Task decision-making method and device, computer equipment and readable storage medium

Abstract

The invention relates to a task decision method and device, computer equipment and a readable storage medium. The method comprises the steps of displaying a session page of a target session; in response to a task decision event for the target task in the session page, displaying target recommendation content of the agent account for the target task in the session page; wherein the target recommendation content is generated based on event interaction data of at least two non-agent accounts participating in the target session for the target task. By adopting the method, the task decision-making process in the group chat scene in the group can be simplified, the decision-making efficiency is improved, and the organization efficiency of member group activities is further improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet, in particular to a task decision method and device, computer equipment and readable storage medium. BACKGROUND

[0002] With the development of Internet technology, more and more users choose to make task decisions through group chats in groups in social software, such as communicating about restaurants for a dinner party or tourist spots.

[0003] However, in the conventional technology, in the group chat scenario in the group, the task decision usually depends on the free discussion of the members in the group, and the members often need to spend time communicating needs such as preferences, budgets, distances, and times, and cannot effectively integrate and analyze these needs, which causes problems such as scattered opinions and difficulty in quickly reaching a consensus, and thus leads to a cumbersome task decision process, low decision efficiency, and low organization efficiency of the member group activity. SUMMARY

[0004] Therefore, it is necessary to provide a task decision method, device, computer equipment and readable storage medium to simplify the task decision process in the group chat scenario in the group, improve the decision efficiency, and thus improve the organization efficiency of the member group activity.

[0005] In a first aspect, the present application provides a task decision method, comprising:

[0006] displaying a session page of a target session;

[0007] in response to a task decision event for a target task in the session page, displaying target recommendation content of an intelligent agent account for the target task in the session page; wherein the target recommendation content is generated based on event interaction data of at least two non-intelligent agent accounts participating in the target session for the target task.

[0008] In one of the embodiments, the task decision event includes at least one of the following:

[0009] in response to a member adding operation for the target session, determining target historical session content in the target session and adding the intelligent agent account to the target session;

[0010] sending event session content in the target session; wherein the event session content includes target characters, account information of the intelligent agent account, and task information of the target task;

[0011] collecting event interaction data of completing the target task;

[0012] The conversation content for the target task includes at least two pieces of conversation content having an opposite relationship.

[0013] In one embodiment, in response to a task decision event for the target task in the conversation page, the target recommendation content for the target task of the intelligent agent account is displayed in the conversation page, including:

[0014] In response to a task decision event for the target task in the conversation page, event interaction data of at least two non-intelligent agent accounts for the target task is extracted from the conversation content related to the target task in the conversation page; or, in response to a task decision event for the target task in the conversation page, an event interaction data collection page for the target task is displayed; wherein the event interaction data collection page is used to collect event interaction data of at least two non-intelligent agent accounts for the target task.

[0015] The target recommendation content generated by the intelligent agent account based on the event interaction data is displayed in the conversation page.

[0016] In one embodiment, the event interaction item in the event interaction data collection page is determined according to the conversation content related to the target task in the conversation page, and / or the content attribute data of the at least one candidate recommendation content corresponding to the target task.

[0017] In one embodiment, in response to a task decision event for the target task in the conversation page, the event interaction data collection page for the target task is displayed, including:

[0018] In response to a task decision event for the target task in the conversation page, an access card for the target task is displayed in the conversation page; wherein the access card includes a page access entry of the event interaction data collection page; in response to a triggering operation on the page access entry, the event interaction data collection page for the target task is displayed.

[0019] In one embodiment, the target recommendation content generated by the intelligent agent account based on the event interaction data of at least one non-intelligent agent account is displayed in the conversation page, including:

[0020] In response to a triggering operation on the content viewing entry in the access card, a content display page is displayed in the conversation page; or, in response to the end of data collection of the event interaction data collection page, the page access entry in the access card is switched to the content viewing entry, and in response to a triggering operation on the content viewing entry, a content display page is displayed in the conversation page; wherein the content display page includes the target recommendation content.

[0021] In one embodiment, the end of data collection of the event interaction data collection page includes at least one of the following:

[0022] reaching a data collection deadline;

[0023] a preset number of non-agent account page fillings are completed;

[0024] a target task corresponding account circle range non-agent account page filling is completed.

[0025] In one embodiment, the target recommendation content of the agent account for the target task is displayed, including at least one of the following:

[0026] Display text content; wherein the text content includes at least one content recommendation result and / or at least one content recommendation result corresponding to the recommendation reference data;

[0027] Display decision recommendation card; wherein the decision recommendation card includes at least two candidate recommendation items for the target task; the candidate recommendation item corresponds to the recommended content for the target task;

[0028] Display decision card; wherein the decision card displays the selection results of at least two candidate recommendation contents by different non-agent accounts.

[0029] In one embodiment, the task decision method further comprises:

[0030] In the case that there is a decision preference dissenting account in the account participating in the target conversation, the decision preference reason of the decision preference dissenting account is collected to update the event interaction data for the target task according to the decision preference reason.

[0031] In one embodiment, collecting the decision preference reason of the decision preference dissenting account includes at least one of the following:

[0032] In the conversation page of the target group corresponding to the decision preference dissenting account, a reason collection page is displayed, and the decision preference reason of the decision preference dissenting account is obtained in response to the editing completion of the decision preference dissenting account in the reason collection page;

[0033] In the independent conversation page corresponding to the decision preference dissenting account and the agent account, the reason inquiry content is displayed, and the decision preference reason of the decision preference dissenting account is obtained according to the reply of the decision preference dissenting account to the reason inquiry content.

[0034] In a second aspect, the present application also provides a task decision method, comprising:

[0035] Display the session page of the target session; in response to a task decision event for the location recommendation task in the session page, display the target recommended location for the location recommendation task of the agent account in the session page; wherein the target recommended location is generated based on the event interaction data of at least two non-agent accounts participating in the target session for the location recommendation task.

[0036] Thirdly, this application also provides a task decision-making method, including:

[0037] Display the session page of the target session; in response to a task decision event for the route recommendation task in the session page, display the target recommended route for the route recommendation task for the agent account in the session page; wherein the target recommended route is generated based on event interaction data of at least two non-agent accounts participating in the target session for the route recommendation task.

[0038] Fourthly, this application also provides a task decision-making device, comprising:

[0039] The first display module is used to display the session page of the target session;

[0040] The second display module is used to respond to a task decision event for the target task in the conversation page and display the target recommended content of the agent account for the target task in the conversation page; wherein, the target recommended content is generated based on the event interaction data of at least two non-agent accounts participating in the target conversation for the target task.

[0041] Fifthly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the various method embodiments provided in the first, second and / or third aspects described above.

[0042] Sixthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method embodiments provided in the first, second, and / or third aspects described above.

[0043] In a seventh aspect, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method embodiments provided in the first, second, and / or third aspects described above.

[0044] The aforementioned task decision-making method, apparatus, computer device, and readable storage medium, in a target session scenario involving at least two non-agent accounts, can display the session page of the target session. Furthermore, when a non-agent account makes a task decision for the target task, generating a task decision event for the target task, an agent account can be added to the participating accounts of the target session. In response to the task decision event for the target task in the session page, the agent account displays the target recommended content for the target task in the session page. The target recommended content is generated based on the event interaction data of the at least two non-agent accounts participating in the target session for the target task. In this way, on the one hand, when non-intelligent agent accounts make task decisions, intelligent agent accounts can quickly and effectively integrate and analyze the event interaction data of non-intelligent agent accounts regarding the target task without requiring them to spend a lot of time communicating. This allows for the provision of recommended content for the target task, simplifying the task decision-making process, improving decision-making efficiency, and thus improving the organizational efficiency of group activities. On the other hand, generating target recommended content based on the event interaction data of non-intelligent agent accounts regarding the target task enables precise analysis of the task decision-making needs of non-intelligent agent accounts, improving the accuracy and relevance of the generated target recommended content, and enhancing the user experience. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is an application environment diagram of the task decision-making method provided in some embodiments of the present invention;

[0047] Figure 2A A flowchart illustrating a task decision-making method provided in some embodiments of the present invention;

[0048] Figure 2B This is a schematic diagram illustrating the display of an agent account in a conversation page provided in some embodiments of the present invention;

[0049] Figure 2C This is a schematic diagram illustrating the display of target recommended content in a conversation page provided in some embodiments of the present invention;

[0050] Figure 2D This is a schematic diagram illustrating the display of an agent account in a dialog box on a conversation page, as provided in some embodiments of the present invention.

[0051] Figure 2E This is a schematic diagram illustrating the addition of an agent account to a target session, provided by some embodiments of the present invention.

[0052] Figure 2F This is a schematic diagram illustrating the addition of an agent account to a target session according to some embodiments of the present invention;

[0053] Figure 2G This is a display diagram illustrating the addition of an agent account to a target session, provided for other embodiments of the present invention.

[0054] Figure 2H A schematic diagram illustrating the collection of event interaction data provided in some embodiments of the present invention;

[0055] Figure 2I This is a schematic diagram illustrating the display of voting prompt information provided in some embodiments of the present invention;

[0056] Figure 2J A flowchart illustrating a task decision-making method provided in other embodiments of the present invention;

[0057] Figure 3A A flowchart illustrating the display of target recommended content provided in some embodiments of the present invention;

[0058] Figure 3B This is a schematic diagram illustrating the display of target recommended content in a conversation page provided in some embodiments of the present invention;

[0059] Figure 3C This is a schematic diagram illustrating the display of target recommended content in a conversation page provided in other embodiments of the present invention;

[0060] Figure 4A A flowchart illustrating the display of target recommended content provided for other embodiments of the present invention;

[0061] Figure 4B This is a schematic diagram illustrating the display of access cards in a session page according to some embodiments of the present invention;

[0062] Figure 4C A schematic diagram showing a decision data collection page provided for other embodiments of the present invention;

[0063] Figure 4D This is a schematic diagram illustrating the display of access cards in a session page according to other embodiments of the present invention;

[0064] Figure 4E This is a schematic diagram illustrating the display of content in a session page according to some embodiments of the present invention;

[0065] Figure 4F This is a schematic diagram showing the access card in a session page according to some embodiments of the present invention;

[0066] Figure 4G A flowchart illustrating the display of target recommended content provided for some embodiments of the present invention;

[0067] Figure 5 A schematic diagram illustrating the display of an independent session page provided in some embodiments of the present invention;

[0068] Figure 6 A flowchart illustrating a task decision-making method provided in some embodiments of the present invention;

[0069] Figure 7 A flowchart illustrating a task decision-making method provided in some embodiments of the present invention;

[0070] Figure 8A This is a flowchart illustrating a group creation method provided in some embodiments of the present invention;

[0071] Figure 8B This is a schematic diagram illustrating the display operation of a message browsing page provided in some embodiments of the present invention;

[0072] Figure 8C A schematic diagram illustrating the display operation of a message browsing page provided in other embodiments of the present invention;

[0073] Figure 8D A schematic diagram illustrating the display operation of a first session page provided in some embodiments of the present invention;

[0074] Figure 8E A schematic diagram illustrating the display operation of a second session page provided in some embodiments of the present invention;

[0075] Figure 8F A schematic diagram illustrating the process of adding an agent account according to some embodiments of the present invention;

[0076] Figure 8G A schematic diagram illustrating the operation of adding an agent account according to other embodiments of the present invention;

[0077] Figure 8H A schematic diagram illustrating the operation of adding an agent account as provided in some embodiments of the present invention;

[0078] Figure 8I This is a schematic diagram illustrating the operation of adding an agent account in some embodiments of the present invention;

[0079] Figure 8J A schematic diagram illustrating the mention operation of an agent account provided in some embodiments of the present invention;

[0080] Figure 8K This is a schematic diagram illustrating group sharing operations provided in some embodiments of the present invention;

[0081] Figure 8L A timing diagram of a group creation method provided in some embodiments of the present invention;

[0082] Figure 9A A flowchart illustrating the steps for creating a target group as provided in some embodiments of the present invention;

[0083] Figure 9B This is a schematic diagram illustrating the creation process of a target group as provided in some embodiments of the present invention;

[0084] Figure 9C A schematic diagram illustrating the creation process of a target group as provided in other embodiments of the present invention;

[0085] Figure 9D This is a schematic diagram illustrating the role editing operation of an intelligent agent provided in some embodiments of the present invention;

[0086] Figure 9E This is a schematic diagram illustrating the display operation of a target group provided in some embodiments of the present invention;

[0087] Figure 9F A schematic diagram illustrating the creation process of a target group as provided in some embodiments of the present invention;

[0088] Figure 9G A timing diagram illustrating the steps for creating a target group as provided in some embodiments of the present invention;

[0089] Figure 10A A flowchart illustrating the steps for creating a target group as provided in other embodiments of the present invention;

[0090] Figure 10B This is a schematic diagram illustrating the creation process of a target group as provided in some embodiments of the present invention;

[0091] Figure 10C This is a schematic diagram illustrating the creation process of a target group in some embodiments of the present invention;

[0092] Figure 10D This is a schematic diagram illustrating the guided operation of the role attribute guidance control provided in some embodiments of the present invention;

[0093] Figure 10E A schematic diagram illustrating the guided operation for adding a guide control to an account, as provided in some embodiments of the present invention;

[0094] Figure 10F A timing diagram illustrating the steps for creating a target group as provided in other embodiments of the present invention;

[0095] Figure 11 This is a structural block diagram of a task decision-making device provided in some embodiments of the present invention;

[0096] Figure 12 This is an internal structural diagram of a computer device provided for some embodiments of the present invention. Detailed Implementation

[0097] 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.

[0098] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments or any combination of multiple embodiments.

[0099] Furthermore, before describing the embodiments of this application, it should be noted that the term "in response to" as used herein refers to a state in which a corresponding event occurs or a condition is met. It is understood that the timing of the execution of a subsequent action performed in response to such an event or condition is not necessarily strongly correlated with the time when the event occurs or the condition is met. For example, in some cases, the subsequent action may be executed immediately upon the occurrence of the event or the fulfillment of the condition; while in other cases, the subsequent action may be executed some time after the occurrence of the event or the fulfillment of the condition. "Triggering operation" refers to an action performed by a user on a visual interface through clicking, dragging, swiping, or other interactive methods, which aims to initiate or activate a specific patent data display, analysis, or editing function. "Published content" refers to content pre-published by a user on a content platform, including text and image content, video content, live content, streaming content, linked content, etc.

[0100] In traditional technologies, task decisions in group chat scenarios typically rely on free discussion among group members. Members often need to spend time exchanging information about preferences, budgets, distance, time, and other requirements, and they cannot effectively integrate and analyze these requirements. This results in problems such as scattered opinions and difficulty in quickly reaching a consensus, leading to a cumbersome task decision-making process, low decision-making efficiency, and affecting the organizational efficiency of group activities.

[0101] In view of this, in order to solve the above-mentioned technical problems, this application provides a task decision method.

[0102] In one exemplary embodiment, the task decision-making method provided in this application can be applied to, for example, Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or located on a cloud or other network server. Terminal 102 can be a user's terminal that logs into the content platform based on an account, and server 104 can be the server of the corresponding content platform. Furthermore, server 104 can manage and store content publishing information related to the content published on the social platform. This content publishing information can include at least one of text content, media content, hashtags, and content titles. Different accounts logged into the content platform can publish content publishing information on a chat page to achieve conversations between different accounts. Therefore, server 104 can also manage and store the content publishing information published by accounts on the chat page, such as text information, image information, and link information on the chat page. Furthermore, accounts logged into the content platform can create closed or semi-closed virtual community spaces, i.e., create groups, to gather different accounts logged into the content platform together. Server 104 can then manage and store content posted by these accounts within their respective groups, such as text and image information. Terminal 102 can have an app installed, which can be a social media app, video app, or shopping app, etc., and this method can be implemented by the app. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0103] Based on the above embodiments, in an exemplary embodiment, this task decision method is applied to... Figure 1 Let's take terminal 102 as an example for illustration. Figure 2A As shown, it includes the following steps:

[0104] S201A, Display the session page of the target session.

[0105] As mentioned earlier, accounts logged into the content platform can publish conversation content to interact with other accounts, and the terminal can then display the conversation page of the target conversation. The conversation page can be understood as a specific interface area within the content platform that allows different accounts to exchange information, share content, and interact with each other.

[0106] Optionally, the accounts participating in the target session may include at least two non-agent accounts; or may include an agent account and at least two non-agent accounts. For example, such as Figure 2BAs shown, this is a conversation page, where Account 1-2 and Account 1-3 are non-smart agent accounts participating in the conversation, and Account 1-1 is the smart agent account "ai meal selection assistant" participating in the conversation.

[0107] The so-called intelligent agent account is an account based on artificial intelligence technology that can automatically interact with group members, provide services, or perform specific tasks within a group. It is a virtual digital entity with autonomous perception, decision-making, and execution capabilities, rather than a real user, and participates in group interactions in a human-like manner. For example, an intelligent agent account can automatically converse with other members in the group, initiate polls or surveys, etc. Optionally, an intelligent agent account can be a virtual object account generated by artificial intelligence models such as large language models.

[0108] Correspondingly, a non-intelligent agent account is an account used by users to log in to a content platform. It lacks autonomous perception, decision-making, and execution capabilities, requiring manual user intervention for information interaction and task execution. It cannot autonomously perform information interaction or task execution; for example, a user logging into a content platform using a non-intelligent agent account manually posts information in a group or edits information on their personal profile. For ease of description, the following will refer to the act of a user logging into a content platform using a non-intelligent agent account to post conversation content in a group as "non-intelligent agent account posting conversation content."

[0109] Optionally, when different accounts participating in a target session are having a session within a shared target group, the displayed session page for the target session will be the session page for the target group. For example... Figure 2B The "Group X" shown indicates that accounts 1-1, 1-2, and 1-3 are having a conversation in a common target group, "Group X". Figure 2B The session page shown is the session page for the target group. Optionally, the participating accounts in the target group include at least two non-agent accounts; or include an agent account and at least two non-agent accounts.

[0110] A group refers to a collection of users created on a content platform based on shared interests, goals, or relationships, who exchange information, share content, and interact within a semi-closed virtual space provided by the platform. Optionally, a target group is a group pre-created before decisions are made regarding a target task, or a group created specifically for making decisions about a target task. No specific limitations are placed on the timing or method of creating target groups. Examples of target group creation will be provided later.

[0111] S202A, in response to a task decision event for a target task in the conversation page, displays the agent account's target recommendation content for the target task in the conversation page.

[0112] The target recommended content is generated based on event interaction data of at least two non-agent accounts participating in the target session regarding the target task.

[0113] The so-called target task refers to the decision-making task discussed by non-agent accounts within the conversation page, such as restaurant recommendation, travel destination recommendation, or route recommendation. Correspondingly, the so-called task decision event refers to the event related to the target task decision, such as the discussion content about the target task posted by non-agent accounts on the conversation page, or non-agent accounts mentioning intelligent agent accounts on the conversation page.

[0114] The so-called event interaction data of non-intelligent agent accounts in relation to the target task refers to the data of operation events related to the target task in which non-intelligent agent accounts participate. For example, the conversation content about the target task published on the conversation page, the response data to the voting task or questionnaire task initiated on the conversation page for the target task, the attention data of the content information published on the content platform related to the target task, and the purchase data of the goods related to the target task on the content platform, etc. There are no specific limitations on this.

[0115] For a specific task, target recommended content can be generated based on event interaction data from at least two non-agent accounts participating in the target session. Thus, when the terminal displays the session page for the target group, it can respond to task decision events related to the target task within the session page and display the target recommended content from the agent accounts. In other words, the target recommended content on the session page is published by the agent accounts. For example, such as... Figure 2C As shown, the target recommended content 2-1 was published by the intelligent agent account "ai meal selection assistant" 1-1.

[0116] The target recommendation content displayed on the terminal can vary depending on the specific task. For example, if the target task is restaurant recommendation, the content will include information about at least one restaurant, such as location, price range, and cuisine. If the target task is tourist destination recommendation, the content will include information about at least one tourist destination, such as weather, attractions, location, and local specialties. If the target task is exercise route recommendation, the content will include information about at least one exercise route, such as traffic congestion level, route length, and landmarks along the way. Of course, the target tasks can also be other types of tasks, such as entertainment recommendations or movie / TV show recommendations, in which case the target recommendation content will correspond to that specific task.

[0117] Optionally, the target recommended content includes multiple pieces of content arranged in descending order of their relevance to the target task. The number of these multiple pieces of content can be a preset number, or it can be the number of pieces with a relevance greater than a preset relevance threshold. Both the preset number and preset relevance threshold can be set based on experience and the needs of actual applications; no specific limitations are imposed on these settings.

[0118] Optionally, when a portion of the target recommended content is displayed on the session page, in response to an upward swipe operation on the target recommended content, or in response to a downward scrolling operation on the decision card containing the target recommended content, the currently displayed target recommended content is collapsed to display the target recommended content that is not currently displayed. Optionally, in response to the display entry in the decision card marked with the text "Show More" below the last currently displayed target recommended content, a content retrieval request is sent to the server; after receiving the content retrieval request, the server sends other target recommended content that is not currently displayed to the terminal; after receiving the other target recommended content, the terminal continues to display the aforementioned other target recommended content in the decision card until it receives a message from the server indicating that there is no more recommended content.

[0119] Optionally, in response to a task decision event for the target task in the conversation page, the terminal sends a task decision request to the server of the content platform to which the target group belongs (hereinafter referred to as the server). After receiving the task decision request, the server obtains event interaction data of at least two non-agent accounts for the target task, and generates target recommended content for the target task based on the event interaction data. The server sends the target recommended content as publishing information of the agent account to the terminal. After receiving the target recommended content, the terminal publishes the target recommended content in the conversation page through the agent account. Optionally, the event interaction data obtained by the server includes historical conversation content of at least two non-agent accounts for the target task, such as conversation content within a preset duration range before the current time, or a preset number of conversation contents before the current time; or, the event interaction data obtained by the server includes all conversation content in the task decision events of at least two non-agent accounts for the target task. The preset duration range and preset number can be set according to experience values, application requirements in actual applications, etc., and are not specifically limited thereto. Optionally, the task decision request may include the event interaction data and / or the account information of the non-agent account participating in the target session, so that the server can obtain the event interaction data from the data storage system based on the account information.

[0120] Optionally, task decision events include: within a preset duration prior to the current moment, the number of conversations related to the target task in the conversation content of the conversation page reaches a preset quantity. Both the preset duration and preset quantity can be set based on experience and application requirements in actual use, and are not specifically limited thereto.

[0121] In an optional embodiment A1, the task decision event includes: in response to a member addition operation for a target session, determining the target historical session content in the target session, and adding the agent account to the target session.

[0122] In this embodiment A1, during the display of the conversation page, the target conversation is a conversation between at least two non-agent accounts regarding the target task. Non-agent account users can add agent accounts to the target conversation to incorporate the agent accounts. Therefore, in response to the member addition operation for the target conversation, the terminal can determine the target historical conversation content within the target conversation and add the agent account to the target conversation. Furthermore, in response to the determination of the target historical conversation content and the addition of the agent account to the target conversation, the terminal displays the target recommended content on the conversation page by having the agent account publish a message.

[0123] Optionally, adding members to an agent's account can include posting "characters that mention the agent's account and the agent's account information" on the conversation page. For example, such as... Figure 2D As shown, "AI Chat Assistant" is the AI ​​agent account. When you enter "@AI Chat Assistant" in the dialog box, a prompt message "Let the AI ​​join the conversation" will appear; then... Figure 2E As shown, the non-AI agent account enters conversation content containing "@AI Chat Assistant" in the dialog box, and as follows... Figure 2F As shown, the above-mentioned conversation content containing "@AI Chat Assistant" is published on the conversation page to add "AI Chat Assistant" to the target conversation, where, as... Figure 2F As shown, after the agent account is added to the target session, the target group name 3-1 displayed on the session page will be followed by the smart identifier 3-2 to indicate that the agent account has been added to the target session.

[0124] Optionally, the member addition operation for an agent account can be a selection operation for the agent account within the member selection page. For example, a non-agent account can open the member selection page as follows: Figure 2G On the member selection page shown, select “AI Chat Assistant” and click the “Complete (1)” component in the lower right corner to add “AI Chat Assistant” to the target session.

[0125] In embodiment A1, optionally, the event interaction data used to generate the target recommendation content includes the target session content in the session page. For example, historical session content within a preset duration range or a preset number of entries before the current time; or, all session content of at least two non-agent accounts for the target task, etc. Optionally, in response to the selection operation of the target session content in the session page, the terminal sends the selected target session content to the server; after receiving the target session content, the server performs semantic analysis and other processing on the target session content to determine the session data related to the target task; the server generates a target prompt word based on the above session data and a preset prompt word, and inputs the target prompt word to the target model corresponding to the agent account deployed on the server to obtain the target recommendation content output by the target model; the server sends the target recommendation content to the terminal so that the target recommendation content is displayed in the terminal's session page. Optionally, after receiving the target session content, the server can call a large language model to optimize and normalize the prompt words in the target session content to obtain task prompt words as session data related to the target task. This session data is then embedded into a preset prompt, and the preset prompt with embedded session data is input into the target model. Optionally, the target model can be an artificial intelligence model such as a multimodal large language model, a graph neural network recommendation model, or an expert hybrid model, or it can be a deep neural network model, a sequence recommendation model, or an information retrieval recommendation model that supports natural speech prompts. Optionally, the server can also further process its output target recommendation content to add the processing results of each recommendation to the target recommendation content. For example, the server can determine and rank the matching degree between the target recommendation content and the target task to add the matching degree of each recommendation to the target recommendation content, and display each recommendation in the target recommendation content in the terminal in descending order of matching degree.

[0126] In an optional embodiment A2, the task decision event includes: sending event session content in the target session; wherein the event session content includes target characters, account information of the agent account, and task information of the target task.

[0127] In embodiment A2, during the display of the conversation page, a non-agent account can send event conversation content in the target conversation to mention the agent account. Consequently, the terminal responds by sending event conversation content in the target conversation and displays the target recommended content on the conversation page as if the agent account had published a message. Optionally, the agent account's account information can be the agent account's username. Optionally, the target character can be a mention character, such as "@".

[0128] Optionally, the session page is the session page of the target group to which at least two non-agent accounts participating in the target session belong. If the target account already exists within the target group, the non-agent accounts can send event session content in the target session to enable the agent account to participate in the target session; for example, such as... Figure 2B As shown, non-agent accounts 1-3 publish the event session content "Use the meal selection assistant, let the AI ​​help you choose, @AI meal selection assistant" on the session page displayed on the terminal. Here, "Use the meal selection assistant, let the AI ​​help you choose" is the task information of the target task, "@" is the target character, and "AI meal selection assistant" is the account information of the intelligent agent account. Correspondingly, if the target account does not exist among the participating accounts in the target group, the non-agent account can add the intelligent agent account to the target group and the target session by sending event session content in the target session; for example, as... Figure 2F As shown, the non-intelligent agent account 3-3 publishes the event conversation content "@AI Chat Assistant What do you think?" and adds the intelligent agent account "AI Chat Assistant" to the target group "National Day Travel gogogo". Here, "What do you think?" is the task information of the target task, "@" is the target character, and "AI Chat Assistant" is the account information of the intelligent agent account.

[0129] In embodiment A2, optionally, the event interaction data used to generate the target recommendation content includes the target session content in the session page. Optionally, in response to a selection operation on the target session content in the session page, the terminal sends the selected target session content to the server; after receiving the target session content, the server performs semantic analysis on the target session content to determine the session data related to the target task; the server generates target prompt words based on the above session data and a preset prompt, and inputs the target prompt words to the target model corresponding to the intelligent agent account deployed on the server to obtain the target recommendation content output by the target model; the server sends the target recommendation content to the terminal so that the target recommendation content is displayed in the terminal's session page.

[0130] In an optional embodiment A3, the task decision event includes: collection of event interaction data for completing the target task.

[0131] In embodiment A3, during the display of the conversation page, the intelligent agent account participating in the target conversation can publish an event interaction data collection page, such as a poll or questionnaire, to collect event interaction data from different non-intelligent agent accounts participating in the target conversation regarding the target task. Therefore, in response to the responses of different non-intelligent agent accounts to the event interaction data collection event, the terminal can determine the intelligent agent account participating in the event interaction data collection event and the response time to the event interaction data collection event, and accordingly determine whether the event interaction data collection for the target task has been completed. Upon completion of the event interaction data collection for the target task, the terminal displays the target recommended content on the conversation page.

[0132] Optionally, in response to a mention of the agent account in the target session, the terminal sends a generation request for event interaction data collection and the target session content from the session page to the server. Upon receiving the generation request and the target session content, the server performs semantic analysis and other processing on the target session content to determine core data related to the target task. Based on the aforementioned core data and the preset requirements for event interaction data collection, the server designs structured prompts for the event interaction data collection page. The server inputs these structured prompts into a specified model corresponding to the event interaction data collection to obtain the event interaction data collection page output by the specified model. The server then sends the event interaction data collection page to the terminal so that the terminal can display the interaction data collection page. Optionally, while sending the generation request and target session content to the server, the terminal can also send the task identifier and account information of the non-agent account participating in the target session to the server. After receiving the task identifier and account information of the non-agent account participating in the target session, the server can further obtain data such as the non-agent account's attention data on content publishing information related to the target task on the content platform and the non-agent account's purchase data on goods related to the target task on the content platform from the data storage system, in order to combine with the target session content to jointly determine the core data.

[0133] For example, such as Figure 2HAs shown, in response to the mention of the intelligent agent account "AI Chat Assistant" by the non-intelligent agent account 4-1, the terminal sends the target session content published by the non-intelligent agent accounts 4-1, 4-2, and 4-3, as well as a generation request for event interaction data collection, to the server. In response to the received target session content and generation request, the server generates a voting page 4-4, which is sent to the terminal as the event interaction data collection page. After receiving the voting page 4-4, the terminal displays the voting page 4-4 published by the intelligent agent account "AI Chat Assistant" in the session page. The voting options include "hot pot" and "barbecue," and these options... This refers to at least two candidate recommendations for the target task; then, in response to different non-agent accounts' selection operations on the options in voting page 4-4, the terminal sends the operation data corresponding to the selection operation as event interaction data to the server; the server, in response to the received operation data, generates voting results 4-5 based on the operation data and sends it to the terminal as the target recommendation content; after receiving voting results 4-5, the terminal displays voting results 4-5 published by the agent account "AI Chat Assistant" on the conversation page. Voting results 4-5 include the selection results of different non-agent accounts on the options in the vote. Among these, the optional options include... Figure 2I As shown, after each non-agent account performs the selection operation to complete the voting, the terminal can output prompt messages 4-6 to indicate that the non-agent account can vote again.

[0134] In embodiment A3, optionally, the event interaction data used to generate the target recommended content includes the collected event interaction data; for example, the voting results of non-agent accounts participating in the target session, the questionnaire answers of non-agent accounts participating in the target session, etc. Optionally, in response to the collection of event interaction data for completing the target task, the terminal sends the collected event interaction data to the server; after receiving the event interaction data, the server generates target prompt words based on the aforementioned event interaction data and a preset prompt, and inputs the target prompt words into the target model corresponding to the agent account deployed on the server, to obtain the target recommended content output by the target model; the server then sends the target recommended content to the terminal as a reply message from the agent account displayed on the terminal's session page. Optionally, the terminal can also, in response to a selection operation on the target session content in the session page, send the selected target session content to the server, so that the server performs semantic analysis and other processing on the target session content to determine the session data related to the target task, and generates target prompt words based on the session data, event interaction data, and the preset prompt.

[0135] In an optional embodiment A4, the task decision event includes: there are at least two opposing conversation contents in the conversation content for the target task.

[0136] In embodiment A4, optionally, during the display of the conversation page, the terminal performs semantic analysis and other processing on the conversation content in the conversation page to determine each piece of content related to the target task, as well as the relationships between the aforementioned tasks, such as opposing relationships, compatible relationships, etc. Then, in response to the presence of at least two opposing conversation pieces in the conversation content related to the target task, the terminal displays the target recommended content in the conversation page by publishing a message through the agent account. Optionally, in response to the selection operation of the target conversation content in the conversation page, the terminal sends the selected target conversation content to the server; after receiving the target conversation content, the server performs semantic analysis and other processing on the target conversation content to determine the conversation data related to the target task; the server generates target prompt words based on the aforementioned conversation data and a preset prompt, and inputs the target prompt words into the target model corresponding to the agent account deployed on the server to obtain the target recommended content output by the target model; the server sends the target recommended content to the terminal as a reply message from the agent account displayed in the terminal's conversation page.

[0137] Optionally, during the display of the conversation page, the terminal sends the conversation content on the conversation page to the server in real time. After receiving the conversation content, the server performs semantic analysis and other processing on the conversation content to determine the content related to the target task and the relationships between these tasks. If the server determines that there are at least two opposing conversation content items related to the target task, the content items related to the target task in the conversation content are used as event interaction data. Based on the content items related to the target task in the conversation content and a preset prompt, the server generates target prompt words and inputs the target prompt words into the target model corresponding to the agent account deployed on the server to obtain the target recommended content output by the target model. The server then sends the target recommended content to the terminal as a reply message from the agent account displayed on the terminal's conversation page. Based on this, the terminal, in response to the presence of at least two opposing conversation content items related to the target task, displays the target recommended content on the conversation page in the form of a message published by the agent account. Optionally, while sending the session content from the session page to the server in real time, the terminal can also send the account information of the non-agent account participating in the target session to the server. After receiving the account information of the non-agent account participating in the target session, the server can further obtain data such as the non-agent account's attention data on content related to the target task published on the content platform and the non-agent account's purchase data on goods related to the target task on the content platform from the data storage system, so as to combine the various content items related to the target task in the above session content to jointly generate target prompt words and obtain target recommended content.

[0138] In one alternative embodiment, it is possible to Figure 1 Based on the application environment shown, the above task decision-making method is implemented through interaction between the server and the terminal, such as... Figure 2J As shown, it includes the following steps:

[0139] S201J, the terminal displays the session page of the target session.

[0140] S202J: In response to a task decision event for a target task in the session page, the terminal sends event interaction data to the server.

[0141] S203J: Upon receiving event interaction data, the server generates target recommendation content for the target task based on the event interaction data.

[0142] S204J: The server sends the target recommended content to the terminal.

[0143] S205J: Upon receiving the target recommended content, the terminal displays the target recommended content published by the agent account on the conversation page.

[0144] In the aforementioned task decision-making method, when a non-agent account makes the task decision, the intelligent agent account enables rapid and effective integration and analysis of the non-agent account's event interaction data regarding the target task, without requiring the non-agent account to spend a significant amount of time communicating. This provides recommended content for the target task, simplifying the task decision-making process, improving decision-making efficiency, and thus enhancing the organizational efficiency of group activities. Furthermore, generating target recommended content based on the non-agent account's event interaction data regarding the target task allows for precise analysis of the non-agent account's task decision-making needs, improving the accuracy and relevance of the generated target recommended content, and enhancing the user experience.

[0145] Based on the above embodiments, in an exemplary embodiment, the display of the target recommended content is further refined. Optionally, such as... Figure 3A As shown, the following steps may be included:

[0146] S301A, in response to a task decision event for a target task in a conversation page, extracts event interaction data of at least two non-agent accounts for the target task from the conversation content related to the target task in the conversation page.

[0147] S302A displays target recommended content generated by the agent account based on event interaction data on the conversation page.

[0148] In response to a task decision event related to a target task on the conversation page, the terminal can determine the conversation content related to the target task within the conversation page, and extract event interaction data for at least two non-agent accounts regarding the target task from the aforementioned conversation content related to the target task. Optionally, the event interaction data includes the conversation content related to the target task, and / or, the conversation data related to the target task within the conversation content related to the target task. Optionally, as the conversation content within the conversation page increases, the agent account can publish conversation content to interact with the non-agent accounts.

[0149] For example, such as Figure 3B As shown, the event interaction data includes the conversation content published by non-agent account 5-1 on the conversation page. This event interaction data also includes conversation content published by other non-agent accounts targeting the specific task. Figure 3B If not shown in the text, then the target recommendation content 5-2 published by the AI ​​chat assistant account will be displayed on the conversation page. Among them, the target recommendation content 5-2 includes the content recommendation result "Fuxing Hot Pot Restaurant", and the corresponding recommendation reference data for the content recommendation result "Fuxing Hot Pot Restaurant" as "because it has the cool environment that @Crispy Dried Fish wants, the spicy flavor that @_Yao Bing likes, and avoids the ingredients that @Fahua Town Road Tsevier is allergic to".

[0150] For example, such as Figure 3C As shown, the event interaction data includes the conversation content posted by non-agent accounts e-1, 5-3, and 5-4 on the conversation page. The conversation page will then display the target recommendation content 5-5 posted by the agent account "AI Chat Assistant". Target recommendation content 5-5 includes two candidate recommendations for the target task, and these two candidate recommendations correspond to the recommended content "hot pot" and "barbecue" for the target task.

[0151] In this embodiment, at least one non-agent account's event interaction data related to the target task is extracted from the conversation content related to the target task to generate target recommendation content for the target task. This can improve the accuracy and relevance of the target recommendation content and enhance the user experience.

[0152] Based on the above embodiments, in an exemplary embodiment, the display of the target recommended content is further refined. Optionally, such as... Figure 4A As shown, the following steps may be included:

[0153] S401A, in response to a task decision event for a target task in a session page, displays an event interaction data collection page for the target task.

[0154] The event interaction data collection page is used to collect event interaction data from at least two non-agent accounts for a target task.

[0155] S402A displays target recommended content generated by the agent account based on event interaction data on the conversation page.

[0156] In this embodiment, during the display of the conversation page, in response to a task decision event for the target task on the conversation page, the terminal can publish an event interaction data collection page for the target task on the conversation page. Then, in response to responses from at least two non-agent accounts to the event interaction data collection page, the terminal obtains the responses from the at least two non-agent accounts as event interaction data for the target task. The terminal sends the aforementioned event interaction data to the server, which generates target recommendation content based on the event interaction data and sends the target recommendation content to the terminal as a reply message from the agent account displayed on the terminal's conversation page. Thus, the terminal can display the target recommendation content published by the agent account on the conversation page. Optionally, after receiving the event interaction data, the server generates target prompt words based on the event interaction data and a preset prompt, and inputs the target prompt words into the target model corresponding to the agent account deployed on the server to obtain the target recommendation content output by the target model.

[0157] Optionally, if the event interaction data collection page is a questionnaire page, the questionnaire questions on the questionnaire page may include questions and corresponding options, and / or questions and corresponding input areas.

[0158] In one optional embodiment, the event interaction items in the event interaction data collection page are determined based on the session content related to the target task in the session page, and / or the content attribute data of at least one candidate recommended content corresponding to the target task.

[0159] In this embodiment, optionally, the session content used to determine the event interaction items in the event interaction data collection page includes the target session content in the session page. For example, session content within a specified duration or a specified number of segments prior to the current moment; or, all session content of at least two non-agent accounts for the target task, etc.

[0160] Optionally, the aforementioned candidate recommended content can be determined based on the historical operation data of at least one non-agent account for each recommended content corresponding to the target task, and / or the currently popular recommended content corresponding to the target task. The aforementioned historical operation data may include follow data, purchase data, collection data, etc.; the aforementioned popular recommended content can be determined based on the recommended content corresponding to the target task that has high interaction data (such as discussion volume, search volume, purchase volume, etc.) within the most recent preset time period (such as this week, the last three days, etc.) (e.g., exceeding a preset interaction threshold, or being ranked first or within a specified number of positions when sorted from high to low). The aforementioned preset time period, interaction data type, preset interaction threshold, and specified number can all be set based on experience, application needs in actual applications, etc., and are not specifically limited in these aspects.

[0161] In one optional embodiment, taking the event interaction data collection page as an example, in response to a task decision event for the target task in the conversation page, the terminal sends a questionnaire generation request and the target conversation content in the conversation page to the server. After receiving the questionnaire generation request and the target conversation content, the server performs semantic analysis and other processing on the target conversation content to determine the core data related to the target task. Based on the above core data and the preset requirements of the questionnaire page (such as the number of questions, the question format, etc.), the server designs structured prompts for the questionnaire page. The server inputs the above structured prompts into the questionnaire generation model to obtain the questionnaire page output by the questionnaire generation model. The server sends the questionnaire page to the terminal as a reply message of the intelligent agent account displayed in the conversation page of the terminal. Optionally, the terminal can also send a task identifier and / or account information of the non-agent account participating in the target session to the server. After receiving the task identifier, the server can determine the currently popular recommended content corresponding to the target task as candidate recommended content for the target task, and then determine the content attribute data of the currently popular recommended content to participate in the generation of the questionnaire page. After receiving the account information of the non-agent account participating in the target session, the server can obtain data such as the non-agent account's attention data to content publishing information related to the target task on the content platform and the non-agent account's purchase data of goods related to the target task on the content platform from the data storage system, based on the account information, to determine the candidate recommended content corresponding to the target task followed by the non-agent account, and then determine the content attribute data of the candidate recommended content corresponding to the target task followed by the non-agent account to participate in the generation of the questionnaire page.

[0162] Optionally, the terminal may send the responses of different agent accounts to the event interaction data collection page to the server after the data collection for the event interaction data collection page is completed; or, it may immediately send the responses of a non-agent account to the event interaction data collection page to the server after obtaining the responses of that agent account.

[0163] In one optional embodiment, S401A may include, in response to a task decision event for the target task in the session page, displaying an access card for the target task in the session page; wherein the access card includes a page access entry for the event interaction data collection page of the target task; and in response to a triggering operation on the page access entry, displaying the event interaction data collection page for the target task.

[0164] Optionally, the event interaction data collection page includes a submission component. In response to a non-agent account's triggering action on the submission component, the terminal sends the non-agent account's responses to the event interaction data collection page to the server, which serve as the non-agent account's event interaction data for the target task. For example, taking a questionnaire page as an example, the event interaction data collection page... Figure 4B As shown, the intelligent agent account "AI Meal Selection Assistant" 6-1 issues an access card 6-2, which includes a page access entry 6-3 marked with the text "Participate Now". In response to a non-intelligent agent account's trigger action on the page access entry 6-3, the terminal can display as follows: Figure 4C The questionnaire page shown includes a submission component 6-4 labeled "Submit Questionnaire".

[0165] Optionally, the server sends an access card and an event interaction data collection page to the terminal; the terminal displays the access card issued by the agent account in the session page, and then, in response to a triggering operation by any non-agent account targeting the page access entry, displays the event interaction data collection page issued by the agent account in the session page. Alternatively, the server sends an access card to the terminal; the terminal first displays the access card issued by the agent account in the session page, and then, in response to a triggering operation by any non-agent account targeting the page access entry, the terminal sends a page retrieval request to the server; after receiving the page retrieval request, the server sends the event interaction data collection page to the terminal; after receiving the event interaction data collection page, the terminal displays the event interaction data collection page issued by the agent account in the session page.

[0166] In one optional embodiment, S402A may include displaying a content display page in the session page in response to a triggered operation on the content viewing entry in the access card; wherein the content display page includes target recommended content. For example, taking the event interaction data collection page as a questionnaire page, such as... Figure 4D As shown, the access card 6-2 published by the intelligent agent account "ai meal selection assistant" 6-1 also includes a content viewing entry 6-5 marked with the text "View Now"; and in response to the trigger operation on the content viewing entry 6-5, the following can be displayed: Figure 4E The content displayed on the page includes target recommendations in decision card 6-6.

[0167] Optionally, after generating the target recommended content, the server sends the target recommended content to the terminal; then, in response to the trigger operation for the content viewing entry in the access card, the terminal displays the content display page published by the agent account on the session page, so as to display the target recommended information on the content display page. Alternatively, in response to the trigger operation for the content viewing entry in the access card, the terminal sends a recommended content retrieval request to the server; after receiving the content retrieval request, the server sends the target recommended content to the terminal; after receiving the target recommended content, the terminal displays the content display page published by the agent account on the session page, so as to display the target recommended information on the content display page.

[0168] In another optional embodiment, S402A may include, in response to the completion of data collection for the event interaction data collection page, switching the page access entry in the access card to a content viewing entry, and in response to a triggering operation on the content viewing entry, displaying a content display page in the session page; wherein the content display page includes target recommended content. For example, taking the event interaction data collection page as a questionnaire page, as... Figure 4F As shown, the intelligent agent account "AI Meal Selection Assistant" 6-1 publishes the conversation content "All group members have finished answering~ The following is the recommended content", and changes the page access entry 6-3 in the published access card 6-2 to the content viewing entry 6-7. In response to the triggered operation on the content viewing entry 6-7, the following can be displayed: Figure 4E The content shown is displayed on the page.

[0169] Optionally, the end of data collection for the event interaction data collection page may include reaching a data collection deadline. This deadline can be a specific moment or a time range starting from the data collection start time corresponding to the event interaction data collection page.

[0170] Optionally, the completion of data collection for the event interaction data collection page may include the completion of page entry for a preset number of non-agent accounts. The preset number can be the total number of all non-agent accounts participating in the target session, such as the total number of all non-agent accounts in the target group, or it can be less than the total number, such as the product of the total number and a preset coefficient, where the preset coefficient is a positive number less than 1. Optionally, the preset number can be pre-set before the target task is initiated, or it can be set specifically for the target task. For example, the server determines the number of non-agent accounts participating in the target session based on the target session content for the target task and the publishing accounts of each piece of content in the target session content sent by the terminal, and uses this as the preset number. Alternatively, the terminal responds to a page switching operation by displaying a quantity selection page, and responds to any non-agent account's quantity selection operation on the quantity selection page, determining that the quantity selected in the quantity selection operation is the preset number.

[0171] Optionally, the completion of data collection for the event interaction data collection page may include the completion of the page for non-intelligent agent accounts within the scope of the target task's corresponding account. The non-intelligent agent accounts within the scope of the target task's corresponding account may include all non-intelligent agent accounts participating in the target dialogue, or a subset of all such non-intelligent agent accounts. Optionally, the server determines the non-intelligent agent accounts within the scope of the target task's corresponding account based on the target session content for the target task and the publishing account of each piece of content in the session page sent by the terminal. Optionally, the terminal responds to the page switching operation by displaying an account selection page, and responds to different non-intelligent agent accounts' account selection operations on the account selection page, determining that the non-intelligent agent account selected in the account selection operation is within the scope of the target task's corresponding account.

[0172] As mentioned earlier, when the preset quantity is the total quantity mentioned above, it can also be considered that the non-intelligent agent accounts within the scope of the target task's corresponding account include all the aforementioned non-intelligent agent accounts. For example, such as Figure 4F As shown, when the data collection for the decision data collection page is completed, the intelligent agent account "AI Chat Assistant" 6-1 publishes the completion message "All group members have finished answering." Optionally, for any non-intelligent agent account within the scope of the account defined for the target task, when the non-intelligent agent account opens the conversation page for the first time during the target conversation, the terminal displays the prompt message published by the intelligent agent account on the conversation page to prompt the non-intelligent agent account to participate in answering questions on the event interaction data collection page.

[0173] The deadline and preset quantity for data collection can be limited based on experience and application requirements in actual use, and no specific limits are set.

[0174] In one exemplary embodiment, it is possible to Figure 1 Based on the application environment shown, the above task decision-making method is implemented through interaction between the server and the terminal, such as... Figure 4G As shown, the following steps may be included:

[0175] S401G, the terminal displays the session page of the target session.

[0176] In response to a task decision event for a target task in the session page, the terminal sends a page generation request and the target session content in the session page to the server.

[0177] In S403G, upon receiving a page generation request, the server generates an access card and an event interaction data collection page based on the target session content. The access card includes the page access entry point for the target task.

[0178] S404G: The server sends access cards and event interaction data collection pages to the terminal.

[0179] After receiving the access card and the event interaction data collection page, the S405G terminal responds to the non-smart agent account's triggered operation on the page access entry and displays the event interaction data collection page.

[0180] S406G: The terminal responds to responses from at least two non-smart agent accounts to the event interaction data collection page, and obtains the responses from at least two smart agent accounts to the event interaction data collection page.

[0181] S407G, the terminal sends its response to the server.

[0182] In the S408G, the server responds to the received response and generates target recommended content based on the response.

[0183] S409G: The server sends the target recommended content to the terminal.

[0184] In response to the completion of data collection for the event interaction data collection page, the terminal in the S410G switches the page access entry in the access card to the content viewing entry.

[0185] In S411G, after receiving the target recommended content, the terminal responds to the triggered operation for the content viewing entry by displaying the content display page; wherein, the content display page includes the target recommended content.

[0186] In this embodiment, event interaction data of at least two non-smart agent accounts for the target task is collected through the event interaction data collection page to generate target recommendation content for the target task. This can improve the accuracy and relevance of the target recommendation content and enhance the user experience.

[0187] Based on the above embodiments, in an optional embodiment, displaying the target recommended content for the target task by the intelligent agent account may include displaying text content; wherein the text content includes at least one content recommendation result and / or recommendation reference data corresponding to at least one content recommendation result. For example, such as Figure 3B As shown, the AI ​​chat assistant account publishes the target recommendation content e-2 in the form of text content on the chat page.

[0188] Based on the above embodiments, in another embodiment, displaying the target recommended content for the target task by the intelligent agent account may include displaying a decision recommendation card; wherein the decision recommendation card includes at least two candidate recommendation items for the target task; the candidate recommendation items correspond to recommended content for the target task. For example, such as Figure 3C As shown, the AI ​​chat assistant account publishes target recommendation content 5-5 in the form of a decision recommendation card on the conversation page. For example, as... Figure 4F As shown, access card 6-2, which includes the content viewing entry point f-7, can be regarded as a decision recommendation card.

[0189] Based on the above embodiments, in another embodiment, displaying the target recommended content for the target task by the intelligent agent account may include displaying a decision card; wherein the decision card displays the selection results of different non-intelligent agent accounts for at least two candidate recommended contents. For example, as... Figure 2H As shown, the AI ​​chat assistant account publishes target recommendation content 4-5 in the form of decision cards on the conversation page; for example, such as Figure 4E As shown, decision card 6-6 is displayed on the conversation page.

[0190] Based on the above embodiments, in an exemplary embodiment, the task decision-making task further includes, when there are dissenting accounts among the participating accounts that do not support the target recommended content, collecting the reasons for the dissenting accounts' decision preferences, so as to update the event interaction data for the target task based on the reasons for the decision preferences.

[0191] In this embodiment, in some cases, not all non-agent accounts participating in the task decision-making event for the target task may support the target recommended content. Therefore, some accounts may disagree on their decision preferences, leading to decision-making disagreements among the non-agent accounts. To improve the accuracy and relevance of the target recommended content and ensure it meets the needs of as many participating accounts as possible, if there are accounts with disagreements on their decision preferences that do not support the target recommended content, the reasons for these disagreements can be collected. Based on these reasons, the event interaction data for the target task is updated, and the process returns to S202A to regenerate the target recommended content. This continues until no accounts with disagreements on their decision preferences support the target recommended content, thus eliminating decision-making disagreements among the non-agent accounts and achieving consensus on the target task. For example, as... Figure 2H As shown, among the four non-intelligent agent accounts participating in the vote, three non-intelligent agent accounts chose "barbecue" and the other non-intelligent agent account chose "hot pot". Therefore, it can be determined that the target recommended content is more inclined to choose "hot pot". Thus, the non-intelligent agent account that chose "hot pot" is the decision preference objection account that does not support the target recommended content. The reasons for the decision preference of the non-intelligent agent account that chose "hot pot" can be collected.

[0192] Optionally, in the presence of accounts with dissenting decision preferences, the server sends a cause collection command to the terminal. Upon receiving the command, the terminal interacts with the dissenting account through its agent account to collect the reasons for the dissenting decision preferences, and then sends these reasons to the server. Upon receiving the reasons, the server updates the event interaction data for the target task based on these reasons, generates new target recommendation content based on the updated event interaction data, and sends the new target recommendation content to the terminal. This allows the terminal to display the new target recommendation content published by the agent account on the session page upon receiving it. Optionally, the server performs semantic analysis and other processing on the reasons for decision preferences to determine preference data related to the target recommendation content. Based on the offset between the preference data and the current event interaction data, the server determines the update direction and extent of the event interaction data, and updates the current event interaction data accordingly to obtain updated event interaction data.

[0193] In one optional embodiment, collecting the reasons for decision preferences from dissenting accounts may include displaying a reason collection page on the session page, and obtaining the reasons for decision preferences from the dissenting account in response to the completion of editing on the reason collection page. Optionally, in response to a reason collection instruction sent by the server, the session page may display mentions of the dissenting account published by the agent account (e.g., the agent account @ the dissenting account in the session page of the target group), and publish the reason collection page; then, the terminal may obtain the reasons for decision preferences edited by the dissenting account in response to the completion of editing on the reason collection page. Optionally, the reason collection page may be a questionnaire page, wherein the questionnaire questions on the questionnaire page are generated based on the session content of the dissenting account regarding the target task.

[0194] In another optional embodiment, collecting the decision preference reasons of the dissenting account may include displaying a reason inquiry in a separate session page corresponding to the dissenting account and the agent account, and obtaining the decision preference reason of the dissenting account based on the dissenting account's response to the reason inquiry. Optionally, the terminal displays a separate session between the agent account and the dissenting account, thereby displaying the publishing reason inquiry content published by the agent account in the separate session page corresponding to the two, and then, in response to the dissenting account's editing operation on the reason inquiry content, determining the dissenting account's response to the reason inquiry content, and obtaining the decision preference reason of the dissenting account based on the above response. For example, the above response is determined as the decision preference reason. For example, such as Figure 5 As shown, this represents the intelligent agent account "AI Chat Assistant" and the decision preference objection account " Figure 2H In the independent chat page corresponding to the non-intelligent agent account "Hot Pot", the intelligent agent account "AI Chat Assistant" publishes the reason inquiry content 7-1.

[0195] In this embodiment, by collecting the reasons for decision preferences from dissenting accounts, and updating the event interaction data for the target task based on these reasons, the target recommended content can be continuously optimized, improving its accuracy and relevance. This ensures that the target recommended content meets the needs of participating accounts as much as possible, thereby enhancing the user experience.

[0196] Based on the above embodiments, in an exemplary embodiment, the target task can be a location recommendation task, such as recommending tourist destinations, shopping destinations, or office locations; correspondingly, the target recommendation content displayed by the terminal is the target recommended location for the location recommendation task; for example, recommending tourist destinations city A and city B; recommending shopping malls C, D, and E; and recommending office buildings F, etc. Optionally, such as... Figure 6 As shown, the task decision-making method may include the following steps:

[0197] S601, Display the session page of the target session.

[0198] S602, in response to a task decision event for a location recommendation task in the session page, the target recommended location for the location recommendation task of the agent account is displayed on the session page; the target recommended location is generated based on the event interaction data of at least two non-agent accounts participating in the target session for the location recommendation task.

[0199] Based on the above embodiments, in an exemplary embodiment, the target task can be a route recommendation task, such as recommending running routes, recommending sightseeing routes, recommending transportation routes, etc.; correspondingly, the target recommendation content displayed by the terminal is a target recommended route for the route recommendation task; for example, recommending multiple running routes, recommending multiple sightseeing routes for the same attraction, recommending transportation routes from the current city to the target city, etc. Optionally, such as... Figure 7 As shown, the task decision-making method may include the following steps:

[0200] S701, Display the session page of the target session.

[0201] S702, in response to a task decision event for the route recommendation task in the session page, displays the target recommended route for the route recommendation task for the agent account in the session page; the target recommended route is generated based on the event interaction data of at least two non-agent accounts participating in the target session for the route recommendation task.

[0202] Below, we will illustrate the creation method of the aforementioned target groups with examples, specifically:

[0203] In one exemplary embodiment, such as Figure 8A As shown, a group creation method is provided, which can be applied to... Figure 1 Taking terminal 102 as an example, the explanation includes:

[0204] S810A, in response to a group creation operation, creates a target group; the participating accounts in the target group include agent accounts.

[0205] The group creation operation can be understood as the operation used to create a target group that includes agent accounts. Participating accounts can be understood as accounts with interactive permissions within the target group, which may include receiving group messages, sending messages / files, and viewing the group member list. For example, participating accounts in the target group may also include user accounts.

[0206] In this context, an intelligent agent account can be understood as a virtual object account generated based on an artificial intelligence model, used to provide conversational services. Artificial intelligence models can include large language models, which can be understood as deep learning models with hundreds of millions or even trillions of parameters.

[0207] The following examples illustrate the creation methods for target groups, including methods B1, B2, and B3. It should be noted that this should not be construed as a limitation on the specific methods for creating target groups.

[0208] Method B1: In response to the group creation operation on the first session page, create the target group; the first session page is a separate session page with the agent account.

[0209] In an optional embodiment, a message browsing page may be displayed in response to a triggering operation on a message tag in the main page; and a first session page may be displayed in response to a triggering operation on an agent-related item in the message browsing page.

[0210] refer to Figure 8B The diagram illustrates the display operation of a message browsing page. The main page (a) may display a navigation bar with at least one navigation label. These navigation labels may include at least one of the following: Homepage label, Shop label, Personal Homepage label, and Messages label (a-1). For example, the Homepage label displays homepage content and at least one filter label for filtering the types of displayed homepage content; filter labels may include, for example, at least one of the following: Follow, Recommended, Live, and Food labels. The Shop label displays product content, which may include products for users to browse, purchase, or post. The Personal Homepage label displays personal homepage content, allowing users to browse and edit their personal homepage content.

[0211] Continue to refer to Figure 8B The system can respond to a trigger operation on message tag a-1 and display message browsing page b. The message browsing page can be embedded in the main page or displayed independently; this application does not impose any limitations on this. Message browsing page b can display a control display area b-1 and a session display area b-2.

[0212] The control display area b-1 may display at least one interactive control, which may include at least one of the following: like and favorite control, new follow control, comment and mention control. The like and favorite control is used to display received likes and favorites; the new follow control is used to display newly followed information; and the comment and mention control is used to display comments and mentions.

[0213] The session display area b-2 can show agent-related items b-3, as well as at least one of the following: created target groups, user private messages, and user groups. Agent-related items can be understood as controls or session items associated with an agent that provides session services; target groups refer to groups whose participating accounts include agent accounts; user groups refer to groups whose participating accounts do not include agent accounts; and user private messages refer to private messages established with user accounts.

[0214] Optional, such as Figure 8B As shown, agent-related item b-3 can be displayed in the conversation display area as a message block. The name of the agent-related item can be "AI Chat Assistant," and this application does not impose any restrictions on the specific name of the agent-related item. The message block corresponding to agent-related item b-3 can display the content of the most recent conversation with the AI ​​Chat Assistant so that users can view unread messages in a timely manner.

[0215] Optional, such as Figure 8C As shown, agent-related item b-3 can also be displayed in the message browsing page b using a functional control. This application does not impose any restrictions on the specific display method of agent-related items.

[0216] refer to Figure 8D The diagram shows the display operation of the first session page. The agent-related item can be an agent control; correspondingly, in response to a trigger operation on agent-related item b-3 in message browsing page b, the first session page c is displayed. The first session page c is an independent session page with the agent account.

[0217] In an optional embodiment, the first session page may display a group creation control, such as "Create AI Group Chat"; correspondingly, in response to the triggering operation of the group creation control in the first session page, an attribute setting page may be displayed; in response to the completion of editing the attribute setting page, the target group is created.

[0218] The attribute settings page can be understood as a page used to set the attributes of the target group. Target group attributes may include at least one of the following: group name, group description, group type, group location, role attributes of the agent account, display mode, and group members to be added.

[0219] In another alternative embodiment, in response to the recognition of a group creation intent based on the session content of the first session page, guidance content for setting attributes of the target group to be created can be displayed on the first session page; in response to the completion of the reply to the guidance content for setting attributes, the target group is created.

[0220] The conversation content on the first conversation page can be understood as the content of the conversation between the current user account and the agent account on the first conversation page.

[0221] The attribute setting guidance content can be understood as content used to guide users in setting attributes for a target group. For example, the attribute setting guidance content can be presented through text, images, and functional controls, and this application does not impose any limitations on this.

[0222] Method B2: In response to mentioning or adding an agent account on the second conversation page, create a target group; the second conversation page is a group chat conversation page; the conversation accounts on the second conversation page do not include agent accounts.

[0223] In an optional embodiment, a message browsing page may be displayed in response to a triggering operation on a message tab on the main page; a second conversation page may be displayed in response to a triggering operation on a user group on the message browsing page; the second conversation page is a group chat conversation page.

[0224] The message browsing page has already been described above and will not be repeated here. A user group can be understood as a group of chat accounts that do not include agent accounts.

[0225] refer to Figure 8E The diagram illustrates the display operation of the second conversation page. The second conversation page d is displayed in response to a trigger operation on user group b-4 in the message browsing page b. For example, the second conversation page d may display a conversation display area d-1 and a content editing area d-2. The conversation display area d-1 may display conversation content between different conversation accounts. The content editing area d-2 may display at least one of the following: a voice input control, a text input control, an emoticon input control, and extended function controls. The voice input control is used to input and send voice content; the text input control is used to input and send text content; the emoticon input control is used to input and send emoticons; and the extended function controls are used to display at least one conversation function control, which may include at least one of the following: an album control, a camera control, a location sharing control, and a group chat control.

[0226] refer to Figure 8FThe diagram illustrates the process of adding an agent account in one embodiment. The second session page d may display a session settings control d-5. Correspondingly, in response to a trigger operation on the session settings control d-5, the group management page e is displayed, which may display an agent addition control e-1. In response to a trigger operation on the agent addition control, the agent account is added to the user group, thus creating the target group.

[0227] For example, group entry settings, group entry management, and function settings can be implemented on the group management page. Group entry settings may include at least one of the following: entry method, entry threshold, and viewing received group entry applications; group entry management may allow setting whether to display historical messages to new members and setting a welcome message; function settings may allow setting whether to publicly display the group chat, setting member message permissions, setting a group map, and whether to automatically remove inactive members, etc. This application does not impose any limitations on the specific management content on the group management page.

[0228] refer to Figure 8G The diagram illustrates the addition of an agent account in another embodiment. In response to a trigger operation on the session settings control d-5, a session settings page f is displayed. The session settings page f may display a group member management control f-1 and a group management control f-2. In response to a trigger operation on the group member management control f-1, a participating account to be added to the user group can be selected, including the agent account. In response to the selection of the agent account, the agent account is added to the user group, thus creating the target group. In response to a trigger operation on the group management control f-2, a group management page is displayed.

[0229] refer to Figure 8H The diagram illustrates the addition of an agent account in another embodiment. In response to a triggering operation on the agent management control e-2 in the group management page e, an agent account list e-3 is displayed. This list e-3 includes different agent accounts, each with different role attributes or function types. Correspondingly, in response to a selection and addition operation of an agent account, the selected agent account is added to the user group, thus creating the target group. Optionally, in response to a selection and editing operation of an agent account, the selected agent account can be edited, such as modifying its role attributes, account name, or deleting it.

[0230] For example, the agent account list can display at least one of the following: agent name and agent tags for different agent accounts. Agent tags can include at least one of the following: application type tag, usage frequency tag, and role attribute tag. Application type tags can include one of the following: group chat decision-making type tag and atmosphere-building type tag; usage frequency tags can include commonly used tags; role attribute tags can include one of the following: mature and stable attribute tag and positive and proactive attribute tag. This application does not limit the specific display method of agent accounts.

[0231] For example, the group management page may also display an agent creation control; correspondingly, in response to the triggering operation of the agent creation control, an agent account can be created and added to the agent account list. Creating an agent account may include setting at least one of the following: agent account role attributes, function type, and agent name.

[0232] refer to Figure 8I The diagram illustrates the addition of an agent account in another embodiment. In this embodiment, the content editing area d-2 of the second session page d can display an agent control d-6; correspondingly, in response to a trigger operation on the agent control d-6 in the content editing area d-2, the agent account can be added to the user group, thus creating the target group.

[0233] refer to Figure 8J The diagram illustrates a mention operation for an agent account in one embodiment. In response to a mention of an agent account entered in the content editing area d-2, the agent account can be added to a user group, thus creating the target group. The mention content can include a mention symbol and the name of the mentioned agent, such as "@AI Chat Assistant".

[0234] Method B3: In response to a group sharing operation on the content sharing page targeting at least two accounts to be shared with, create a target group; wherein, at least two accounts to be shared with include an agent account.

[0235] The content sharing page can be understood as a page used to share target content, which may include at least one of the following: notes, images, product cards, and live stream links.

[0236] It should be noted that the term "content sharing page" should be interpreted broadly. For example, a content sharing page may include a live streaming sharing page accessed by clicking a live streaming sharing control, a note sharing page accessed by clicking a note sharing control, or a product sharing page accessed by clicking a product sharing control, etc. This application does not limit the content type of the target content, nor does it limit the specific display link of the content sharing page.

[0237] refer to Figure 8K The diagram illustrates a group sharing operation. When the selected account for sharing on the content sharing page includes an agent account, a target group creation control can be displayed; correspondingly, a target group can be created in response to triggering the target group creation control. For example, when the selected account for sharing on the content sharing page does not include an agent account, a user group creation control can be displayed; correspondingly, a user group can be created in response to triggering the user group creation control. For example, the content sharing page may also display separate send controls; correspondingly, in response to triggering the separate send controls, the content to be shared can be shared separately to the selected accounts. For example, the content sharing page may also display an editing area for editing the sharing description; correspondingly, in response to triggering the target group creation control, the content to be shared and / or the sharing description can be sent to the created target group; wherein the sharing description can include at least one of the following: sharing message text, sharing description emoticons, and sharing description images.

[0238] In an optional embodiment, the target group's group name may display an AI identifier to distinguish it from user groups.

[0239] S820A: During the creation of the target group, in response to the role editing operation of the agent account, the role attributes of the agent account are configured so that the agent account can provide session services in the target group based on the set role attributes.

[0240] Role attributes can include at least one of the following: defining the agent's role characteristics, behavioral patterns, conversational style, and service scope within the target group. Role attributes determine how the agent understands the group context, how it interacts with group members, and what type of conversational services it provides.

[0241] Optionally, the attribute type of a role can include at least one of the following: role trait type, professional field type, service function type, and social role type. For example, a role trait type can include at least one of the following: mature and stable, positive and optimistic, humorous and witty, meticulous and detail-oriented, and tactful. A professional field type can include at least one of the following: gourmet / foodie, travel expert, technical expert, and fitness instructor. A service function type can include at least one of the following: conflict mediator, creative facilitator, and schedule manager. A social role type can include at least one of the following: topic facilitator, atmosphere facilitator, and knowledge sharer. It should be noted that this application does not impose any limitations on the specific type or content of the role attributes.

[0242] The following is an example of how to configure character attributes. This application does not impose any restrictions on the specific configuration method of character attributes.

[0243] In an optional embodiment, in response to a mention or addition operation of the agent account in the second session page, a role configuration request may be sent to the platform server, so that the platform server can determine the role attributes of the agent account based on the historical session content in the second session page and configure the role attributes of the agent account.

[0244] The second conversation page has already been described above and will not be repeated here. The historical conversation content of the second conversation page can be understood as the conversation content between different conversation accounts within the user group to which the second conversation page belongs, before the addition of the agent account. For example, historical conversation content may include attribute requirements for the agent account, as well as chat content that reflects the user group's communication style or topic preferences.

[0245] For example, attribute requirement content can be extracted from historical session content; the role attributes of the agent account can be determined based on the attribute requirement content.

[0246] For example, historical conversation content can be input into the first role determination model to obtain the role attributes of the agent account. The first role determination model can be a traditional machine learning model or a large language model; this application does not impose any limitations on it.

[0247] In an optional embodiment, in response to a group sharing operation on a content sharing page targeting at least two accounts to be shared, a role configuration request may be sent to the platform server, so that the platform server determines the role attributes of the agent account based on the content to be shared and / or the account attributes of the non-agent accounts among the at least two accounts to be shared, and configures the role attributes of the agent account.

[0248] The content sharing page has already been described above and will not be repeated here. The content to be shared can be understood as the content to be shared through the content sharing page. Account attributes may include at least one of the following: basic profile attributes, interest preference attributes, and social relationship attributes. Optionally, account attributes can be publicly available attributes of the account to be shared, thereby ensuring the protection of personal privacy.

[0249] Optionally, the role attribute of the agent account can be determined based on at least one of the account attributes of the currently logged-in account and the account attributes of the account to be shared (excluding agent accounts).

[0250] For example, basic profile attributes may include at least one of the following: the user's age, gender, and occupation. Interest preference attributes may include at least one of the following: the type of content posted or forwarded, and the content of the target topics followed. Social relationship attributes may include at least one of the following: number of likes, number of followers, and activity level.

[0251] For example, the content to be shared and / or the account attributes of non-agent accounts from at least two accounts to be shared can be input into the second role determination model to obtain the role attributes of the agent accounts. The second role determination model can be a traditional machine learning model or a large language model; this application does not impose any limitations on it.

[0252] In an optional embodiment, in response to a triggered operation of the role editing control for the agent account on the group management page corresponding to the target group, the role editing page for the agent account can be displayed, and in response to the role editing operation for the agent account on the role editing page, the role attributes of the agent account can be configured.

[0253] For example, the character editing page may display at least one selectable character attribute tag, thereby configuring the character attributes of the agent account according to the selected character attribute tag.

[0254] For example, the character editing page can display an editable character attribute editing bar, thereby configuring the character attributes of the intelligent agent account according to the edited character attribute content.

[0255] In one implementation, the character editing control may include Figure 8F Add control e-1 to the agent; correspondingly, in response to the trigger operation of adding control e-1 to the agent, the role editing page of the agent account can be displayed, thereby enabling the configuration of the role attributes of the agent account through the role editing operation on the role editing page.

[0256] In another implementation, the character editing control may includeFigure 8G The agent management control e-2 is configured accordingly. In response to the trigger operation of adding control e-1 to the agent, the role editing options for different agent accounts can be displayed in the agent account list e-3. In response to the trigger operation of the role editing options for the selected agent account, the role editing operation for the agent account in the role editing page can be performed to configure its role attributes.

[0257] In another implementation, the group management page may include a separate role editing control for configuring the role attributes of agent accounts. This application does not impose any limitations on the specific display location of the role editing control.

[0258] In an optional embodiment, in response to a triggered operation of the role editing control for the agent account in the first session page, the role editing page for the agent account is displayed, and in response to the role editing operation for the agent account in the role editing page, the role attributes of the agent account are configured; the first session page is an independent session page for the agent account.

[0259] In one implementation, an attribute settings page may be displayed in response to a triggered operation on a group creation control in a first session page, the attribute settings page of which may display a role editing control.

[0260] In another implementation, after the first session page is established, a character editing control can be displayed on the first session page.

[0261] In another implementation, in response to the recognition of a group creation intent based on the session content of the first session page, attribute setting guidance content for the target group to be created can be displayed on the first session page, including role editing controls.

[0262] The aforementioned group creation method creates a target group in response to a group creation operation. The participating accounts in the target group include agent accounts, thus introducing agents into the group interaction environment and providing a basic framework for subsequent group-based conversation services. By configuring the role attributes of agent accounts during the target group creation process in response to role editing operations, customizable agent attribute configuration is achieved. This allows agents to flexibly adapt to diverse scenarios in different groups based on customized role attributes, overcoming the limitations of traditional, rigid agent models. Simultaneously, during agent participation in group conversations, the method effectively empowers the group interaction process, improving group collaboration and decision-making efficiency, and enhancing the user experience.

[0263] Based on the above embodiments, the above group creation method can be implemented through the interaction between the terminal and the server.

[0264] refer to Figure 8L The sequence diagram shown is for the group creation method, including:

[0265] The S810L terminal responds to the group creation operation by interacting with the server to create a target group, and the participating accounts in the target group include agent accounts.

[0266] During the creation of the target group, the S820L terminal responds to the role editing operation of the agent account by sending role configuration information to the server.

[0267] The S830L server configures the role attributes of the agent account based on the role configuration information, so that the agent account can provide session services in the target group based on the set role attributes.

[0268] Optionally, the role configuration information may include historical session content from the second session page.

[0269] Optionally, the role configuration information may include the content to be shared and / or the account attributes of the non-agent accounts among at least two accounts to be shared.

[0270] Optionally, the role configuration information may include role editing information edited on the role editing page. The role editing page may be triggered by the role editing control for the agent account on the group management page, or it may be triggered by the role editing control for the agent account on the first session page; this application does not impose any limitations on this.

[0271] Based on the technical solutions of the above embodiments, this application also provides an optional embodiment in which the steps for creating the target group are refined.

[0272] See Figure 9A The diagram illustrates the steps for creating a target group in one embodiment, including:

[0273] S910A, in response to a triggered operation on the group creation control in the first session page, displays the property settings page.

[0274] S920A: In response to completion of editing on the property settings page, create the target group.

[0275] The attribute settings page may include an account addition page and / or a group settings page.

[0276] refer to Figure 9BThe diagram illustrates the creation process of the target group in the first embodiment. In response to a triggering operation on the group creation control c-1 in the first session page c, the account addition page g is displayed; in response to the completion of adding participating accounts on the account addition page g, the target group is created; and on the account addition page g, session accounts from the first session page are selected.

[0277] For example, in response to the selection of participating accounts on the account addition page g, the avatars of the selected participating accounts and the total number of additions can be displayed on the account addition page g; in response to the addition completion control on the account addition page g, the target group can be created.

[0278] refer to Figure 9C The diagram illustrates the creation process of the target group in the second embodiment. In response to a triggering operation on the group creation control c-1 in the first session page c, the group settings page h is displayed; and in response to the completion of setting the attributes of the group settings page h, the target group is created.

[0279] For example, the group settings page h may display at least one of the following: a group name editing area, a group description editing area, a group type selection control, a group location selection control, a role editing control, and a personal homepage display control.

[0280] Among these features, users can edit the group name in the group name editing area, such as "National Day Travel"; edit the group description in the group description area, such as "Welcome to join the 'National Day Travel' discussion group. Let's talk about our favorite destinations and route guides"; select the group type using the group type selection control, such as at least one of travel, food, and study types; configure the role attributes of the agent account using the role editing control, so that the agent account can provide conversation services in the target group based on the set role attributes; and select whether to display the target group chat on the personal homepage using the personal homepage display control.

[0281] refer to Figure 9D The diagram illustrates a role editing operation for an agent in one embodiment. In response to a triggering operation on the role editing control h-1 in the group settings page h, the role editing page I is displayed. Furthermore, in response to a role editing operation for the agent account on role editing page I, role attributes for the agent account are configured. For example, in response to a role editing operation for the agent account on the role editing page, the user can return to the group settings page.

[0282] For example, the character editing page I may display at least one of the following: a character nickname editing bar, a character attribute label, and a character attribute editing bar.

[0283] The nickname editing section is used to edit the nickname of the intelligent agent, such as "Travel Planning Assistant".

[0284] The tag type for the role attribute tag can include at least one of the following: role trait type, professional field type, service function type, and social role type.

[0285] For example, role trait types may include at least one of the following: mature and stable, positive and optimistic, humorous and witty, meticulous and detail-oriented, and tactful. Professional field types may include at least one of the following: gourmet / foodie, travel expert, technical expert, and fitness instructor. Service function types may include at least one of the following: conflict mediator, creative facilitator, and schedule manager. Social role types may include at least one of the following: topic facilitator, atmosphere maker, and knowledge sharer. (Continue to refer to...) Figure 9D , Figure 9D The document shows that character attribute tags may include at least one of the following: "foodie tag," "composed tag," "lively and cute tag," and "tactful tag." This application does not impose any limitations on the specific types or content of character attribute tags.

[0286] The role attribute editing panel is used to edit the role description information of the intelligent agent. The role description information may include at least one of the following: application background, personality, ability, and language style.

[0287] refer to Figure 9E The diagram illustrates the display operation of the target group. The target group can be created and displayed in response to a triggering operation of the control creation function on the group settings page (h). For example, after creating the target group, the agent account can send a welcome message in the target group, such as "Welcome to the 'National Day Travel' discussion group! Let's talk about our favorite places and travel routes!" For example, the welcome message can be generated based on the group name, group description, and the agent's role attributes.

[0288] refer to Figure 9F The diagram illustrates the creation process of the target group in the third embodiment. In response to a triggering operation on the group creation control c-1 in the first session page c, the account addition page g is displayed; in response to the completion of adding participating accounts on the account addition page g, the group settings page h is displayed; and in response to the completion of setting the attributes on the group settings page h, the target group is created.

[0289] It should be noted that this application does not impose any restrictions on the display order between the account addition page and the group settings page.

[0290] For example, in response to a triggered operation on the group creation control in the first session page, the group settings page can be displayed; in response to the completion of the property settings on the group settings page, the account addition page can be displayed; and in response to the completion of the addition of participating accounts on the account addition page, the target group can be created.

[0291] For example, in response to a triggering operation on the group creation control in the first session page, both the account addition page and the group settings page can be displayed simultaneously; and in response to completion of editing on the attribute settings page, the target group can be created. The attribute settings page includes the account addition page and the group settings page.

[0292] Based on the above embodiments, the creation steps of the target group can be achieved through the interaction between the terminal and the server.

[0293] refer to Figure 9G The diagram shown is a timing diagram of the steps for creating a target group in one embodiment, including:

[0294] The S910G terminal responds to the triggered operation of the group creation control on the first session page and displays the account addition page.

[0295] The S920G terminal responds to the completion of adding participating accounts on the account addition page and displays the group settings page.

[0296] Upon completion of setting the attributes on the group settings page, the S930G terminal sends the set group attribute information and the account information of the added participating accounts to the server.

[0297] The group attribute information may include at least one of the following: group name, group description, group type, group location, and agent role attributes.

[0298] The S940G server creates the target group based on group attribute information and account information.

[0299] Based on the technical solutions of the above embodiments, this application also provides an optional embodiment, in which another step for creating a target group is provided.

[0300] See Figure 10A The following is a step in creating a target group, as shown in another embodiment:

[0301] S1010A: In response to the recognition of a group creation intent based on the session content of the first session page, guide content for setting attributes of the target group to be created is displayed on the first session page.

[0302] The intention to create a group can be understood as the desire or need to create a group that includes intelligent agents.

[0303] In an optional embodiment, if the conversation content is identified to contain target keywords or target phrases, guidance on setting attributes for the target group to be created can be displayed on the first conversation page. The target keywords can be set by a technician based on needs or experience, or determined through extensive experimentation; this application does not impose any limitations on this. For example, the target keywords could include "create AI group chat".

[0304] In another optional embodiment, the content type of the session content can be identified using a semantic recognition model. If the content type indicates that the session content contains an intent to create a group, then the attribute setting guidance content for the target group to be created can be displayed on the first session page. The semantic recognition model can be a traditional machine learning model or a large language model; this application does not impose any limitations on it.

[0305] In another alternative embodiment, in response to the identification of a group creation intent based on the conversation content of the first conversation page, an inquiry message for creating the target group can be displayed; in response to obtaining a positive response to the inquiry message, guidance content for setting attributes of the target group to be created can be displayed on the first conversation page.

[0306] S1020A: In response to the completion of the reply to the guidance content for setting attributes, create the target group.

[0307] In an optional embodiment, the attribute setting guidance content may include account addition guidance content and / or group setting guidance content. The account addition guidance content can be understood as content used to guide users to add participating accounts for the target group session; the group setting guidance content can be understood as content used to guide users to set group attribute information.

[0308] For example, the group settings guidance content may include at least one of the following: group name guidance content, group introduction guidance content, role attribute guidance content, modification suggestion content, inquiry content, and settings feedback content. The modification suggestion content can be understood as attribute setting suggestions in response to attribute settings; the inquiry content may include inquiries regarding attribute setting responses or modification suggestions; and the settings feedback content can be understood as the notification information provided upon successful attribute setting.

[0309] In one implementation, a target group can be created in response to the completion of a first reply to add guiding content to an account; the first reply is associated with participating accounts in the target group. In another implementation, a target group can be created in response to the completion of a second reply to set guiding content for a group; the second reply is associated with attribute configuration information for the target group. In yet another implementation, a target group can be created in response to the completion of both the first reply to add guiding content to an account and the second reply to set guiding content for a group.

[0310] In an optional embodiment, the attribute setting guidance content may include text guidance content and / or control guidance content. Text guidance content can be understood as guidance information based on natural language descriptions, used to guide users to provide the various group attribute information required for creating a group through free text input. Control guidance content can be understood as guidance information based on graphical user interface elements, used to guide users to provide the various group attribute information required for creating a group through standardized operations such as clicking and checking boxes using structured interactive components such as buttons, selectors, and forms. Group attribute information may include at least one of the following: group name, group description, agent role attributes, and accounts to be added.

[0311] In one implementation, a target group can be created in response to completion of a content reply to the text-guided content, based on attribute settings associated with the target group.

[0312] refer to Figure 10B The diagram shown illustrates the creation process of the target group in the fourth embodiment. From Figure 10B The example shows that group introduction prompts can include phrases like "Let's set a group introduction." Furthermore, after replying with "National Day travel," suggestions for modification and inquiries can be displayed. Upon receiving a positive response, the prompts can continue with further guidance, such as "Give me a likable persona to keep the group chat lively." Based on this approach, responses can be obtained for different attribute settings, leading to the creation of the target group upon completion of the attribute setting prompts.

[0313] For example, the completion of the response to the attribute setting guidance content may include at least one of the following: obtaining a preset type of attribute setting response, i.e., completing the preset guidance process; or recognizing that the session content includes a session message used to indicate the end of the guidance process, such as "Create target group according to the above attribute setting response".

[0314] In another implementation, in response to a triggering operation on a guide control in the control guide content, the property setting page corresponding to the guide control can be displayed; in response to a property editing operation on the property setting page corresponding to the guide control, the guide properties of the target group corresponding to the guide control can be configured; and in response to the completion of the configuration of the guide properties corresponding to each guide control in the control guide content, the target group can be created.

[0315] The guide control can correspond to at least one control type. The property settings pages differ for different control types.

[0316] refer to Figure 10C The diagram illustrates the creation process of the target group in the fifth embodiment. Wherein, Figure 10C The example illustrates that the guidance controls may include at least one of a role attribute guidance control c-2 and an account addition guidance control c-3. The role attribute guidance control c-2 guides the user to edit the role attributes of the agent; the account addition guidance control guides the user to add a participating account for the target group.

[0317] For example, the role attribute guide control and the account addition guide control can be displayed in sequence. For instance, if the agent's role attributes are obtained through the role attribute guide control, the account addition guide control can be displayed. This application does not impose any limitation on the display order of the guide controls.

[0318] Optionally, users can be guided to create a target group through a combination of text-based and control-based guidance.

[0319] refer to Figure 10D The diagram illustrates the guided operation of the character attribute guide control. It can respond to a triggering operation on the character attribute guide control c-2 in the first session page c, displaying the character editing page I, and then returning to the first session page upon completion of editing the agent account on the character editing page I.

[0320] The character editing page I has already been described above and will not be repeated here. Optionally, in response to the completion of the editing operation on the character editing page I, a completion message can be replied on the first session page c.

[0321] refer to Figure 10E The diagram illustrates the guided operation of the account addition guide control. It can display the account addition page g in response to a triggering operation on the account addition guide control c-3 in the first session page c; and return to the first session page in response to the completion of account addition for participating accounts in the account addition page g.

[0322] The account addition page g has already been described above and will not be repeated here. Optionally, in response to the completion of account addition on the account addition page g, a completion message can be replied on the first session page c.

[0323] Similarly, in response to a triggering action on the group settings controls in the first session page, the group settings page can be displayed; and in response to the completion of group settings on the group settings page, the user can return to the first session page. Optionally, in response to the completion of editing on the group settings page, a completion message can be displayed on the first session page.

[0324] Based on the above embodiments, the creation steps of the target group can be achieved through the interaction between the terminal and the server.

[0325] refer to Figure 10F The diagram shown is a timing diagram of the steps for creating a target group in another embodiment, including:

[0326] S1010F: The terminal sends the session content of the first session page to the server.

[0327] S1020F: In response to the server's recognition of a group creation intent based on the session content, the server sends guidance content on setting attributes for the target group to be created to the terminal.

[0328] The attribute setting guidance includes guidance on adding an account and guidance on group settings.

[0329] The S1030F terminal displays the attribute setting guide on the first session page.

[0330] S1040F: Upon completion of the first reply to the account addition guidance content, the terminal sends the account information of the added participating account to the server.

[0331] S1050F: Upon completion of the first reply to the group setting guidance content, the terminal sends the set group attribute information to the server.

[0332] The group attribute information may include at least one of the following: group name, group description, group type, group location, and agent role attributes.

[0333] S1060F: The server creates the target group based on account information and group attribute information.

[0334] S1070F: The server sends a message to the terminal indicating that the target group has been created.

[0335] S1080F: Upon receiving the creation completion message, the terminal displays the session page for the target group.

[0336] 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 in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0337] Based on the same inventive concept, this application also provides a task decision-making apparatus for implementing the task decision-making method described above. The solution provided by this apparatus is similar to the solution described in the above method; therefore, the specific limitations in one or more task decision-making apparatus embodiments provided below can be found in the limitations of the task decision-making method described above, and will not be repeated here.

[0338] In one exemplary embodiment, such as Figure 11 As shown, a task decision-making device is provided, including: a first display module 1110 and a second display module 1120, wherein:

[0339] The first display module 1110 is used to display the session page of the target session;

[0340] The second display module 1120 is used to respond to a task decision event for the target task in the session page and display the target recommended content of the agent account for the target task in the session page; wherein the target recommended content is generated based on the event interaction data of at least two non-agent accounts participating in the target session for the target task.

[0341] Optionally, when the target task is a location recommendation task, the second display module 1120 is specifically used to respond to a task decision event for the location recommendation task in the session page and display the target recommended location for the location recommendation task in the session page; wherein, the target recommended location is generated based on the event interaction data of at least two non-agent accounts participating in the target session for the location recommendation task.

[0342] Optionally, when the target task is a route recommendation task, the second display module 1120 is specifically used to respond to a task decision event for the route recommendation task in the session page and display the target recommended route for the location recommendation task in the session page; wherein, the target recommended route is generated based on the event interaction data of at least two non-agent accounts participating in the target session for the route recommendation task.

[0343] In one exemplary embodiment, the task decision event includes at least one of the following:

[0344] In response to a member addition operation for a target session, determine the target historical session content in the target session and add the agent account to the target session;

[0345] Send event session content in the target session; the event session content includes the target character, the agent account's account information, and the target task's task information;

[0346] Collect event interaction data to complete the target task;

[0347] There are at least two conflicting conversations in the conversation content related to the target task.

[0348] In one exemplary embodiment, the second display module 920 includes:

[0349] An event response unit is used to respond to a task decision event for a target task in a session page, and to extract event interaction data of at least two non-agent accounts for the target task from the session content related to the target task in the session page; or, in response to a task decision event for a target task in a session page, to display an event interaction data collection page for the target task; wherein the event interaction data collection page is used to collect event interaction data of at least two non-agent accounts for the target task.

[0350] The content display unit is used to display target recommended content generated by the agent account based on event interaction data on the conversation page.

[0351] In one exemplary embodiment, the data collection items in the event interaction data collection page are determined based on the session content related to the target task in the session page, and / or the content attribute data of at least one candidate recommended content corresponding to the target task.

[0352] In an exemplary embodiment, the event response unit is specifically configured to: in response to a task decision event for the target task in the session page, display an access card for the target task in the session page; wherein the access card includes a page access entry for the event interaction data collection page of the target task; and in response to a trigger operation on the page access entry, display the event interaction data collection page for the target task.

[0353] In an exemplary embodiment, the content display unit is specifically configured to: display a content display page on the session page in response to a trigger operation on the content viewing entry in the access card; or, in response to the end of data collection on the event interaction data collection page, switch the page access entry in the access card to the content viewing entry, and display the content display page on the session page in response to a trigger operation on the content viewing entry; wherein the content display page includes target recommended content generated by the agent account based on event interaction data of at least one non-agent account.

[0354] In one exemplary embodiment, the completion of data collection for the event interaction data collection page includes at least one of the following:

[0355] The data collection deadline has been reached;

[0356] The page for the preset number of non-smart agent accounts has been filled out;

[0357] The page for non-agent accounts within the defined account range corresponding to the target task has been filled out.

[0358] In one exemplary embodiment, the second display module 920 is specifically used for:

[0359] Display text content; wherein the text content includes at least one content recommendation result and / or recommendation reference data corresponding to at least one content recommendation result;

[0360] Display a decision recommendation card; wherein the decision recommendation card includes at least two candidate recommendation items for the target task; the candidate recommendation items correspond to the recommended content for the target task;

[0361] Display decision cards; in these decision cards, different non-agent accounts show the selection results of at least two candidate recommendations.

[0362] In one exemplary embodiment, the task decision-making device further includes:

[0363] The reason collection module is used to collect the reasons for the decision preferences of accounts that disagree with the decision preferences in the target session, so as to update the event interaction data for the target task based on the reasons for the decision preferences.

[0364] In one exemplary embodiment, the cause collection module is specifically configured to perform any of the following:

[0365] In the conversation page of the target group corresponding to the decision preference objection account, the reason collection page is displayed, and in response to the decision preference objection account completing the editing on the reason collection page, the reason for the decision preference objection account is obtained;

[0366] In the separate conversation pages corresponding to the decision preference objection account and the agent account, the reason query content is displayed, and the reason for the decision preference objection account is obtained based on the decision preference objection account's reply to the reason query content.

[0367] Each module in the aforementioned task decision-making device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0368] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 12 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements a task decision-making method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0369] Those skilled in the art will understand that Figure 12 The 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.

[0370] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described method embodiments.

[0371] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method embodiments.

[0372] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the above-described method embodiments.

[0373] 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.

[0374] 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 of the embodiments described above. 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.

[0375] 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.

[0376] 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 task decision-making method, characterized in that, include: Display the session page of the target session; In response to a task decision event for a target task on the session page, the agent account displays target recommended content for the target task on the session page; The target recommended content is generated based on event interaction data of at least two non-agent accounts participating in the target session regarding the target task.

2. The method according to claim 1, characterized in that, Task decision events include at least one of the following: In response to the member addition operation for the target session, the target historical session content in the target session is determined, and the agent account is added to the target session; Send event session content in the target session; wherein, the event session content includes target characters, account information of the agent account, and task information of the target task; Collect event interaction data to complete the target task; There are at least two opposing conversational messages in the conversational content related to the target task.

3. The method according to claim 1, characterized in that, The response to a task decision event for a target task on the session page, displaying target recommendation content for the target task from the agent account on the session page, includes: In response to a task decision event for the target task in the session page, extract at least two event interaction data of the non-agent accounts for the target task from the session content related to the target task within the session page; or, In response to a task decision event for a target task in the session page, an event interaction data collection page for the target task is displayed; wherein, the event interaction data collection page is used to collect event interaction data for the target task from at least two of the non-agent accounts. The conversation page displays target recommended content generated by the agent account based on the event interaction data.

4. The method according to claim 3, characterized in that, The data collection items in the event interaction data collection page are determined based on the session content related to the target task in the session page, and / or the content attribute data of at least one candidate recommended content corresponding to the target task.

5. The method according to claim 3, characterized in that, The response to a task decision event for a target task on the session page, displaying an event interaction data collection page for the target task, includes: In response to a task decision event for a target task in the session page, an access card for the target task is displayed on the session page; wherein the access card includes a page access entry for the event interaction data collection page of the target task; In response to a triggered operation on the page access entry, an event interaction data collection page for the target task is displayed.

6. The method according to claim 5, characterized in that, The step of displaying target recommended content generated by the agent account based on event interaction data from at least one of the non-agent accounts on the conversation page includes: In response to a triggered operation on the content viewing entry in the access card, a content display page is displayed on the session page; or, In response to the completion of data collection for the event interaction data collection page, the page access entry in the access card is switched to the content viewing entry, and in response to the triggering operation of the content viewing entry, the content display page is displayed in the session page; The content display page includes the target recommended content.

7. The method according to claim 6, characterized in that, The completion of data collection for the event interaction data collection page includes at least one of the following: The data collection deadline has been reached; The page for the preset number of non-smart agent accounts has been filled out; The page for non-agent accounts within the defined account range corresponding to the target task has been filled out.

8. The method according to any one of claims 1-7, characterized in that, The target recommended content for the target task by the display agent account includes at least one of the following: Display text content; wherein the text content includes at least one content recommendation result and / or at least one recommendation reference data corresponding to the content recommendation result; Display a decision recommendation card; wherein the decision recommendation card includes at least two candidate recommendation items for the target task; the candidate recommendation items correspond to recommended content for the target task; Display decision cards; wherein, the decision cards show the selection results of different non-agent accounts for at least two candidate recommended contents.

9. The method according to any one of claims 1-7, characterized in that, The method further includes: If there are accounts among the accounts participating in the target session that disagree with the decision preferences and do not support the target recommended content, the reasons for the decision preferences of the dissenting accounts are collected, and the event interaction data for the target task is updated based on the reasons for the decision preferences.

10. The method according to claim 9, characterized in that, The reasons for decision preference objections collected from the decision preference objection accounts include at least one of the following: In the conversation page, a reason collection page is displayed, and in response to the completion of editing on the reason collection page by the decision preference objection account, the reason for the decision preference objection account is obtained; In the independent session page corresponding to the decision preference objection account and the agent account, the reason for the decision preference objection is displayed, and the reason for the decision preference objection is obtained based on the response of the decision preference objection account to the reason for the reason.

11. A task decision-making method, characterized in that, include: Display the session page of the target session; In response to a task decision event for a location recommendation task on the session page, the target recommended location for the location recommendation task by the agent account is displayed on the session page; The target recommended location is generated based on event interaction data of at least two non-agent accounts participating in the target session regarding the location recommendation task.

12. A task decision-making method, characterized in that, include: Display the session page of the target session; In response to a task decision event for the route recommendation task on the session page, the target recommended route for the route recommendation task by the agent account is displayed on the session page; The target recommended route is generated based on event interaction data of at least two non-agent accounts participating in the target session regarding the route recommendation task.

13. A task decision-making device, characterized in that, include: The first display module is used to display the session page of the target session; The second display module is used to display target recommendation content of the agent account for the target task in response to a task decision event for the target task in the session page; The target recommended content is generated based on event interaction data of at least two non-agent accounts participating in the target session regarding the target task.

14. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1-12.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-12.

16. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-12.

Images