple accounts of the same user into the same segment, with the secondary principle being that different segments receive roughly the same amount of data.

[0128] The minimum datasets of multiple accounts in different segments can be processed separately, thereby increasing the parallelism of processing and improving the ability to push message data.

[0129] Once any server comes online, a hash algorithm can be used to determine its corresponding segment, and the server can then register its services with that segment. Therefore, the segment can determine which servers belong to its network based on the servers where services are registered.

[0130] A segment can identify a server with relatively idle CPU and memory among several servers belonging to the local machine, and select that server as the one that meets the requirements. A segment can then trigger the server that meets the requirements to read the stored minimum dataset of multiple accounts.

[0131] After the server obtains the minimum dataset of multiple accounts from the segment, it can establish a long-lived WebSocket connection with the client containing the minimum dataset of multiple accounts for data communication. During each data communication session, the minimum dataset of multiple accounts processed in that batch can be aggregated and sent to the client in bulk, effectively reducing the push frequency under massive data volumes. Simultaneously, leveraging the data storage capabilities of the stream structure, failed pushes can be resent, ensuring the stability of data pushes.

[0132] The user device on which the client is located can render the interface based on a minimal dataset of multiple accounts.

[0133] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0134] In one exemplary embodiment, the aggregation system provided in this application can be implemented by a computer device. The internal structure diagram of the computer device can be as follows: Figure 6 As shown, the computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The database stores the data involved in the above embodiments. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps involved in an aggregation system.

[0135] Those skilled in the art will understand that Figure 6The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0136] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0137] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes described in the above embodiments. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0138] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0139] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A multi-account message processing aggregation system, characterized in that, The aggregation system includes AI processing equipment, servers, data sharing equipment, and storage units; The AI ​​processing device is used to receive the corresponding message dataset pushed by the server of the instant messaging application for the account hosted on this system. The AI ​​processing device is used to call AI services to process message data in the message dataset that can be automatically processed by AI, and obtain a preliminary message dataset; The server is used to receive the initial screening message dataset, aggregate the initial screening message dataset, and obtain a session set; The server is configured to determine, based on its perception of the current view of the client of the instant messaging application, the minimum set of message data required for the user device to render the interface from the set of sessions. The data sharing device is used to receive the minimum message data set uploaded by each of the servers, and to aggregate the minimum message data set of the same user to obtain the minimum dataset of multiple accounts of the same user, and to send the minimum dataset of multiple accounts to the storage unit. The server is used to obtain the minimum dataset of multiple accounts from the storage unit and send it to the corresponding client so that the user device can perform interface rendering based on the minimum dataset of multiple accounts.

2. The system according to claim 1, characterized in that, When the AI ​​processing device invokes AI services to process message data in the message dataset that can be automatically processed by AI to obtain a preliminary message dataset, the specific steps include: The AI ​​processing device is used to call AI services to perform semantic recognition on message data in the message dataset; The AI ​​processing device is also used to determine, based on the semantic recognition results, whether the preset question-and-answer knowledge base includes matching response content; The AI ​​processing device is also used to respond based on the response content if the preset question-and-answer knowledge base includes matching response content; The AI ​​processing device is also used to add the message data to the initial screening message dataset if the preset question-and-answer knowledge base does not contain matching response content.

3. The system according to claim 2, characterized in that, When the AI ​​processing device responds based on the response content, it specifically includes: The AI ​​processing device is also used to invoke the PRA device to simulate a client and reply with the response content.

4. The system according to claim 1, characterized in that, When the server aggregates the initial screening message dataset to obtain a session set, the specific steps include: The server is used to aggregate the initial message dataset based on the sender and receiver of the messages to obtain a session set.

5. The system according to claim 1, characterized in that, When the server determines the minimum set of message data required for user device interface rendering from the session set based on its perception of the current view of the instant messaging application's client, it specifically includes: The server is used to receive operation data sent by the user device for the client of the instant messaging application; The server is also used to structure the operation data to obtain view information of the view currently in which the client is located; The server is further configured to determine, based on the session identifier carried in the view information, the session in the session set that matches the session identifier, so as to obtain the minimum message data set required by the user device for interface rendering.

6. The system according to claim 1, characterized in that, When the data sharing device sends the minimum dataset of multiple accounts to the storage unit, it specifically includes: The data sharing device is used to divide the minimum dataset of multiple accounts of multiple users into different storage units, so that the minimum dataset of multiple accounts of the same user is sent to the same storage unit and the amount of data received by different storage units is basically the same.

7. The system according to claim 1, characterized in that, The storage unit is also used to determine, based on the server's idle time, which of the several servers belonging to this end meets the requirements; The storage unit is also used to send the minimum dataset of multiple accounts stored in this device to a server that meets the requirements.

8. The system according to claim 7, characterized in that, The server is also used to determine the storage unit to which the local terminal belongs after the local terminal goes online, and to register services with the storage unit. The storage unit is also used to determine several servers belonging to this device based on the servers that register services with this device.

9. The system according to claim 7, characterized in that, When the storage unit determines a server that meets the requirements from among several servers belonging to this end based on the server's idle time, the specific steps include: The storage unit is also used to obtain the idle time of several servers belonging to this end; The storage unit is also used to sort the servers in ascending order of their free time, and select the servers that are ranked first as the servers that meet the requirements. Alternatively, the storage unit may also be used to sort the servers in descending order of their idle time, and select the servers ranked last as the servers that meet the requirements.

10. The system according to claim 7, characterized in that, The server's idle time includes at least one of the server's processor idle time and memory idle time.

Images