Message interaction method and related device

Abstract

This application discloses a message interaction method and related devices. The relevant embodiments can be applied to various scenarios such as cloud technology, artificial intelligence, and intelligent transportation. It can display the session page of a target session in a message interaction client. The session page includes a message content input area and a recommended message content area. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. In response to a trigger operation on the target message interaction control, the method sends the target recommended message content corresponding to the target message interaction control to the client corresponding to the session member of the target session. It also receives and displays the reply content sent by the intelligent virtual session object regarding the target recommended message content. This embodiment allows direct input of recommended message content based on a trigger operation on the message interaction control in the recommended message content area, eliminating the need for manual input and improving the efficiency of human-computer message interaction in the target session.

Description

Technical Field

[0001] This application relates to the field of computer technology, specifically to a message interaction method and related equipment. Background Technology

[0002] With the rapid development of artificial intelligence and natural language processing technologies, human-computer dialogue has become increasingly common. For example, intelligent robots can engage in conversations with users and respond to their questions via text.

[0003] In current technologies, users typically input their questions into text input boxes on the chat page and then send the text to the AI ​​chatbot, which then responds. This method of manually inputting questions by the user results in low human-computer interaction efficiency during the chatbot process. Summary of the Invention

[0004] This application provides a message interaction method and related equipment. The related equipment may include message interaction devices, electronic devices, computer-readable storage media, and computer program products, which can improve the efficiency of human-computer message interaction in a target session.

[0005] This application provides a message interaction method, including:

[0006] The message interaction client displays the session page of the target session. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent.

[0007] In response to a trigger operation on the target message interaction control, the target recommended message content corresponding to the target message interaction control is sent to the client corresponding to the session member of the target session;

[0008] Receive and display the response content sent by the intelligent virtual session object in response to the target recommended message content.

[0009] Accordingly, embodiments of this application provide a message interaction device, including:

[0010] The first display unit is used to display the session page of the target session in the message interaction client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent.

[0011] The sending unit is configured to, in response to a trigger operation on the target message interaction control, send the target recommendation message content corresponding to the target message interaction control to the client corresponding to the session member of the target session;

[0012] The second display unit is used to receive and display the response content sent by the intelligent virtual session object to the target recommended message content.

[0013] Optionally, in some embodiments of this application, the sending unit may include a first display subunit and a first sending subunit, as follows:

[0014] The first display subunit is used to display the target recommended message content corresponding to the target message interaction control in the message content input area in response to a trigger operation on the target message interaction control;

[0015] The first sending subunit is configured to, in response to the completion operation of editing the target recommendation message content, send the edited target recommendation message content to the client corresponding to the session member of the target session.

[0016] Optionally, in some embodiments of this application, the recommended message content area is located above the message content input area.

[0017] Optionally, in some embodiments of this application, the message interaction controls in the recommended message content area are arranged in a target order, which is determined based on the historical message content of the real session object and the content type of the recommended message content.

[0018] Optionally, in some embodiments of this application, the message interaction controls in the recommended message content area are arranged at intervals.

[0019] Optionally, in some embodiments of this application, the recommended message content is determined based on the historical message content of the real conversation object and the comment information corresponding to the historical message content.

[0020] Optionally, in some embodiments of this application, the session page further includes a comment control for the reply content; the message interaction device may also include a comment unit, which is used to comment on the reply content in response to a trigger operation on the comment control.

[0021] Optionally, in some embodiments of this application, the second display unit may include a first receiving subunit, a conversion subunit, and a second display subunit, as follows:

[0022] The first receiving subunit is used to receive the initial response content sent by the intelligent virtual session object, the initial response content including the feature entity corresponding to the target recommendation message content;

[0023] The conversion subunit is used to convert the feature entity into response content in the target message format;

[0024] The second display subunit is used to display the reply content under the target message format.

[0025] Optionally, in some embodiments of this application, the second display unit may include a second receiving subunit, a third display subunit, and a fourth display subunit, as follows:

[0026] The second receiving subunit is used to receive the response content sent by the intelligent virtual session object in response to the target recommendation message content;

[0027] The third display subunit is used to display the reply content on the conversation page. The reply content includes at least one piece of related information, which is information associated with the target recommended message content.

[0028] The fourth display subunit is used to display the edited response content in response to the editing operation on the associated information.

[0029] Optionally, in some embodiments of this application, the message interaction device may further include a fifth display subunit and a second sending subunit, as follows:

[0030] The fifth display subunit is used to display the input message content in the message content input area in response to a content input operation on the message content input area;

[0031] The second sending subunit is used to send the edited message content to the client corresponding to the session member of the target session in response to the operation of completing the editing of the message content.

[0032] Optionally, in some embodiments of this application, the message interaction device may further include a selection unit, which is used to select recommended message content. Specifically, the selection unit may include a first determining subunit, a second determining subunit, and a selection subunit, as follows:

[0033] The first determining subunit is used to determine the first recommendation score corresponding to each candidate recommendation message content based on the historical message content of the real session object;

[0034] The second determining subunit is used to determine the second recommendation score corresponding to each candidate recommendation message content based on the content type of each candidate recommendation message content.

[0035] A selection sub-unit is used to select recommended message content from each candidate recommended message content based on the first recommendation score and the second recommendation score.

[0036] Optionally, in some embodiments of this application, the selection subunit may be specifically used to obtain the weights corresponding to the first recommendation score and the second recommendation score; for each candidate recommendation message content, based on the weights, the first recommendation score and the second recommendation score are fused to obtain the target recommendation score corresponding to each candidate recommendation message content; and based on the target recommendation score, recommendation message content is selected from each candidate recommendation message content.

[0037] Optionally, in some embodiments of this application, the first determining subunit may be specifically used to count the frequency information of each candidate recommended message content in the historical message content of the real session object; and determine the first recommendation score corresponding to the candidate recommended message content based on the frequency information.

[0038] Optionally, in some embodiments of this application, the message interaction device may further include a determining unit, which is used to determine the content of candidate recommended messages. Specifically, the determining unit may include an acquisition subunit, an analysis subunit, and a third determining subunit, as follows:

[0039] The acquisition subunit is used to acquire the content of the first candidate recommendation message associated with the current session time;

[0040] The analysis subunit is used to perform event distribution analysis on the historical message content of the real session object based on the current session time to obtain the second candidate recommended message content;

[0041] The third determining subunit is used to determine the candidate recommendation message content based on the first candidate recommendation message content and the second candidate recommendation message content.

[0042] Optionally, in some embodiments of this application, the analysis subunit may be specifically used to select target historical message content whose session time falls within a target time period from the historical message content of the real session object based on the current session time; perform feature extraction on the target historical message content to obtain at least one feature entity corresponding to the target historical message content; count the frequency of each feature entity in the target historical message content to obtain frequency information of each feature entity; and select second candidate recommended message content from each feature entity of the target historical message content based on the frequency information.

[0043] An electronic device provided in this application includes a processor and a memory. The memory stores multiple instructions, and the processor loads the instructions to execute the steps in the message interaction method provided in this application.

[0044] This application also provides a computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the steps in the message interaction method provided in this application.

[0045] Furthermore, embodiments of this application also provide a computer program product, including a computer program or instructions, which, when executed by a processor, implement the steps in the message interaction method provided in embodiments of this application.

[0046] This application provides a message interaction method and related device, which can display a session page of a target session in a message interaction client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to recommended message content. The message content input area is used to input the message content to be sent. In response to a trigger operation on the target message interaction control, the recommended message content corresponding to the target message interaction control is sent to the client corresponding to the session member of the target session. The method also receives and displays the reply content sent by the intelligent virtual session object in response to the target recommended message content. This embodiment allows direct input of recommended message content based on a trigger operation on the message interaction control in the recommended message content area, eliminating the need for manual input and improving the efficiency of human-computer message interaction in the target session. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1a This is a schematic diagram of a scenario for the message interaction method provided in an embodiment of this application;

[0049] Figure 1b This is a flowchart of the message interaction method provided in the embodiments of this application;

[0050] Figure 1c This is an explanatory diagram of the message interaction method provided in the embodiments of this application;

[0051] Figure 1d This is another illustrative diagram of the message interaction method provided in the embodiments of this application;

[0052] Figure 1e This is a schematic diagram of a page illustrating the message interaction method provided in an embodiment of this application;

[0053] Figure 1f This is another illustrative diagram of the message interaction method provided in the embodiments of this application;

[0054] Figure 2 This is another flowchart of the message interaction method provided in the embodiments of this application;

[0055] Figure 3 This is a schematic diagram of the structure of the message interaction device provided in the embodiments of this application;

[0056] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0057] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0058] This application provides a message interaction method and related equipment. The related equipment may include a message interaction device, an electronic device, a computer-readable storage medium, and a computer program product. Specifically, the message interaction device may be integrated into an electronic device, which may be a terminal or a server, etc.

[0059] It is understood that the message interaction method in this embodiment can be executed on a terminal, on a server, or jointly by a terminal and a server. The above examples should not be construed as limiting this application.

[0060] like Figure 1a As shown, the message interaction method jointly executed by the terminal and the server is taken as an example. The message interaction system provided in this application embodiment includes a terminal 10 and a server 11, etc.; the terminal 10 and the server 11 are connected through a network, such as through a wired or wireless network, etc., wherein the message interaction device can be integrated into the terminal.

[0061] Terminal 10 can be used to: display a session page of a target session in a message interaction client, the session page including a message content input area and a recommended message content area, the session members of the target session including intelligent virtual session objects and real session objects, the recommended message content area including at least one message interaction control corresponding to recommended message content, the message content input area for inputting message content to be sent, and the recommended message content being determined based on the historical message content of the real session object; in response to a trigger operation on the target message interaction control, sending the target recommended message content corresponding to the target message interaction control to the client corresponding to the session member of the target session; and receiving and displaying the reply content sent by the intelligent virtual session object in response to the target recommended message content. Terminal 10 may include a mobile phone, smart TV, tablet computer, laptop computer, or personal computer (PC, Personal Computer), etc. A client may also be set on terminal 10, which may be an application client or a browser client, etc.

[0062] The server 11 can be used to: determine a first recommendation score corresponding to each candidate recommended message content based on the historical message content of the real session object; determine a second recommendation score corresponding to each candidate recommended message content based on the content type of each candidate recommended message content; select recommended message content from each candidate recommended message content based on the first recommendation score and the second recommendation score; and send the selected recommended message content to the terminal 10. The server 11 can be a single server, a server cluster composed of multiple servers, or a cloud server. In the message interaction method or apparatus disclosed in this application, multiple servers can form a blockchain, and the servers are nodes on the blockchain.

[0063] The steps of selecting recommended message content by server 11 can also be performed by terminal 10.

[0064] The message interaction method provided in this application relates to natural language processing in the field of artificial intelligence.

[0065] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to have perception, reasoning, and decision-making capabilities. AI technology is a comprehensive discipline involving a wide range of fields, encompassing both hardware and software technologies. AI software technology mainly includes computer vision, speech processing, natural language processing, as well as machine learning / deep learning, autonomous driving, and intelligent transportation.

[0066] Natural Language Processing (NLP) is an important area within computer science and artificial intelligence. It studies the theories and methods for enabling effective communication between humans and computers using natural language. NLP is a science that integrates linguistics, computer science, and mathematics. Therefore, research in this field involves natural language—the language people use in daily life—and thus it has a close connection with linguistic research. NLP technologies typically include text processing, semantic understanding, machine translation, question answering, and knowledge graphs.

[0067] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the preferred order of the embodiments.

[0068] This embodiment will be described from the perspective of a message interaction device, which can be integrated into an electronic device, such as a server or a terminal.

[0069] The message interaction method of this application embodiment can be applied to various intelligent question-and-answer scenarios. This embodiment can be applied to various scenarios such as cloud technology, artificial intelligence, smart transportation, and assisted driving.

[0070] like Figure 1b As shown, the specific flow of this message interaction method can be as follows:

[0071] 101. Display the session page of the target session in the message interaction client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent.

[0072] The messaging client can be used for messaging; it can be an application client or a browser client. In some embodiments, the messaging client can be a mini-program, specifically a smart dialogue mini-program, which can provide human-computer dialogue services such as casual conversation and recording of daily life.

[0073] The target session can specifically include at least two session members. The session page can be used to display the content of sent or received messages.

[0074] A mini program is an application or script file developed using a specific programming language that can be used without downloading or installing. Mini programs typically run on an application platform; users can scan the mini program's QR code using the application's scanning function to open it, or they can search for the mini program within the application. Mini programs are characterized by low development difficulty, small memory footprint, practical functionality, and ease of use, as they do not require manual installation on the terminal's operating system.

[0075] In this context, the intelligent virtual conversation object, also known as an intelligent agent, can be viewed as an intelligent chatbot. In a dialogue system, the intelligent agent can generate responses to the user's input messages. The dialogue system is a computer system designed for dialogue with humans. The real conversation object can be a real user, not a virtual conversation member.

[0076] The message interaction control can be displayed in various ways, and this embodiment does not limit this. For example, the message interaction control can be displayed as a "bubble" on the session page. In response to the triggering operation of the message interaction control, corresponding recommendation message content can be sent to the client corresponding to the session member of the target session.

[0077] Specifically, the recommended message content can be determined based on the historical message content of the real conversation object. This real conversation object can be the real conversation object corresponding to the message interaction client, or it can refer to other real conversation objects besides the one corresponding to the message interaction client; this embodiment does not impose any restrictions. The recommended message content is the input recommendation provided to the user by the message interaction system (specifically, it can be an intelligent dialogue mini-program).

[0078] The message content input area can be used to input the message content to be sent. The format of the message content is not limited, and it can be text, audio, image, video, etc.

[0079] In current technologies, users need to input messages in the text input box (specifically, the message content input area) on the conversation page to interact with the intelligent chatbot. Manually inputting messages is inefficient. In this embodiment, a recommended message content area can be added to the conversation page. This area includes message interaction controls corresponding to the recommended message content, specifically clickable text "bubbles." In this way, text is recommended to the user, who can click on the text "bubbles" of interest as input to send the "bubble" text to the client corresponding to the conversation member of the target conversation, thereby increasing interaction efficiency.

[0080] Optionally, in this embodiment, the recommended message content area is located above the message content input area.

[0081] In other embodiments, the recommended message content area may also be located below the message content input area; this embodiment does not impose any restrictions on this.

[0082] Optionally, in this embodiment, the message interaction controls in the recommended message content area are arranged in a target order, which is determined based on the historical message content of the real conversation object and the content type of the recommended message content.

[0083] Specifically, the target order can be determined based on the target recommendation score of the recommended message content corresponding to each message interaction control, and the target recommendation score of the recommended message content can be determined based on the historical message content of the real session object and the content type of the recommended message content.

[0084] Specifically, the recommended message content corresponding to each message interaction control can be sorted based on the target recommendation score, such as sorting from largest to smallest, to obtain the sorted recommended message content. Based on the order in the sorted recommended message content, the arrangement order of each message interaction control in the recommended message content area can be determined.

[0085] Optionally, in this embodiment, the message interaction controls in the recommended message content area are arranged at intervals.

[0086] In a specific scenario, such as Figure 1cAs shown, the recommended message content area is located above the message content input area. This area includes five interactive controls, or text "bubbles," corresponding to five recommended message contents: "Investment Income," "Early Rising," "Citrus Fruits," "Communication," and "Dinner." Users can click to select the text within a bubble as the interactive text, i.e., the selected target recommended message content, and send that text to the client corresponding to a member of the target session. Additionally, in some embodiments, the bubble text can be edited before being sent to the client corresponding to a member of the target session.

[0087] Optionally, in this embodiment, the message interaction method may further include:

[0088] Based on the historical message content of the real session object, determine the first recommendation score corresponding to each candidate recommendation message content;

[0089] The second recommendation score for each candidate recommendation message is determined based on its content type.

[0090] Based on the first recommendation score and the second recommendation score, recommended message content is selected from each candidate recommended message content.

[0091] The real session object can be the real session object corresponding to the message interaction client, or it can refer to other real session objects besides the real session object corresponding to the message interaction client. This embodiment does not limit this.

[0092] The content type of the candidate recommendation message can be the event type corresponding to that message. There are various content types, such as food, commuting, expenses, and income; this embodiment does not impose any limitations on this. This embodiment can set different second recommendation scores for recommendation message content of different content types.

[0093] Specifically, for new users of this messaging client, i.e., users using the messaging client for the first time, there is no historical message content. In this case, the real session object here can refer to other real session objects with interaction history. For users with historical message content, the real session object here can include the real session object corresponding to this messaging client, as well as other real session objects with interaction history.

[0094] Optionally, in this embodiment, the step "selecting recommended message content from each candidate recommended message content based on the first recommendation score and the second recommendation score" may include:

[0095] Obtain the weights corresponding to the first recommendation score and the second recommendation score;

[0096] For each candidate recommended message content, based on the weight, the first recommendation score and the second recommendation score are fused to obtain the target recommendation score corresponding to each candidate recommended message content;

[0097] Based on the target recommendation score, recommended message content is selected from each candidate recommended message content.

[0098] The weights corresponding to the first recommendation score and the second recommendation score can be set according to the actual situation, and this embodiment does not impose any restrictions on this. For example, the weights corresponding to the first recommendation score and the second recommendation score can be determined according to the specific scenario requirements, whether the focus is on historical message content or on the content type of candidate recommended message content.

[0099] Specifically, this fusion method can be weighted calculation, etc.

[0100] In some embodiments, candidate recommended message content with a target recommendation score greater than a preset value can be selected as recommended message content. This preset value can be set according to actual conditions. In other embodiments, the candidate recommended message content can be sorted according to the target recommendation score, such as sorting from largest to smallest, to obtain sorted candidate recommended message content. The top n candidate recommended message contents in the sorted candidate recommended message content are then selected as recommended message content.

[0101] Optionally, the step "determine the first recommendation score corresponding to each candidate recommendation message content based on the historical message content of the real session object" may include:

[0102] For each candidate recommended message content, the frequency information of the candidate recommended message content appearing in the historical message content of the real session object is statistically analyzed;

[0103] Based on the frequency information, a first recommendation score is determined corresponding to the candidate recommendation message content.

[0104] Among them, the frequency information of the candidate recommended message content in the historical message content of the real session object can be the number of times the candidate recommended message content appears in the historical message content of the real session object. This number represents the probability of the candidate recommended message content appearing in the historical message content of the real session object.

[0105] The higher the frequency, the higher the first recommendation score of the candidate recommended message content; conversely, the lower the frequency, the lower the first recommendation score of the candidate recommended message content.

[0106] For users with an interaction history, for each candidate recommended message, the first frequency of the candidate recommended message in the historical message content of the real session object corresponding to the message interaction client can be counted. Based on the first frequency, the self-recommendation score corresponding to the candidate recommended message is determined. Then, the second frequency of the candidate recommended message in the historical message content of other real session objects can be counted. Based on the second frequency, the associated recommendation score corresponding to the candidate recommended message is determined. Thus, based on the self-recommendation score and the associated recommendation score corresponding to the candidate recommended message, the first recommendation score of the candidate recommended message is determined.

[0107] In a specific scenario, the target recommendation score for each candidate recommended message content (which can be considered as a candidate bubble) can be calculated. Then, the candidate recommended message contents are sorted from highest to lowest based on their target recommendation scores. The top N candidate recommended message contents are then used as the recommended message contents and presented in the recommended message content area, where N is the number of bubble texts to be displayed to the user. The calculation of the target recommendation score for each candidate recommended message content can be based on the following strategies:

[0108] A. Overall Strategy: Calculate the frequency of candidate recommended message content in the historical message content of all real conversation objects, and determine the recommendation score s1 of the candidate recommended message content based on the frequency information; where all real conversation objects can specifically refer to real conversation objects that have corresponding historical message content.

[0109] B. Habitual Strategy: Calculate the frequency of the candidate recommended message content in the historical message content of the real session object corresponding to the message interaction client, and determine the recommendation score s2 of the candidate recommended message content based on the frequency information;

[0110] C. Content type strategy: Determine the recommendation score s3 of the candidate recommended message content based on its content type;

[0111] Next, obtain the weights corresponding to the three strategies mentioned above. For example, the weight of the overall strategy is w1, the weight of the habit strategy is w2, and the weight of the content type strategy is w3. It can be understood that the weights of each strategy can be set according to the specific needs of the scenario. Based on the weights corresponding to each strategy, the target recommendation score for the candidate recommended message content can be calculated as s = w1*s1 + w2*s2 + w3*s3. Therefore, the candidate recommended message content can be ranked according to the target recommendation score.

[0112] In this embodiment, the candidate recommendation message content can be retrieved from candidate feature entities, and the specific retrieval process can be referred to the following description.

[0113] In this embodiment, the candidate feature entity can be a word or a character, etc., and this is not limited to this. Specifically, the candidate feature entity can be obtained by preprocessing the historical message content of the real conversation object, such as event distribution analysis.

[0114] Optionally, in this embodiment, the message interaction method may further include:

[0115] Retrieve the content of the first candidate recommended message associated with the current session time;

[0116] Based on the current session time, the event distribution analysis is performed on the historical message content of the real session object to obtain the second candidate recommended message content;

[0117] The content of the candidate recommendation message is determined based on the content of the first candidate recommendation message and the content of the second candidate recommendation message.

[0118] For example, if the current conversation time is in the morning, text such as "morning tea", "breakfast", "morning run", and "subway" can be retrieved as candidate bubbles, which are the first candidate recommended message content.

[0119] Specifically, the event distribution analysis of historical message content can involve named entity recognition, intent classification, and sentiment classification of historical message content to extract feature entities, and then perform statistical analysis on each feature entity to select the second candidate recommended message content.

[0120] For example, a certain intelligent dialogue mini-program is mainly used to record users' expenses, income, mood, habits, etc. It is possible to obtain the historical message content corresponding to the actual conversation objects of this intelligent dialogue mini-program, perform named entity recognition on it, and extract corresponding feature entities such as numbers, events, and times from the historical message content through named entity recognition.

[0121] Named Entity Recognition (NER) specifically identifies entities with specific meanings in text information, mainly including names of people, places, organizations, proper nouns, as well as text such as time, quantity, currency, and ratio values.

[0122] Specifically, both the first and second candidate recommendation message contents can be designated as candidate recommendation message contents. These candidate recommendation message contents can be viewed as bubble text that might be of interest to potential users.

[0123] Optionally, the step "based on the current session time, perform event distribution analysis on the historical message content of the real session object to obtain the second candidate recommended message content" may include:

[0124] Based on the current session time, select the target historical message content whose session time falls within the target time period from the historical message content of the real session object;

[0125] Feature extraction is performed on the target historical message content to obtain at least one feature entity corresponding to the target historical message content;

[0126] The frequency of each feature entity appearing in the target historical message content is statistically analyzed to obtain the frequency information of each feature entity.

[0127] Based on the frequency information, a second candidate recommended message content is selected from each feature entity of the target historical message content.

[0128] The target time period can be determined based on the current session time and is associated with it. For example, if the current session time is 8:30 a.m., the target time period can be set to the period from 7:00 a.m. to 10:00 a.m.

[0129] In some embodiments, feature entities with frequency information higher than a preset value can be selected as the second candidate recommended message content. This preset value can be set according to the actual situation. In other embodiments, based on frequency information, each feature entity in the target historical message content can be sorted, such as sorted from largest to smallest, to obtain sorted feature entities. The top n feature entities in the sorted feature entities are then determined as the second candidate recommended message content.

[0130] In some embodiments, if the real session object corresponding to the message interaction client does not have historical message content, event distribution analysis can be performed on the historical message content of other real session objects to obtain second candidate recommended message content. Specifically, features can be extracted from the historical message content of other real session objects to obtain corresponding feature entities. Then, the frequency of each feature entity appearing in the historical message content of other real session objects can be counted, and second candidate recommended message content can be selected from each feature entity based on this frequency. For example, by statistically determining the feature entities such as "taxi," "cycling," and "subway" that the user most frequently inputs during the time period corresponding to the current session time (i.e., the target time period), these feature entities can be identified as second candidate recommended message content.

[0131] In other embodiments, if the real session object corresponding to the message interaction client has historical message content, its second candidate recommended message content can be selected from two aspects. First, it can be selected based on the event distribution strategy corresponding to the real session object within the target time period. Specifically, event distribution analysis can be performed on the historical message content of the real session object within the target time period to obtain the second candidate recommended message content. Second, it can be selected based on the event distribution strategy corresponding to other real session objects within the target time period. Specifically, event distribution analysis can be performed on the historical message content of other real session objects within the target time period to obtain the second candidate recommended message content.

[0132] In a specific scenario, such as Figure 1d The diagram shows the process of selecting recommended message content in the recommended message content area, described in detail below:

[0133] a. Initialization: Preprocess the historical message content of the real conversation object, such as named entity recognition, intent classification, sentiment classification, etc., and extract feature entities such as numbers (such as income and expenditure amounts), events (such as moods), and time from the historical message content.

[0134] b. Recall candidate recommended message content, which can be divided into two cases based on whether the real session object corresponding to the message interaction client has historical message content.

[0135] i) Cold start, which is when the real session object corresponding to the message interaction client does not have historical message content, there are two strategies for recalling candidate recommended message content. The first is a time strategy, which can obtain candidate bubbles associated with the current session time, i.e., the first candidate recommended message content; the second is based on the current session time, and can be obtained by performing event distribution analysis on the historical message content of other real session objects. Specifically, the feature entities obtained in step a can be statistically analyzed to obtain the second candidate recommended message content.

[0136] ii) For historical users, i.e., the real session objects corresponding to the message interaction clients, where historical message content exists, there are three strategies for recalling candidate recommended message content. The first is a time-based strategy, which specifically retrieves candidate bubbles associated with the current session time, i.e., the first candidate recommended message content. The second is an event distribution strategy for the user within the target time period, which specifically analyzes the event distribution of the historical message content of the real session object based on the current session time to obtain the second candidate recommended message content. The third is an event distribution strategy for other users within the target event period, which specifically analyzes the event distribution of the historical message content of other real session objects based on the current session time to obtain the second candidate recommended message content.

[0137] c. Based on the target recommendation scores of the candidate recommendation message content, sort the candidate recommendation message content, and select the top N candidate recommendation message content as the recommended message content. The target recommendation score for each candidate recommendation message content can be calculated based on the following strategies:

[0138] i) Overall strategy: Calculate the frequency of candidate recommended message content in the historical message content of all real session objects, and determine the recommendation score s1 of the candidate recommended message content based on the frequency information; where all real session objects can specifically refer to real session objects that have corresponding historical message content.

[0139] ii) Habitual strategy: Calculate the frequency of the candidate recommended message content in the historical message content of the real session object corresponding to the message interaction client, and determine the recommendation score s2 of the candidate recommended message content based on the frequency information;

[0140] iii) Content type strategy: Determine the recommendation score s3 of the candidate recommended message content based on its content type;

[0141] Next, obtain the weights corresponding to the three strategies mentioned above. For example, the weight of the overall strategy is w1, the weight of the habit strategy is w2, and the weight of the content type strategy is w3. It can be understood that the weights of each strategy can be set according to the specific needs of the scenario. Based on the weights corresponding to each strategy, the target recommendation score of the candidate recommended message content can be calculated as s = w1*s1 + w2*s2 + w3*s3.

[0142] It should be noted that the recommended message content in the recommended message content area of ​​the conversation page can be specifically calculated and recommended to the user in real time.

[0143] 102. In response to a trigger operation on the target message interaction control, send the target recommendation message content corresponding to the target message interaction control to the client corresponding to the session member of the target session.

[0144] The triggering operation for the target message interaction control can specifically be a click operation or a swipe operation on the target message interaction control, and this embodiment does not limit this.

[0145] In some embodiments, in response to a triggering operation of the target message interaction control, the target recommended message content can be directly sent to the client corresponding to the session member of the target session, and the target recommended message content can be displayed on the session page. In other embodiments, in response to a triggering operation of the target message interaction control, the target recommended message content can be displayed in the message content input area, so that the user can also edit the target recommended message content in the message content input area, and then send the edited target recommended message content to the client corresponding to the session member of the target session.

[0146] Optionally, in this embodiment, the step "in response to a trigger operation on the target message interaction control, sending the target recommendation message content corresponding to the target message interaction control to the client corresponding to the session member of the target session" may include:

[0147] In response to a trigger operation on the target message interaction control, the target recommended message content corresponding to the target message interaction control is displayed in the message content input area;

[0148] In response to the completion of the editing operation for the target recommendation message content, the edited target recommendation message content is sent to the client corresponding to the session member of the target session.

[0149] The message content input area can include a send control, which allows for the editing of the target recommended message content. Specifically, this can be a trigger operation on the send control, such as a click operation.

[0150] Optionally, in this embodiment, the message interaction method may further include:

[0151] In response to a content input operation in the message content input area, the input message content is displayed in the message content input area;

[0152] In response to the completion of the editing operation on the message content, the edited message content is sent to the client corresponding to the session member of the target session.

[0153] In this embodiment, the message content can also be edited directly in the message content input area. The input message content can take many forms, such as text, audio, images, and video; this embodiment does not impose any restrictions on this.

[0154] 103. Receive and display the response content sent by the intelligent virtual session object to the target recommended message content.

[0155] In some embodiments, feature extraction can be performed on the target recommendation message content to obtain the feature entities of the target recommendation message content, and then the corresponding response content can be generated based on the feature entities.

[0156] The response can take many forms, such as text, voice, images, and video, and this embodiment does not limit this.

[0157] Optionally, in this embodiment, the recommended message content is determined based on the historical message content of the real conversation object and the comment information corresponding to the historical message content.

[0158] The comments corresponding to historical messages can include both positive and negative comments.

[0159] The comment information corresponding to historical message content can include comments on the replies to those historical messages. Based on this comment information, the user's satisfaction level with the generated reply can be determined. If the comment is negative, it indicates that the accuracy of the corresponding reply generated by the message interaction client is low, and therefore, recommendations for related message content can be reduced.

[0160] Optionally, in this embodiment, the conversation page further includes a comment control for the reply content; the message interaction method may also include:

[0161] In response to a triggered operation on the comment control, a comment is made on the reply content.

[0162] The triggering operation for the comment control can specifically be a click operation or a swipe operation on the comment control, and this embodiment does not limit this.

[0163] Specifically, comment controls can include positive comment controls and negative comment controls, such as... Figure 1e As shown, below the reply "Did you go to bed early as usual today? Going to bed early is a good habit" on the conversation page, there is a comment control for this reply. When a positive comment control is triggered, a positive comment can be made on the reply; when a negative comment control is triggered, a negative comment can be made on the reply.

[0164] Optionally, in this embodiment, the step of "receiving and displaying the response content sent by the intelligent virtual session object to the target recommended message content" may include:

[0165] Receive initial response content sent by the intelligent virtual session object, wherein the initial response content includes the feature entity corresponding to the target recommendation message content;

[0166] Convert the feature entity into reply content in the target message format;

[0167] Display the response content in the target message format.

[0168] The target message format may specifically be in the form of a card, but this embodiment does not limit this.

[0169] In one specific embodiment, the feature entities corresponding to the target recommendation message content include "300 yuan", "hair dryer", "bill", and "consumption". These feature entities can be converted into card-style reply content and presented to the user, such as... Figure 1e As shown, the response content is displayed in card format.

[0170] Optionally, in this embodiment, the step of "receiving and displaying the response content sent by the intelligent virtual session object to the target recommended message content" may include:

[0171] Receive the response content sent by the intelligent virtual session object in response to the target recommended message content;

[0172] The reply content is displayed on the conversation page. The reply content includes at least one piece of related information, which is information associated with the target recommended message content.

[0173] In response to the editing operation on the associated information, the edited reply content is displayed.

[0174] The editing operation of related information can specifically be a selection operation of related information or a deletion operation of related information, etc., and this embodiment does not limit it.

[0175] like Figure 1f As shown, the recommended message content area includes message interaction controls corresponding to five recommended message contents: "Dinner", "Exercise", "Alcohol", "Drinking", and "Transportation". When a user triggers an operation on the message interaction control corresponding to the recommended message content "Alcohol", the recommended message content "Alcohol" is sent to the client corresponding to the session member of the target session, and the recommended message content "Alcohol" is displayed on the session page.

[0176] Simultaneously, based on the recommended message content "wine" sent by the messaging client, a corresponding reply can be generated and sent to the messaging client via a smart virtual conversation object, thereby being displayed on the conversation page, such as... Figure 1f As shown, the reply contains several related information items associated with the recommended message content "alcohol," namely "drinking," "alcohol," and the category "habit."

[0177] Users can edit the related information "drinking" in the reply content, changing the keyword from "drinking" to "alcohol". This editing process can specifically involve selecting the related information "alcohol" (such as clicking). Furthermore, users can delete the related information "habit" in the reply content or change it to another category; this embodiment does not impose restrictions on this.

[0178] The message interaction method provided in this application can prompt users with recommended message content displayed on the conversation page, indicating the areas of expertise of the intelligent virtual conversation object. This guides users to interact within those areas of expertise. For example, if an intelligent dialogue mini-program excels at recording daily life events, specifically in areas like photos, exercise, diet, and budgeting, text "bubbles" can be used to guide user interaction. This prevents users from blindly trying areas the mini-program is not good at, thus improving the user experience. Simultaneously, it also guides new users in understanding and using the product.

[0179] As described above, this embodiment can display the session page of a target session in a message interaction client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent. In response to a trigger operation on the target message interaction control, the target recommended message content corresponding to the target message interaction control is sent to the client corresponding to the session member of the target session. The response content sent by the intelligent virtual session object to the target recommended message content is received and displayed. This embodiment allows direct input of recommended message content based on the trigger operation of the message interaction control in the recommended message content area, eliminating the need for manual input and improving the efficiency of human-computer message interaction in the target session.

[0180] Based on the method described in the preceding embodiments, the following will provide a more detailed explanation by taking the specific integration of the message interaction device into a terminal as an example.

[0181] This application provides a message interaction method, such as... Figure 2 As shown, the specific flow of this message interaction method can be as follows:

[0182] 201. The terminal displays the session page of the target session in the message interaction client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent. The recommended message content is determined based on the historical message content of the real session object.

[0183] In some embodiments, the messaging client can be a mini-program, specifically a smart dialogue mini-program, which can provide human-computer dialogue services such as casual conversation and life recording.

[0184] In this context, the intelligent virtual conversation object, also known as an intelligent agent, can be viewed as an intelligent chatbot. In a dialogue system, the intelligent agent can generate responses to the user's input messages. The dialogue system is a computer system designed for dialogue with humans. The real conversation object can be a real user, not a virtual conversation member.

[0185] The message interaction control can be displayed in various ways, and this embodiment does not limit this. For example, the message interaction control can be displayed as a "bubble" on the session page. In response to the triggering operation of the message interaction control, corresponding recommendation message content can be sent to the client corresponding to the session member of the target session.

[0186] The recommended message content is determined based on the historical message content of the real conversation object. This real conversation object can be the real conversation object corresponding to the message interaction client, or it can refer to other real conversation objects besides the one corresponding to the message interaction client; this embodiment does not impose any restrictions. The recommended message content is the input recommendation provided to the user by the message interaction system (specifically, it can be a smart dialogue mini-program).

[0187] The message content input area can be used to input the message content to be sent. The format of the message content is not limited, and it can be text, audio, image, video, etc.

[0188] Optionally, in this embodiment, the message interaction method may further include:

[0189] Based on the historical message content of the real session object, determine the first recommendation score corresponding to each candidate recommendation message content;

[0190] The second recommendation score for each candidate recommendation message is determined based on its content type.

[0191] Based on the first recommendation score and the second recommendation score, recommended message content is selected from each candidate recommended message content.

[0192] The content type of the candidate recommendation message can be the event type corresponding to that message. There are various content types, such as food, commuting, expenses, and income; this embodiment does not impose any limitations on this. This embodiment can set different second recommendation scores for recommendation message content of different content types.

[0193] Specifically, for new users of this messaging client, i.e., users using the messaging client for the first time, there is no historical message content. In this case, the real session object here can refer to other real session objects with interaction history. For users with historical message content, the real session object here can include the real session object corresponding to this messaging client, as well as other real session objects with interaction history.

[0194] Optionally, in this embodiment, the step "selecting recommended message content from each candidate recommended message content based on the first recommendation score and the second recommendation score" may include:

[0195] Obtain the weights corresponding to the first recommendation score and the second recommendation score;

[0196] For each candidate recommended message content, based on the weight, the first recommendation score and the second recommendation score are fused to obtain the target recommendation score corresponding to each candidate recommended message content;

[0197] Based on the target recommendation score, recommended message content is selected from each candidate recommended message content.

[0198] Optionally, the step "determine the first recommendation score corresponding to each candidate recommendation message content based on the historical message content of the real session object" may include:

[0199] For each candidate recommended message content, the frequency information of the candidate recommended message content appearing in the historical message content of the real session object is statistically analyzed;

[0200] Based on the frequency information, a first recommendation score is determined corresponding to the candidate recommendation message content.

[0201] For users with an interaction history, for each candidate recommended message, the first frequency of the candidate recommended message in the historical message content of the real session object corresponding to the message interaction client can be counted. Based on the first frequency, the self-recommendation score corresponding to the candidate recommended message is determined. Then, the second frequency of the candidate recommended message in the historical message content of other real session objects can be counted. Based on the second frequency, the associated recommendation score corresponding to the candidate recommended message is determined. Thus, based on the self-recommendation score and the associated recommendation score corresponding to the candidate recommended message, the first recommendation score of the candidate recommended message is determined.

[0202] In this embodiment, the candidate recommendation message content can be retrieved from candidate feature entities, and the specific retrieval process can be referred to the following description.

[0203] In this embodiment, the candidate feature entity can be a word or a character, etc., and this is not limited to this. Specifically, the candidate feature entity can be obtained by preprocessing the historical message content of the real conversation object, such as event distribution analysis.

[0204] Optionally, in this embodiment, the message interaction method may further include:

[0205] Retrieve the content of the first candidate recommended message associated with the current session time;

[0206] Based on the current session time, the event distribution analysis is performed on the historical message content of the real session object to obtain the second candidate recommended message content;

[0207] The content of the candidate recommendation message is determined based on the content of the first candidate recommendation message and the content of the second candidate recommendation message.

[0208] For example, if the current conversation time is in the morning, text such as "morning tea", "breakfast", "morning run", and "subway" can be retrieved as candidate bubbles, which are the first candidate recommended message content.

[0209] Specifically, the event distribution analysis of historical message content can involve named entity recognition, intent classification, and sentiment classification of historical message content to extract feature entities, and then perform statistical analysis on each feature entity to select the second candidate recommended message content.

[0210] Optionally, the step "based on the current session time, perform event distribution analysis on the historical message content of the real session object to obtain the second candidate recommended message content" may include:

[0211] Based on the current session time, select the target historical message content whose session time falls within the target time period from the historical message content of the real session object;

[0212] Feature extraction is performed on the target historical message content to obtain at least one feature entity corresponding to the target historical message content;

[0213] The frequency of each feature entity appearing in the target historical message content is statistically analyzed to obtain the frequency information of each feature entity.

[0214] Based on the frequency information, a second candidate recommended message content is selected from each feature entity of the target historical message content.

[0215] The target time period can be determined based on the current session time and is associated with it. For example, if the current session time is 8:30 a.m., the target time period can be set to the period from 7:00 a.m. to 10:00 a.m.

[0216] 202. In response to the triggering operation of the target message interaction control, the terminal sends the target recommendation message content corresponding to the target message interaction control to the client corresponding to the session member of the target session.

[0217] The triggering operation for the target message interaction control can specifically be a click operation or a swipe operation on the target message interaction control, and this embodiment does not limit this.

[0218] In some embodiments, in response to a triggering operation of the target message interaction control, the target recommended message content can be directly sent to the client corresponding to the session member of the target session, and the target recommended message content can be displayed on the session page. In other embodiments, in response to a triggering operation of the target message interaction control, the target recommended message content can be displayed in the message content input area, allowing the user to edit the target recommended message content in the message content input area before sending the edited target recommended message content to the client corresponding to the session member of the target session.

[0219] Optionally, in this embodiment, the step "in response to a trigger operation on the target message interaction control, sending the target recommendation message content corresponding to the target message interaction control to the client corresponding to the session member of the target session" may include:

[0220] In response to a trigger operation on the target message interaction control, the target recommended message content corresponding to the target message interaction control is displayed in the message content input area;

[0221] In response to the completion of the editing operation for the target recommendation message content, the edited target recommendation message content is sent to the client corresponding to the session member of the target session.

[0222] The message content input area can include a send control, which allows for the editing of the target recommended message content. Specifically, this can be a trigger operation on the send control, such as a click operation.

[0223] 203. The terminal receives and displays the response content sent by the intelligent virtual session object to the target recommended message content.

[0224] In some embodiments, feature extraction can be performed on the target recommendation message content to obtain the feature entities of the target recommendation message content, and then the corresponding response content can be generated based on the feature entities.

[0225] The response can take many forms, such as text, voice, images, and video, and this embodiment does not limit this.

[0226] Optionally, in this embodiment, the step of "receiving and displaying the response content sent by the intelligent virtual session object to the target recommended message content" may include:

[0227] Receive initial response content sent by the intelligent virtual session object, wherein the initial response content includes the feature entity corresponding to the target recommendation message content;

[0228] Convert the feature entity into reply content in the target message format;

[0229] Display the response content in the target message format.

[0230] The target message format may specifically be in the form of a card, but this embodiment does not limit this.

[0231] As can be seen from the above, this embodiment can display the conversation page of the target conversation in the message interaction client on the terminal. The conversation page includes a message content input area and a recommended message content area. The conversation members of the target conversation include intelligent virtual conversation objects and real conversation objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent. The recommended message content is determined based on the historical message content of the real conversation object. In response to the trigger operation of the target message interaction control, the target recommended message content corresponding to the target message interaction control is sent to the client corresponding to the conversation member of the target conversation. The response content sent by the intelligent virtual conversation object to the target recommended message content is received and displayed. This embodiment can directly input the recommended message content based on the trigger operation of the message interaction control in the recommended message content area, without the need for manual input, thus improving the efficiency of human-computer message interaction in the target conversation.

[0232] To better implement the above methods, embodiments of this application also provide a message interaction device, such as... Figure 3 As shown, the message interaction device may include a first display unit 301, a sending unit 302, and a second display unit 303, as follows:

[0233] (1) First display unit 301;

[0234] The first display unit 301 is used to display the session page of the target session in the message interaction client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent.

[0235] Optionally, in some embodiments of this application, the recommended message content area is located above the message content input area.

[0236] Optionally, in some embodiments of this application, the message interaction controls in the recommended message content area are arranged in a target order, which is determined based on the historical message content of the real session object and the content type of the recommended message content.

[0237] Optionally, in some embodiments of this application, the message interaction controls in the recommended message content area are arranged at intervals.

[0238] Optionally, in some embodiments of this application, the recommended message content is determined based on the historical message content of the real conversation object and the comment information corresponding to the historical message content.

[0239] (2) Transmitting unit 302;

[0240] The sending unit 302 is used to send the target recommendation message content corresponding to the target message interaction control to the client corresponding to the session member of the target session in response to a trigger operation on the target message interaction control.

[0241] Optionally, in some embodiments of this application, the sending unit may include a first display subunit and a first sending subunit, as follows:

[0242] The first display subunit is used to display the target recommended message content corresponding to the target message interaction control in the message content input area in response to a trigger operation on the target message interaction control;

[0243] The first sending subunit is configured to, in response to the completion operation of editing the target recommendation message content, send the edited target recommendation message content to the client corresponding to the session member of the target session.

[0244] (3) Second display unit 303;

[0245] The second display unit 303 is used to receive and display the response content sent by the intelligent virtual session object to the target recommended message content.

[0246] Optionally, in some embodiments of this application, the session page further includes a comment control for the reply content; the message interaction device may also include a comment unit, which is used to comment on the reply content in response to a trigger operation on the comment control.

[0247] Optionally, in some embodiments of this application, the second display unit may include a first receiving subunit, a conversion subunit, and a second display subunit, as follows:

[0248] The first receiving subunit is used to receive the initial response content sent by the intelligent virtual session object, the initial response content including the feature entity corresponding to the target recommendation message content;

[0249] The conversion subunit is used to convert the feature entity into response content in the target message format;

[0250] The second display subunit is used to display the reply content under the target message format.

[0251] Optionally, in some embodiments of this application, the second display unit may include a second receiving subunit, a third display subunit, and a fourth display subunit, as follows:

[0252] The second receiving subunit is used to receive the response content sent by the intelligent virtual session object in response to the target recommendation message content;

[0253] The third display subunit is used to display the reply content on the conversation page. The reply content includes at least one piece of related information, which is information associated with the target recommended message content.

[0254] The fourth display subunit is used to display the edited response content in response to the editing operation on the associated information.

[0255] Optionally, in some embodiments of this application, the message interaction device may further include a fifth display subunit and a second sending subunit, as follows:

[0256] The fifth display subunit is used to display the input message content in the message content input area in response to a content input operation on the message content input area;

[0257] The second sending subunit is used to send the edited message content to the client corresponding to the session member of the target session in response to the operation of completing the editing of the message content.

[0258] Optionally, in some embodiments of this application, the message interaction device may further include a selection unit, which is used to select recommended message content. Specifically, the selection unit may include a first determining subunit, a second determining subunit, and a selection subunit, as follows:

[0259] The first determining subunit is used to determine the first recommendation score corresponding to each candidate recommendation message content based on the historical message content of the real session object;

[0260] The second determining subunit is used to determine the second recommendation score corresponding to each candidate recommendation message content based on the content type of each candidate recommendation message content.

[0261] A selection sub-unit is used to select recommended message content from each candidate recommended message content based on the first recommendation score and the second recommendation score.

[0262] Optionally, in some embodiments of this application, the selection subunit may be specifically used to obtain the weights corresponding to the first recommendation score and the second recommendation score; for each candidate recommendation message content, based on the weights, the first recommendation score and the second recommendation score are fused to obtain the target recommendation score corresponding to each candidate recommendation message content; and based on the target recommendation score, recommendation message content is selected from each candidate recommendation message content.

[0263] Optionally, in some embodiments of this application, the first determining subunit may be specifically used to count the frequency information of each candidate recommended message content in the historical message content of the real session object; and determine the first recommendation score corresponding to the candidate recommended message content based on the frequency information.

[0264] Optionally, in some embodiments of this application, the message interaction device may further include a determining unit, which is used to determine the content of candidate recommended messages. Specifically, the determining unit may include an acquisition subunit, an analysis subunit, and a third determining subunit, as follows:

[0265] The acquisition subunit is used to acquire the content of the first candidate recommendation message associated with the current session time;

[0266] The analysis subunit is used to perform event distribution analysis on the historical message content of the real session object based on the current session time to obtain the second candidate recommended message content;

[0267] The third determining subunit is used to determine the candidate recommendation message content based on the first candidate recommendation message content and the second candidate recommendation message content.

[0268] Optionally, in some embodiments of this application, the analysis subunit may be specifically used to select target historical message content whose session time falls within a target time period from the historical message content of the real session object based on the current session time; perform feature extraction on the target historical message content to obtain at least one feature entity corresponding to the target historical message content; count the frequency of each feature entity in the target historical message content to obtain frequency information of each feature entity; and select second candidate recommended message content from each feature entity of the target historical message content based on the frequency information.

[0269] As can be seen from the above, this embodiment can display the session page of the target session in the message interaction client through the first display unit 301. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent. In response to the trigger operation of the target message interaction control, the sending unit 302 sends the target recommended message content corresponding to the target message interaction control to the client corresponding to the session member of the target session. The second display unit 303 receives and displays the reply content sent by the intelligent virtual session object in response to the target recommended message content. This embodiment can directly input the recommended message content based on the trigger operation of the message interaction control in the recommended message content area, without the need for manual input, thus improving the efficiency of human-computer message interaction in the target session.

[0270] This application also provides an electronic device, such as... Figure 4 The diagram shows a schematic representation of the structure of an electronic device according to an embodiment of this application. This electronic device can be a terminal or a server, specifically:

[0271] The electronic device may include components such as a processor 401 with one or more processing cores, a memory 402 with one or more computer-readable storage media, a power supply 403, and an input unit 404. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein:

[0272] The processor 401 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 402, and by calling data stored in the memory 402, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, the processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and applications, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 401.

[0273] The memory 402 can be used to store software programs and modules. The processor 401 executes various functional applications and data processing by running the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

[0274] The electronic device also includes a power supply 403 that supplies power to the various components. Preferably, the power supply 403 can be logically connected to the processor 401 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 403 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0275] The electronic device may also include an input unit 404, which can be used to receive input digital or character information, and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0276] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 401 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 402 according to the following instructions, and the processor 401 runs the applications stored in the memory 402 to realize various functions, as follows:

[0277] The system displays a session page for a target session in a messaging client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to recommended message content. The message content input area is used to input the message content to be sent. In response to a trigger operation on the target message interaction control, the system sends the target recommended message content corresponding to the target message interaction control to the client corresponding to the session member of the target session. The system also receives and displays the reply content sent by the intelligent virtual session object in response to the target recommended message content.

[0278] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0279] As described above, this embodiment can display the session page of a target session in a message interaction client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to the recommended message content. The message content input area is used to input the message content to be sent. In response to a trigger operation on the target message interaction control, the target recommended message content corresponding to the target message interaction control is sent to the client corresponding to the session member of the target session. The response content sent by the intelligent virtual session object to the target recommended message content is received and displayed. This embodiment allows direct input of recommended message content based on the trigger operation of the message interaction control in the recommended message content area, eliminating the need for manual input and improving the efficiency of human-computer message interaction in the target session.

[0280] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0281] Therefore, embodiments of this application provide a computer-readable storage medium storing a plurality of instructions that can be loaded by a processor to execute steps in any of the message interaction methods provided in embodiments of this application. For example, the instructions can execute the following steps:

[0282] The system displays a session page for a target session in a messaging client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to recommended message content. The message content input area is used to input the message content to be sent. In response to a trigger operation on the target message interaction control, the system sends the target recommended message content corresponding to the target message interaction control to the client corresponding to the session member of the target session. The system also receives and displays the reply content sent by the intelligent virtual session object in response to the target recommended message content.

[0283] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0284] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0285] Since the instructions stored in the computer-readable storage medium can execute the steps in any of the message interaction methods provided in the embodiments of this application, the beneficial effects that any of the message interaction methods provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0286] According to one aspect of this application, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various alternative implementations of the above-described message interaction aspects.

[0287] The above provides a detailed description of a message interaction method and related devices provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A message interaction method, characterized in that, include: The presentation describes a session page for a target session in a messaging client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to recommended message content. The message content input area is used to input the message content to be sent. The presentation involves: obtaining first candidate recommended message content associated with the current session time; performing event distribution analysis on the historical message content of the real session objects based on the current session time to obtain second candidate recommended message content; determining candidate recommended message content based on the first and second candidate recommended message contents; determining a first recommendation score for each candidate recommended message content based on the historical message content of the real session objects; determining a second recommendation score for each candidate recommended message content based on its content type; and selecting recommended message content from the candidate recommended message contents based on the first and second recommendation scores. In response to a trigger operation on the target message interaction control, the target recommended message content corresponding to the target message interaction control is sent to the client corresponding to the session member of the target session; Receive and display the response content sent by the intelligent virtual session object in response to the target recommended message content.

2. The method according to claim 1, characterized in that, The step of responding to a trigger operation on a target message interaction control by sending the target recommendation message content corresponding to the target message interaction control to the client corresponding to the session member of the target session includes: In response to a trigger operation on the target message interaction control, the target recommended message content corresponding to the target message interaction control is displayed in the message content input area; In response to the completion of the editing operation for the target recommendation message content, the edited target recommendation message content is sent to the client corresponding to the session member of the target session.

3. The method according to claim 1, characterized in that, The recommended message content area is located above the message content input area.

4. The method according to claim 1, characterized in that, The message interaction controls in the recommended message content area are arranged in a target order, which is determined based on the historical message content of the real conversation object and the content type of the recommended message content.

5. The method according to claim 1, characterized in that, The message interaction controls in the recommended message content area are arranged at intervals.

6. The method according to claim 1, characterized in that, The recommended message content is determined based on the historical message content of the real conversation object and the comment information corresponding to the historical message content.

7. The method according to claim 6, characterized in that, The conversation page also includes a comment control for the reply content; the method further includes: In response to a triggered operation on the comment control, a comment is made on the reply content.

8. The method according to claim 1, characterized in that, The step of receiving and displaying the response content sent by the intelligent virtual session object in response to the target recommended message content includes: Receive initial response content sent by the intelligent virtual session object, wherein the initial response content includes the feature entity corresponding to the target recommendation message content; Convert the feature entity into reply content in the target message format; Display the response content in the target message format.

9. The method according to claim 1, characterized in that, The step of receiving and displaying the response content sent by the intelligent virtual session object in response to the target recommended message content includes: Receive the response content sent by the intelligent virtual session object in response to the target recommended message content; The reply content is displayed on the conversation page. The reply content includes at least one piece of related information, which is information associated with the target recommended message content. In response to the editing operation on the associated information, the edited reply content is displayed.

10. The method according to claim 1, characterized in that, The method further includes: In response to a content input operation in the message content input area, the input message content is displayed in the message content input area; In response to the completion of the editing operation on the message content, the edited message content is sent to the client corresponding to the session member of the target session.

11. The method according to claim 1, characterized in that, The step of selecting recommended message content from each candidate recommended message content based on the first recommendation score and the second recommendation score includes: Obtain the weights corresponding to the first recommendation score and the second recommendation score; For each candidate recommended message content, based on the weight, the first recommendation score and the second recommendation score are fused to obtain the target recommendation score corresponding to each candidate recommended message content; Based on the target recommendation score, recommended message content is selected from each candidate recommended message content.

12. The method according to claim 1, characterized in that, The step of determining the first recommendation score corresponding to each candidate recommendation message content based on the historical message content of the real session object includes: For each candidate recommended message content, the frequency information of the candidate recommended message content appearing in the historical message content of the real session object is statistically analyzed; Based on the frequency information, a first recommendation score is determined corresponding to the candidate recommendation message content.

13. The method according to claim 1, characterized in that, The step of performing event distribution analysis on the historical message content of the real session object based on the current session time to obtain the second candidate recommended message content includes: Based on the current session time, select the target historical message content whose session time falls within the target time period from the historical message content of the real session object; Feature extraction is performed on the target historical message content to obtain at least one feature entity corresponding to the target historical message content; The frequency of each feature entity appearing in the target historical message content is statistically analyzed to obtain the frequency information of each feature entity. Based on the frequency information, a second candidate recommended message content is selected from each feature entity of the target historical message content.

14. A message interaction device, characterized in that, include: The first display unit is used to display the session page of the target session in the message interaction client. The session page includes a message content input area and a recommended message content area. The session members of the target session include intelligent virtual session objects and real session objects. The recommended message content area includes at least one message interaction control corresponding to recommended message content. The message content input area is used to input the message content to be sent. The process involves: obtaining first candidate recommended message content associated with the current session time; performing event distribution analysis on the historical message content of the real session objects based on the current session time to obtain second candidate recommended message content; determining candidate recommended message content based on the first and second candidate recommended message content; determining a first recommendation score for each candidate recommended message content based on the historical message content of the real session objects; determining a second recommendation score for each candidate recommended message content based on the content type of each candidate recommended message content; and selecting recommended message content from the candidate recommended message content based on the first and second recommendation scores. The sending unit is used to send the target recommendation message content corresponding to the target message interaction control to the client corresponding to the session member of the target session in response to a trigger operation on the target message interaction control; The second display unit is used to receive and display the response content sent by the intelligent virtual session object to the target recommended message content.

15. An electronic device, characterized in that, It includes a memory and a processor; the memory stores an application program, and the processor runs the application program within the memory to perform the operations in the message interaction method according to any one of claims 1 to 13.

16. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a plurality of instructions adapted for loading by a processor to perform the steps of the message interaction method according to any one of claims 1 to 13.

17. A computer program product comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the message interaction method according to any one of claims 1 to 13.

Images