Conversation control method, device, equipment and storage medium for instant messaging

Abstract

The present invention discloses a conversation control method, apparatus, device, and storage medium for instant messaging. The method comprises: upon receiving a conversation message sent by a client to a user via a first terminal, determining whether an intelligent conversation mode needs to be enabled based on the operating data of a second terminal currently used by the user; if enabled, performing intent recognition on the conversation message to obtain the client's conversation intent; generating response information for the current conversation based on the conversation intent, user profile, and client profile; and pushing the response information to the first terminal according to the intelligent conversation mode to conduct a conversation with the client. In the present invention, the intelligent conversation mode is enabled when the user currently has difficulty in achieving an autonomous response, and personalized response information is automatically generated based on the client's intent, user, and client profile information, thereby achieving intelligent response, which can effectively improve the intelligence of instant messaging, achieve conversation intelligence, and thus enhance the client's instant messaging conversation experience.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a conversation control method, apparatus, device and storage medium for instant messaging. Background Art

[0002] Instant Messaging (IM) is a real-time communication system that allows two or more people to use the Internet to exchange text messages, files, voice and video in real time. It is the most popular communication method on the Internet, and various instant messaging software provides increasingly rich communication service functions.

[0003] Currently, instant messaging software is primarily used for proactive communication between users. However, with its widespread and intensive use, existing IM software is lacking in intelligence and can no longer meet users' growing demand for intelligent communication, diminishing the user experience. For example, when a user receives a conversation message from another user and is unable to respond promptly due to other commitments, most IM software is unable to intelligently process the conversation, thus losing the immediate meaning of the conversation. Even if some IM software can automatically process the conversation, it often relies on a few pre-edited text responses, which are rudimentary and lack intelligence. Summary of the Invention

[0004] The present invention provides a conversation control method, apparatus, device and storage medium for instant messaging, so as to solve the problem that the instant messaging software is not intelligent enough and the user experience of instant messaging is reduced.

[0005] A method for controlling a session in an instant messaging system is provided, comprising:

[0006] Upon receiving a conversation message sent by a client to a user via a first terminal, determining whether to enable the smart conversation mode based on the operating data of a second terminal currently used by the user;

[0007] If you need to enable the intelligent conversation mode, perform intent recognition on the conversation message to obtain the customer's conversation intent;

[0008] Obtain the customer's customer profile and the user's user profile, and generate response information for the current session based on the session intent, user profile, and customer profile;

[0009] The response information is pushed to the first terminal according to the intelligent conversation mode to conduct a conversation with the customer.

[0010] Optionally, generate response information for the current session based on the session intent, user profile, and customer profile, including:

[0011] Obtain a pre-built intent graph and preset response template data pre-stored in the database. The nodes of the intent graph are the semantic content of the intent, and the edges of the intent graph are the semantic associations between the nodes.

[0012] Determine the target response template from the preset response template data based on the conversation intent and the intent map;

[0013] The target response template is modified according to the user portrait and the customer portrait to obtain response data that matches the user portrait and the customer portrait as the response information of the current session.

[0014] Optionally, the operation data includes an operation status. Determining whether to enable the smart conversation mode based on the operation data of the second terminal currently used by the user includes:

[0015] Determining whether the operating state of the second terminal is a power-on state;

[0016] If the running state of the second terminal is the power-on state, obtaining application data on the second terminal, the application data including the operation log of each application;

[0017] Determine whether to enable smart conversation mode based on the operation logs of each application.

[0018] Optionally, the application data also includes the type of each application. Determining whether to enable the smart conversation mode based on the operation log of each application includes:

[0019] determining, based on the operation logs of each application, whether the second terminal currently has an application in use;

[0020] If the second terminal does not currently have an application in use, determining a target application that the user has operated within a preset time period;

[0021] Determine whether the target application is a first-category application, which includes instant messaging and audio and video playback applications;

[0022] If the target application is not a first-category application, it is determined that the smart session mode needs to be enabled;

[0023] If the target application is a first-category application, it is determined that the smart session mode does not need to be enabled.

[0024] Optionally, after determining whether the smart conversation mode needs to be enabled based on the operating data of the second terminal currently used by the user, the method further includes:

[0025] If the intelligent conversation mode does not need to be enabled, controlling the second terminal to enable the conversation assistance function, and controlling the auxiliary function menu of the conversation assistance function to be displayed in the conversation window of the second terminal;

[0026] receiving an operation instruction of the user on the auxiliary function menu, and feeding back corresponding display data to the second terminal in response to the operation instruction, so that the second terminal displays the display data;

[0027] Receive a user's data selection instruction for display data, and determine the target data selected by the user according to the data selection instruction;

[0028] Generate response information for the current session based on the target data, push the response information to the first terminal, and conduct a conversation with the customer.

[0029] Optionally, the auxiliary function menu includes a briefcase and a quick answer, and the corresponding display data fed back to the second terminal in response to the operation instruction includes:

[0030] If an operation instruction for a quick answer is received from the user, the preset speech data is obtained from the database, and the preset speech data is fed back to the second terminal as display data;

[0031] If a briefcase operation instruction is received from the user, the briefcase data is obtained from the database, and the display screen data of the second terminal is obtained;

[0032] A briefcase display interface is generated according to the display screen data and the briefcase data, and the briefcase display interface is fed back to the second terminal.

[0033] Optionally, the auxiliary function menu includes a customer portrait, and responding to the operation instruction to feed back corresponding display data to the second terminal includes:

[0034] If an operation instruction for a customer portrait is received from the user, the customer portrait data corresponding to the customer is determined, and display screen data of the second terminal is obtained;

[0035] A customer information display interface is generated according to the customer portrait data and the display screen data, and the customer information display interface is fed back to the second terminal as display data.

[0036] Provided is a session control device for instant messaging, comprising:

[0037] a determination module, configured to, upon receiving a conversation message sent by a client to a user via a first terminal, determine whether the smart conversation mode needs to be enabled based on the operating data of a second terminal currently used by the user;

[0038] The recognition module is used to identify the intent of the conversation message and obtain the customer's conversation intent if the intelligent conversation mode needs to be enabled;

[0039] The generation module is used to obtain the customer's customer profile and the user's user profile, and generate the response information of the current session based on the session intention, user profile and customer profile;

[0040] The push module is used to push the response information to the first terminal according to the intelligent conversation mode to conduct a conversation with the customer.

[0041] A computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the above-mentioned conversation control method for instant messaging are implemented.

[0042] A computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and wherein the computer program implements the steps of the above-mentioned conversation control method for instant messaging when executed by a processor.

[0043] In one solution provided by the above-mentioned conversation control method, apparatus, device and storage medium for instant messaging, when a conversation message sent by a client to a user via a first terminal is received, it is determined whether the intelligent conversation mode needs to be enabled based on the operating data of the second terminal currently used by the user; if the intelligent conversation mode needs to be enabled, the conversation message is subjected to intent recognition to obtain the client's conversation intention; the client's customer profile and the user's user profile are obtained, and response information for the current conversation is generated based on the conversation intention, user profile and customer profile; the response information is pushed to the first terminal according to the intelligent conversation mode to conduct a conversation with the client. In the present invention, the conversation mode is determined by the operating data of the second terminal used by the user, and the intelligent conversation mode is enabled when the user currently has difficulty in achieving an autonomous response, automatically identifying the intent of the client's conversation message, and then automatically generating personalized response information based on the client's intention, user and client profile information, thereby achieving intelligent response, which can effectively improve the intelligence of instant messaging and achieve conversation intelligence, thereby improving the client's instant messaging conversation experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments of the present invention. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0045] Figure 1 This is a schematic diagram of an application environment of a session control method for instant messaging according to an embodiment of the present invention;

[0046] Figure 2This is a flow chart of a session control method for instant messaging according to an embodiment of the present invention;

[0047] Figure 3 yes Figure 2 A schematic diagram of an implementation flow of step S10;

[0048] Figure 4 yes Figure 2 A schematic diagram of an implementation flow of step S30;

[0049] Figure 5 is another flow chart of a session control method for instant messaging according to an embodiment of the present invention;

[0050] Figure 6 is a schematic diagram of a session control device for instant messaging according to an embodiment of the present invention;

[0051] Figure 7 It is a structural diagram of a computer device in one embodiment of the present invention.

[0052] Figure 8 FIG. 2 is another structural diagram of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0053] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0054] The instant messaging session control method provided by the embodiment of the present invention can be applied in the following situations: Figure 1 In the application environment, the first terminal is the terminal device used by the client corresponding to the user, the second terminal is the terminal device used by the user, and the server represents the server of the instant messaging software (instant messaging application). That is, the second terminal is the terminal device bound to or logged in by the user's instant messaging account; the first terminal is the terminal device bound to or logged in by the client's (the user's client or friend) instant messaging account. When the user and the client conduct instant messaging, the third-party server provides instant messaging session services for both. Among them, the first terminal and the second terminal communicate with the server (server) through the network respectively. When the client sends a session message to the user, the server sends the session message sent by the client through the first terminal to the user's second terminal.

[0055] When receiving a conversation message sent by a customer to a user via a first terminal, determine whether it is necessary to enable the intelligent conversation mode based on the operating data of the second terminal currently used by the user; if it is necessary to enable the intelligent conversation mode, perform intent recognition on the conversation message to obtain the customer's conversation intention; obtain the customer's customer portrait and the user's user portrait, and generate response information for the current conversation based on the conversation intention, user portrait, and customer portrait; push the response information to the first terminal according to the intelligent conversation mode to conduct a conversation with the customer. In the present invention, the conversation mode is determined by the operating data of the second terminal used by the user, and the intelligent conversation mode is enabled when it is currently difficult for the user to achieve autonomous response. The intention of the customer's conversation message is automatically recognized, and then personalized response information is automatically generated based on the customer's intention, user, and customer portrait information, thereby achieving intelligent response, which can effectively improve the intelligence of instant messaging and achieve conversation intelligence, thereby improving the customer's instant messaging conversation experience.

[0056] The first terminal and the second terminal may be, but are not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server may be implemented as an independent server or a server cluster consisting of multiple servers.

[0057] In one embodiment, if Figure 2 As shown, a session control method for instant messaging is provided, which is applied in Figure 1 The server in the example is used as an example to illustrate the following steps:

[0058] S10: When receiving a conversation message sent by a client to a user via a first terminal, determining whether to enable the smart conversation mode according to the operation data of a second terminal currently used by the user.

[0059] When a client needs to have a conversation, they send a conversation message to the user via their first terminal. The server receives the conversation message sent by the client via the first terminal and sends it to the user's instant messaging software account. The user can then receive the conversation message via their second terminal and respond to the conversation message, completing the conversation with the client. However, after receiving the conversation message on the second terminal, the user may be unable to respond in time for various reasons. In this case, if the server determines that the user is unable to respond in time, it will provide a smart response function to intelligently reply to the conversation message.

[0060] That is, when receiving a conversation message sent by a client to a user through a first terminal, the server will obtain the operating data of the second terminal currently used by the user, and then determine whether the user can respond to the conversation message in a timely and autonomous manner based on the operating data of the second terminal currently used by the user. If it is determined that the user cannot respond to the conversation message in a timely and autonomous manner, the server determines that it is necessary to enable the intelligent conversation mode to achieve instant response and improve the user conversation experience; if it is determined that the user can respond to the conversation message in a timely and autonomous manner, the server determines that it is not necessary to enable the intelligent conversation mode.

[0061] S20: If the intelligent conversation mode needs to be enabled, the conversation message is subjected to intent recognition to obtain the customer's conversation intent.

[0062] After determining whether the smart conversation mode needs to be enabled based on the operating data of the second terminal currently used by the user, if it is determined that the user is unable to respond to the conversation message independently in a timely manner, that is, the server needs to enable the smart conversation mode, then the server needs to use the intent recognition model to identify the intent of the conversation message sent by the customer and obtain the customer's conversation intention.

[0063] In order to ensure the accuracy of intent recognition, the context information of the conversation message can also be obtained as intent recognition data, and then the intent recognition data is input into the intent recognition model for intent recognition to obtain the customer's conversation intention, so as to improve the diversity of input data and improve the accuracy of intent recognition.

[0064] S30: Obtain the customer profile of the customer and the user profile of the user, and generate response information for the current session based on the session intention, the user profile, and the customer profile.

[0065] While identifying the intent of the conversation message, the server also needs to obtain the client's customer profile and the user's user profile, and then generate the response information for the current conversation based on the conversation intent, user profile, and customer profile. Different users / customers have different speaking styles, and different users have different wording for different customers, resulting in different responses to different customers. Therefore, to ensure the humanization and personalization of intelligent responses, the generated response information is adapted to the user profile and customer profile. The response information matches the portrait information (including the user portrait and customer portrait), which means that the generated response information is consistent with the situation described in the portrait information. The generated conversation message can make the user feel that it is the message they want to send, so that the source of the response information in the conversation scenario is no longer limited to the pre-entered template data. Under the influence of the portrait information, it can simulate the user's true intention and automatically generate the response message accurately and adaptively, thereby improving the intelligence, humanization, and personalization of the response message.

[0066] In this embodiment, the customer profile is constructed based on the customer's historical instant messaging data, including the customer's basic information (such as age, gender, occupation, hobbies, etc.), as well as the customer's preferences and wording style in historical communications with the user. The user profile is constructed based on the user's historical instant messaging data, including the user's basic information (such as age, gender, occupation, hobbies, etc.), as well as the user's preferences and wording style in historical communications with the customer.

[0067] In other embodiments, to maintain conversational diversity and further enhance conversational intelligence, customer and user profiles also include preferred response types from past communications between the parties, including voice, text, image, and text-image links. This facilitates determining the presentation format of response information based on the preferred response types from past communications between the parties. For example, after generating a response, the response information can be converted based on the preferred response types from past communications and presented in the form of language, images, or text, which can then be pushed to the client's second terminal, enhancing conversational intelligence.

[0068] S40: Push the response information to the first terminal according to the intelligent conversation mode, and conduct a conversation with the customer.

[0069] After generating a response message for the current session based on the session intent, user profile, and customer profile, the server pushes the response message to the first terminal in intelligent conversation mode, allowing the client to continue the conversation. In intelligent conversation mode, the user is currently unable to actively respond. Therefore, after automatically generating the response message, the server pushes it directly to the first terminal and records it in the conversation log for subsequent query.

[0070] In one embodiment, to improve user automation and avoid situations where the user can currently make an active response, after automatically generating a response message, the response message needs to be entered into the conversation input box of the second terminal and a timer is set. If a send instruction from the user is received within a preset time period, the response message is directly pushed to the second terminal, completing the response to the conversation message, thereby conversing with the client. If an edit instruction from the user for the response message is received within a preset time period, the user edit instruction is executed to edit the response message, and after receiving the send instruction, the edited response message is pushed to the second terminal. If no send instruction or edit instruction from the user is received within the preset time period, the response message is directly pushed to the second terminal after the preset time period. If a send instruction or edit instruction from the user is received within the preset time period, it indicates that the current user can make an autonomous response. After sending the current response message, the server exits the intelligent conversation mode and enters the assisted conversation mode or the normal conversation mode (user autonomous response, without any assisted conversation function), allowing the user to autonomously answer the conversation.

[0071] In this embodiment, when a conversation message is received from a customer through a first terminal and sent to a user, it is determined whether the intelligent conversation mode needs to be enabled based on the operating data of the second terminal currently used by the user; if the intelligent conversation mode needs to be enabled, the conversation message is subjected to intent recognition to obtain the customer's conversation intent; the customer's customer profile and the user's user profile are obtained, and response information for the current conversation is generated based on the conversation intent, user profile, and customer profile; the response information is pushed to the first terminal according to the intelligent conversation mode to conduct a conversation with the customer. In the present invention, the conversation mode is determined based on the operating data of the second terminal used by the user, and the intelligent conversation mode is enabled when the user currently has difficulty in achieving an autonomous response. The intention of the customer's conversation message is automatically recognized, and then personalized response information is automatically generated based on the customer's intent, the user, and the customer's portrait information, thereby achieving intelligent response, which can effectively improve the intelligence of instant messaging and achieve conversation intelligence, thereby improving the customer's instant messaging conversation experience.

[0072] In one embodiment, the operation data includes the operation status. Figure 3 As shown, in step S10, that is, determining whether to enable the smart conversation mode based on the operating data of the second terminal currently used by the user, specifically includes the following steps:

[0073] S11: Determine whether the operating state of the second terminal is the power-on state.

[0074] In this embodiment, the second terminal is a terminal device bound to or logged into the user's instant messaging account; the first terminal is a terminal device bound to or logged into the client's (the user's client or friend's) instant messaging account. When the server receives a conversation message sent by the client to the user via the first terminal, it determines the terminal device currently logged into the user's instant messaging account as the second terminal (the second terminal may be the terminal device bound to the instant messaging account or the terminal device most recently logged into the instant messaging account). It then determines whether the operating status of the second terminal is powered on, which includes powered on (the device is powered on and operating normally) and powered off (the device is powered off and cannot operate). If the operating status of the second terminal is powered off, indicating that the user cannot use the second terminal normally and cannot respond to conversation messages in a timely manner, it can be determined that the intelligent conversation mode needs to be enabled to intelligently respond to the client's conversation messages to ensure the timeliness of the conversation.

[0075] S12: If the running state of the second terminal is the power-on state, obtain application data on the second terminal.

[0076] In this embodiment, the operating data of the second terminal also includes application data on the second terminal. After determining whether the operating status of the second terminal is on, if the operating status of the second terminal is on, indicating that the second terminal is turned on and operating normally, and the user is using the second terminal, it is necessary to further determine whether the user can actively respond to conversation messages. The server then needs to obtain the application data on the second terminal, which includes the type and operation log of each application.

[0077] S13: Determine whether to enable the smart conversation mode based on the operation logs of each application.

[0078] After obtaining the application data on the second terminal, the server needs to determine whether the smart conversation mode needs to be enabled based on the type and operation log of each application in the application data.

[0079] That is, after determining whether the operating status of the second terminal is the power-on state, if the operating status of the second terminal is the power-on state, the server needs to obtain the types of all applications on the second terminal and the operation logs of each application; then, based on the types of all applications on the second terminal and / or the operation logs of each application, determine whether the smart session mode needs to be enabled.

[0080] Determining whether the smart conversation mode needs to be enabled based on the operation logs of each application can be as follows: determining whether there is an application used by the user from the current time point to the previous preset time period (such as 30 seconds, 1 minute, or 3 minutes ago) based on the operation logs of each application; if there is an application used by the user from the current time point to the previous preset time period, it can be determined that the user has recently operated the second terminal and the user may be using the second terminal, and it is determined that the user can make a timely and autonomous response to the customer's conversation message, and it is determined that the smart conversation mode does not need to be enabled; if there is no application used by the user from the current time point to the previous preset time period, indicating that the user has not used the second terminal in the recent period, it is determined that the user may not be able to make a timely and autonomous response to the customer's conversation message, and it is determined that the smart conversation mode needs to be enabled. Determining whether the smart conversation mode needs to be enabled by this method can effectively determine the user's idle state, thereby effectively and objectively determining whether the user can make an autonomous response. This is simple and has a high accuracy rate, and can effectively ensure the accuracy of conversation message responses.

[0081] In this embodiment, by determining whether the operating status of the second terminal is in the power-on state, the second terminal is a terminal device bound to or logged in by the user's session account; if the operating status of the second terminal is in the power-on state, the application data on the second terminal is obtained, and the application data includes the type and operation log of each application; according to the type and operation log of each application, it is determined whether the smart session mode needs to be enabled. The steps of determining whether the smart session mode needs to be enabled based on the operating data of the second terminal currently used by the user are refined, and whether the smart session mode needs to be enabled is determined by the operation log of each application on the second terminal. This is simple and has high accuracy, and can effectively ensure the accuracy of the session message response.

[0082] In one embodiment, the application data further includes the type of each application. In step S13, determining whether to enable the smart conversation mode is performed based on the type of each application and the operation log, specifically includes the following steps:

[0083] S131: Determine, based on the operation logs of each application, whether the second terminal has an application currently in use.

[0084] In this embodiment, the application data includes not only the operation logs of each application on the second terminal but also the type of each application. After obtaining the application data on the second terminal, it is necessary to determine whether any application is currently in use on the second terminal based on the operation logs of each application on the second terminal. If it is determined that an application is currently in use on the second terminal, it indicates that the user is frequently using the second terminal and can promptly reply to instant messaging conversation messages. Therefore, it is determined that the intelligent conversation mode does not need to be enabled, which is simple and intuitive.

[0085] S132: If the second terminal does not currently have an application in use, determine a target application that has been operated by the user within a preset time period.

[0086] After determining whether the second terminal currently has an application in use, if it is determined that the second terminal currently has no application in use, it indicates that the current user has not performed any operation on the second terminal, and further judgment is required to determine whether the user has operated any application on the second terminal within a preset time period before the current time point, and determine the target application that the user has operated within the preset time period before the current time point. The preset time period before can be 30 seconds, 1 minute, 2 minutes, 3 minutes, 4 minutes, or 5 minutes ago.

[0087] If the user has not operated the application on the second terminal within a preset time period before the current time point, it means that the user has not used the second terminal in recent time, and the user is likely to be away from the first terminal, or cannot reply to the instant communication conversation message in time for various reasons, then it is determined that the smart conversation mode needs to be enabled for smart response.

[0088] S133: Determine whether the target application is a first-category application.

[0089] After determining the target application that the user has operated within a preset time period, it is determined whether the target application is a first category application, wherein the first category application includes instant messaging and audio and video playback applications.

[0090] S134: If the target application is not a first-category application, determining that the intelligent conversation mode needs to be enabled;

[0091] S135: If the target application is a first-category application, it is determined that the smart conversation mode does not need to be enabled.

[0092] After determining whether the target application is a first-category application, if the target application is not a first-category application, it means that the user has not performed any instant feedback-type application operations in recent time, and that the used application may be an application that performs background operations according to a certain logic. The user is likely to be away from the first terminal, or cannot reply to the instant communication conversation message in time for various reasons. In this case, it is determined that the intelligent conversation mode needs to be enabled to reduce the possibility of misjudgment.

[0093] If the target application is a first-category application, it means that the user has recently performed application operations with instant feedback and can reply to instant communication conversation messages in a timely manner. In this case, it is determined that there is no need to enable the intelligent conversation mode. Instead, the user can respond independently to reduce the load of running the intelligent conversation mode.

[0094] In other embodiments, the first category of applications also includes applications whose types are marked by users. The user's idle state is judged based on the applications marked by users, thereby further improving the accuracy of the judgment.

[0095] In this embodiment, based on the operation logs of each application, it is determined whether the second terminal currently has an application in use. If no application is currently in use on the second terminal, the target application operated by the user within a preset time period is determined. It is then determined whether the target application is a first-category application, which includes instant messaging and audio and video playback applications. If the target application is not a first-category application, it is determined that the intelligent conversation mode needs to be enabled. If the target application is a first-category application, it is determined that the intelligent conversation mode does not need to be enabled. This method clarifies the specific steps for determining whether the intelligent conversation mode needs to be enabled based on the operation logs of each application, and further improves the accuracy of the judgment by determining the conversation mode based on the application's operation logs and type. By judging the current operation logs of applications such as instant messaging and audio and video playback, the user's idle state can be intuitively, quickly, and accurately determined, and the timeliness of the user's response to customer conversation messages can be evaluated.

[0096] In one embodiment, if Figure 4 As shown, in step S30, the response information of the current session is generated according to the session intention, user profile and customer profile, which specifically includes the following steps:

[0097] S31: Obtain a pre-built intent map and obtain preset response template data pre-stored in a database.

[0098] After determining the customer's conversational intent and obtaining the user and customer profiles, it's necessary to obtain a pre-built intent graph and pre-stored response template data in the database. The nodes of the intent graph represent the semantic content of the intent, and the edges represent the semantic connections between nodes.

[0099] In this embodiment, the preset response template data includes multiple response templates pre-created based on different intents. The multiple response templates in the preset response template data include text, images, and graphic pages, each of which corresponds to at least one intent tag. The graphic pages can be network pages such as HTML5 pages, web pages, and WeChat article pages.

[0100] S32: Determine a target response template in the preset response template data according to the conversation intention and the intention map.

[0101] After obtaining the pre-built intent map and preset response template data, the target response template is determined in the preset response template data according to the conversation intent and the intent map.

[0102] For example, the conversation intent is first matched with multiple intentions (nodes) in the intent graph to obtain the target node intention that matches the conversation intent; then the target node intention is matched with multiple response templates in the preset response template data, that is, the target node intention is matched with the intention label corresponding to each response template, and it is determined whether the match is successful; if the match is successful, the matched response template is used as the target response template.

[0103] After matching the target node intention with multiple response templates in the preset response template data, if the match is not successful, other nodes directly associated with the target node intention are determined and recorded as secondary node intentions; then the secondary node intention is matched with multiple response templates in the preset response template data, that is, the secondary node intention is matched with the intention label corresponding to each response template, and the matched response template is recorded as the pending response template; if the number of pending response templates is equal to 1, that is, there is only one pending response template, then the pending response template is directly used as the target response template.

[0104] If the number of pending response templates is greater than 1, that is, there are multiple pending response templates, it is necessary to use the edge between the secondary node intent corresponding to each pending response template and the target node intent as a weight, and determine the target response template from the multiple pending response templates based on the corresponding weight to ensure the match between the target response template and the customer intent (i.e., the session intent), thereby ensuring the accuracy of the response message. For example, the pending response template corresponding to the secondary node intent with the largest weight is directly recorded as the target response template, and the pending response template that best matches the customer intent is used as the target response template to ensure the accuracy of the target response template.

[0105] In other embodiments, if the number of pending response templates is greater than 1, the target response template can also be determined from the multiple pending response templates based on the historical conversation records between the user and the customer, and the edges between the secondary node intentions and the target node intentions corresponding to each pending response template. For example, if the number of pending response templates is greater than 1, it is determined whether the pending response template has appeared in the historical conversation records between the user and the customer; if it has appeared, the pending response template is removed, and the target response template is determined from the remaining pending response templates based on the edges between the secondary node intentions and the target node intentions corresponding to each pending response template as weights, thereby avoiding repeated push of response messages with the same content and improving the accuracy of the response messages.

[0106] S33: Modify the target response template according to the user portrait and the customer portrait to obtain response data that matches the user portrait and the customer portrait as the response information of the current session.

[0107] After determining the target response template, the target response template is modified according to the user portrait and customer portrait to obtain response data that matches the user portrait and customer portrait, which is used as the response information for the current session. This ensures that the response information not only meets the customer's intentions but also matches the user and customer portraits, thereby improving the intelligence of the response.

[0108] For example, after determining the target response template, the type of the target response template can be determined first. If the type of the target response template is text (such as a preset speech text), the wording style of the historical conversation between the user and the customer is determined based on the user portrait and the customer portrait, and the text is modified according to the wording style to obtain the response information of the current conversation, such as adding or modifying the text mood particles according to the wording style to obtain the response information of the current conversation; modifying the text response template according to the wording style of the historical conversation makes the response message closer to the customer's actual impression of the user, thereby improving the customer's conversation experience. If the type of the target response template is an image or a graphic web page, the wording style of the historical conversation between the user and the customer is determined based on the user portrait and the customer portrait, and a recommended sentence for the image or graphic web page is generated, and the recommended sentence and the image (or graphic web page) are used as the response information of the current conversation. Among them, the recommended sentence includes the mood particles that conform to the corresponding wording style, the keywords or title of the image (or graphic web page), etc. In this embodiment, recommended sentences are generated for image and graphic response templates based on the wording style of historical conversations, thereby improving the personification and personalization of the response message, thereby enhancing the customer's conversation experience.

[0109] In this embodiment, a pre-constructed intent graph is obtained, along with preset response template data pre-stored in a database. The nodes of the intent graph represent the semantic content of the intent, and the edges of the intent graph represent the semantic associations between the nodes. Then, based on the session intent and the intent graph, a target response template is determined from the preset response template data. Finally, the target response template is modified based on the user profile and customer profile to obtain response data that matches the user profile and customer profile, which serves as the response information for the current session. This clarifies the specific steps for generating the response information for the current session based on the session intent, user profile, and customer profile. First, based on the session intent and the intent graph, a target response template is determined from the preset response template data, ensuring a match between the target response template and the customer intent. Then, based on the user profile and customer profile, the target response template is modified to make the resulting response message more consistent with the user intent, thereby improving the personalization of the response message and, in turn, enhancing the intelligence of the session.

[0110] In one embodiment, if Figure 5 As shown, after step S10, that is, after determining whether the smart conversation mode needs to be enabled based on the operating data of the second terminal currently used by the user, the method further specifically includes the following steps:

[0111] S50: If the intelligent conversation mode does not need to be enabled, the second terminal is controlled to enable a conversation assistance function, and an auxiliary function menu of the conversation assistance function is controlled to be displayed in the conversation window of the second terminal.

[0112] After determining whether the intelligent conversation mode needs to be enabled based on the operating data of the second terminal currently used by the user, if it is determined that the intelligent conversation mode does not need to be enabled, indicating that the user can make timely and autonomous responses, the server can provide the user with a conversation assistance function, thereby reducing the situation where the user may need to query information or organize language when answering, reducing the difficulty of data query, and improving the conversation experience.

[0113] That is, if the smart conversation mode does not need to be enabled, the server controls the user's second terminal to turn on the conversation assistance function, and controls the auxiliary function menu of the conversation assistance function to be displayed in the conversation window of the second terminal, so that the user can view the relevant information and use the function of the conversation assistance function during the conversation. The auxiliary function menu can be displayed on the screen of the second terminal in parallel with the conversation window of the current conversation. For example, the auxiliary function menu can be on the left or right side of the conversation window, or above the conversation window. The auxiliary function menu can be displayed in a blank position in the conversation window in the form of a floating window, so that the user can browse and use it. In other embodiments, in order to reduce the impact of the auxiliary function menu on the conversation window,

[0114] The auxiliary function menu can also be displayed between the information recording area and the information input field of the conversation window. For example, the auxiliary function menu includes quick response, briefcase,

[0115] and customer portraits, and arrange quick responses, briefcases, and customer portraits horizontally between the information recording area and the information input bar in the conversation window to facilitate users to use and browse the corresponding data.

[0116] In this embodiment, the second terminal needs to pre-install a program package corresponding to the conversation assistance function so that the second terminal can directly call the program package to run the conversation assistance function.

[0117] 0S60: Receive the user's operation instruction for the auxiliary function menu, and feed back corresponding display data to the second terminal in response to the operation instruction, so that the second terminal displays the display data.

[0118] After the auxiliary function menu for controlling the conversation auxiliary function is displayed in the conversation window of the second terminal, it is determined whether an operation instruction for the auxiliary function menu from the user is received. If an operation instruction for the auxiliary function menu from the user is received, corresponding display data is fed back to the second terminal in response to the user's operation instruction, so that the second terminal displays the display data.

[0119] 5S70: Receive the user's data selection instruction for display data, and determine the target data selected by the user according to the data selection instruction.

[0120] Label data.

[0121] In response to the operation instruction, the corresponding display data is fed back to the second terminal so that the second terminal displays the display data. The server receives the user's data selection instruction for the display data and determines the target data selected by the user according to the data selection instruction.

[0122] 0S80: Generate response information of the current session according to the target data, and push the response information to the first terminal, and communicate with the client.

[0123] Users can have conversations.

[0124] After determining the target data selected by the user according to the data selection instruction, it is necessary to generate response information of the current session according to the target data, and push the response information to the first terminal to conduct a session with the client.

[0125] Among them, the auxiliary function menu may include quick answers and customer portraits; the user's operation instructions may be click instructions 5 or voice instructions; the display data corresponding to the quick answer is the preset speech data; the display data corresponding to the customer portrait is the customer

[0126] User portrait data, including customer's gender, age, occupation and preferences. Taking click command as an example, users can obtain preset speech data (including multiple preset speech texts) by clicking on quick response. By clicking on the corresponding preset speech text in the preset speech data, users can directly enter the preset speech text as the target data into the information input field, which is convenient for users to make information.

[0127] Edit the message to generate a response message and send it, or directly send the preset text as the response message, which is simple and fast; click 0 Customer Portrait to view the customer's portrait information (customer portrait data). By viewing the customer's portrait information, you can timely understand the customer's gender, age, occupation, preferences and other information, which helps users better respond to conversations. In this embodiment, quick responses can be used to quickly respond to certain customer conversation intentions, which is simple and convenient and reduces the user's editing time; the customer's corresponding portrait information is displayed through the customer portrait, allowing users to timely understand the customer's relevant information and respond to the customer's conversation messages based on the customer portrait information, making the response information more in line with the customer's preferences.

[0128] In this embodiment, after determining whether to enable smart conversation mode based on the operating data of the second terminal currently used by the user, if smart conversation mode is not required, the second terminal is controlled to enable a conversation assistance function and an auxiliary function menu of the conversation assistance function is controlled to be displayed in the conversation window of the second terminal. The user's operation instruction for the auxiliary function menu is then received, and corresponding display data is fed back to the second terminal in response to the operation instruction, so that the second terminal displays the display data. The user's data selection instruction for the display data is then received, and the target data selected by the user is determined based on the data selection instruction. Finally, response information for the current conversation is generated based on the target data, and the response information is pushed to the first terminal to conduct a conversation with the customer. When the user is able to independently respond to the customer's conversation message, the conversation assistance function is enabled to provide the user with different function assistance menus, and then data is displayed in response to the user's operation, which can provide conversation assistance for the user and further improve the conversation experience.

[0129] In one embodiment, the auxiliary function menu includes a briefcase and a quick answer. In step S60, that is, responding to the operation instruction and feeding back corresponding display data to the second terminal, the steps specifically include the following:

[0130] S61: If an operation instruction for a quick response from the user is received, the preset speech data is obtained from the database, and the preset speech data is fed back to the second terminal as display data.

[0131] In this embodiment, the auxiliary function menu includes a briefcase and a quick answer. After the auxiliary function menu for controlling the conversation auxiliary function is displayed in the conversation window of the second terminal, if a user operation instruction for a quick answer is received, the preset speech data associated with the quick answer is obtained from the database in response to the user's operation on the quick answer, and the preset speech data is fed back to the second terminal as display data, so that the second terminal displays the preset speech data. The preset speech data includes multiple preset speech texts, and the multiple preset speech texts correspond to different intent tags.

[0132] After the preset speech data is fed back to the second terminal as display data, if a user selection instruction for any preset speech text in the preset speech data is received, the preset speech text will be used as the target data selected by the user, so that response information can be generated subsequently based on the selected preset speech text.

[0133] S62: If an operation instruction for the briefcase is received from the user, the briefcase data is obtained from the database, and the display screen data of the second terminal is obtained.

[0134] After the auxiliary function menu for controlling the conversation auxiliary function is displayed in the conversation window of the second terminal, if a user operation instruction on the briefcase is received, the user operation on the briefcase is responded to and the briefcase data associated with the briefcase is retrieved from the database. The briefcase data includes multiple product pages and multiple information pages. The product pages and information pages can be displayed as images or as graphic and text pages. That is, the briefcase data includes multiple images and / or graphic and text pages.

[0135] In addition, since the terminal device does not have the required display data (including screen size, brightness and resolution, etc.),

[0136] Therefore, different terminal devices have different display effects on briefcase data. In order to ensure that the second terminal can fully display the briefcase data and ensure the uniformity of the display effect, if the user's operation instruction for the briefcase is received, the briefcase data is obtained from the database.

[0137] When the briefcase data is displayed, the server also needs to obtain the display screen data of the second terminal so that the briefcase data can be fully displayed according to the display screen data of the second terminal.

[0138] S63: Generate a briefcase display interface according to the display screen data and the briefcase data, and feed the briefcase display interface back to the second terminal.

[0139] 0 After acquiring the briefcase data and the display screen data of the second terminal,

[0140] Generate a briefcase display interface and feed it back to the second terminal. For example, the display screen data includes data such as screen size, screen brightness, and screen resolution. The size of the briefcase display interface is determined according to the screen size, and the briefcase data is rendered using a preset page template according to the screen brightness and screen resolution to obtain the briefcase display interface.

[0141] Then the rendered briefcase display interface is fed back to the second terminal, so that the second terminal displays the briefcase display interface, so that the briefcase display interface is adapted to the second terminal, thereby improving the data display effect of the briefcase display interface.

[0142] The briefcase display interface may include interface controls and text controls such as a floating layer (div), a floating button, an input box, a progress bar, and may also be bound to a Software Development Kit (SDK) attribute trigger event to facilitate user information query, browsing, and selection.

[0143] The preset page template is a page display template designed in advance according to the data content of the briefcase data. For example, if the briefcase data includes multiple product pages and multiple information pages, the preset page template at least includes the product column and information column.

[0144] The product section includes multiple product pages with different intent tags; the information section includes multiple information pages with different intent tags. In other embodiments, the briefcase data also includes other content, and the preset page template can also be designed in other styles accordingly.

[0145] After the briefcase display interface is fed back to the second terminal as display data, if a user selects any page in the briefcase display interface 5, the selected page is used as the target data selected by the user, so that the selected page can be selected according to the selected page.

[0146] The page fetched generates response information.

[0147] In this embodiment, the auxiliary function menu includes a briefcase and a quick answer. After the auxiliary function menu for controlling the conversation auxiliary function is displayed in the conversation window of the second terminal, if the user's operation instruction for the quick answer is received, the database

[0148] The system obtains preset speech data from the user's operation instruction and feeds the preset speech data back to the second terminal as display data; if a user operation instruction for the briefcase is received, the briefcase data is obtained from the database and the display screen data of the second terminal is obtained; a briefcase display interface is generated based on the display screen data and the briefcase data, and the briefcase display interface is fed back to the second terminal. This embodiment clarifies the specific process of feeding back corresponding display data to the second terminal in response to the user's operation instruction, provides users with functions such as quick response and briefcase, allows users to select multiple types of data according to their selection, improves the diversity of response messages, and different auxiliary function menus have different data display methods, further improving the user's conversation experience.

[0149] In one embodiment, the auxiliary function menu includes a customer portrait. In step S60, corresponding display data is fed back to the second terminal in response to the operation instruction, which specifically includes the following steps:

[0150] S64: If an operation instruction for the customer portrait is received from the user, the customer portrait data corresponding to the customer is determined, and the display screen data of the second terminal is obtained.

[0151] In this embodiment, the auxiliary function menu also includes a customer profile. After the auxiliary function menu for controlling the conversation auxiliary function is displayed in the conversation window of the second terminal, if a user operation instruction on the customer profile is received, indicating that the user has no or limited understanding of the customer, the user operation on the customer profile is responded to, and the customer profile data corresponding to the customer is determined based on the customer's user ID and other information, so that the customer profile data can be displayed to the user to facilitate understanding of the customer.

[0152] In addition, since different terminal devices have different display data (including screen size, brightness, and resolution, etc.), different terminal devices will display different customer profile data. To ensure that the second terminal can fully display the customer profile data and ensure the uniformity of the display effect, when receiving the user's operation instruction for the user profile, the server also needs to obtain the display screen data of the second terminal while obtaining the customer profile data from the database, so that the customer profile data can be fully displayed based on the display screen data of the second terminal.

[0153] S65: Generate a customer information display interface based on the customer portrait data and the display screen data, and feed the customer information display interface back to the second terminal as display data.

[0154] After acquiring the customer portrait data and the display screen data of the second terminal, a customer information display interface is generated based on the display screen data and the customer portrait data, and the customer information display interface is fed back to the second terminal as display data. For example, the display screen data includes data such as screen size, screen brightness, and screen resolution. The size of the customer information display interface is determined based on the screen size, and the customer portrait data is rendered using a preset page template based on the screen brightness and screen resolution to obtain the customer information display interface. The rendered customer information display interface is then fed back to the second terminal so that the second terminal displays the customer information display interface. This makes the customer information display interface compatible with the second terminal, improves the data display effect of the customer information display interface, and thereby improves the user experience.

[0155] After the customer information display interface is fed back to the second terminal as display data, if a user selects any information in the customer information display interface, the selected information is used as the user-selected target data, so that a response message can be generated based on the selected information. For example, if the user selects a customer nickname in the customer information display interface, the customer nickname and the customer's conversational intent are combined to generate a response message containing the customer nickname, thereby bringing the user closer to the customer and improving the conversational intelligence and experience.

[0156] In this embodiment, the auxiliary function menu includes a customer portrait. After the auxiliary function menu for controlling the conversation auxiliary function is displayed in the conversation window of the second terminal, if an operation instruction for the customer portrait from the user is received, the customer portrait data corresponding to the customer is determined, and the display screen data of the second terminal is obtained. Then, a customer information display interface is generated based on the customer portrait data and the display screen data, and the customer information display interface is fed back to the second terminal as display data. This clarifies the specific process of responding to the user's operation instruction and feeding back the corresponding display data to the second terminal, providing the user with a customer portrait function so that the user can understand customer information in a timely manner, thereby effectively assisting in the response and further improving the user's conversation experience.

[0157] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0158] In one embodiment, a session control device for instant messaging is provided, which corresponds one-to-one to the session control method for instant messaging in the above embodiment. Figure 6 As shown, the conversation control device for instant messaging includes a determination module 601, an identification module 602, a generation module 603 and a push module 604. The functional modules are described in detail as follows:

[0159] Determining module 601, configured to determine whether to enable the smart conversation mode based on the operation data of the second terminal currently used by the user upon receiving a conversation message sent by the client to the user via the first terminal;

[0160] Identification module 602, used to identify the intent of the conversation message and obtain the customer's conversation intent if the intelligent conversation mode needs to be enabled;

[0161] A generation module 603 is used to obtain the customer profile of the client and the user profile of the user, and generate response information for the current session based on the session intention, the user profile, and the customer profile;

[0162] The push module 604 is configured to push the response information to the first terminal according to the intelligent conversation mode, and to conduct a conversation with the client.

[0163] Optionally, the generating module 603 is specifically configured to:

[0164] Obtain a pre-built intent graph and preset response template data pre-stored in the database. The nodes of the intent graph are the semantic content of the intent, and the edges of the intent graph are the semantic associations between the nodes.

[0165] Determine the target response template from the preset response template data based on the conversation intent and the intent map;

[0166] The target response template is modified according to the user portrait and the customer portrait to obtain response data that matches the user portrait and the customer portrait as the response information of the current session.

[0167] Optionally, the operation data includes an operation status, and the determination module 601 is specifically configured to:

[0168] Determining whether the operating state of the second terminal is a power-on state;

[0169] If the running state of the second terminal is the power-on state, obtaining application data on the second terminal, the application data including the operation log of each application;

[0170] Determine whether to enable smart conversation mode based on the operation logs of each application.

[0171] Optionally, the application data further includes the type of each application, and the determination module 601 is further specifically configured to:

[0172] Determining whether there is an application currently in use on the second terminal according to the operation log of each application;

[0173] If the second terminal does not currently have an application in use, determining a target application that the user has operated within a preset time period;

[0174] Determine whether the target application is a first-class application. First-class applications include instant messaging and audio and video playback.

[0175] Applications that place categories;

[0176] 0 If the target application is not a first-category application, it is determined that the smart session mode needs to be enabled;

[0177] If the target application is a first-category application, it is determined that the smart session mode does not need to be enabled.

[0178] Optionally, the conversation control apparatus for instant messaging further includes a control module 605 and a feedback module 606; after determining whether the intelligent conversation mode needs to be enabled based on the operation data of the second terminal currently used by the user, the method further includes:

[0179] The control module 605 is used to: if the intelligent conversation mode does not need to be enabled, control the second terminal to enable the conversation assistance function 5 and control the auxiliary function menu of the conversation assistance function to be displayed in the conversation window of the second terminal;

[0180] The feedback module 606 is configured to receive an operation instruction from a user on the auxiliary function menu, and in response to the operation instruction, feedback corresponding display data to the second terminal so that the second terminal displays the display data;

[0181] The determination module 601 is further used to: receive a user's data selection instruction for display data, and determine the display data according to the data selection instruction.

[0182] Determine the target data selected by the user;

[0183] The generating module 603 is also used to generate the response information of the current session according to the target data so that the pushing module 604 can push the response information to the target data.

[0184] The answer information is pushed to the first terminal and a conversation is conducted with the customer.

[0185] Optionally, the auxiliary function menu includes a briefcase and a quick answer, and the feedback module 606 is specifically configured to:

[0186] If the user receives a quick response instruction, the preset speech data is obtained from the database and the preset speech data is

[0187] The technical data is fed back to the second terminal as display data;

[0188] 5. If a briefcase operation instruction is received from the user, the briefcase data is obtained from the database, and the display screen data of the second terminal is obtained;

[0189] A briefcase display interface is generated according to the display screen data and the briefcase data, and the briefcase display interface is fed back to the second terminal.

[0190] Optionally, the auxiliary function menu includes a customer portrait, and the feedback module 606 is specifically configured to: 0 if an operation instruction on the customer portrait is received from the user, determine the customer portrait data corresponding to the customer, and obtain display screen data of the second terminal;

[0191] A customer information display interface is generated according to the customer portrait data and the display screen data, and the customer information display interface is fed back to the second terminal as display data.

[0192] For the specific definition of the instant messaging session control device, please refer to the definition of the instant messaging session control method above, and will not be repeated here. The various modules in the instant messaging session control device can be implemented in whole or in part through software, hardware, or a combination thereof. The modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0193] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the data generated and used by the aforementioned conversation control method for instant messaging, including customer portraits, user portraits, conversation messages, conversation intentions, response information, as well as preset speech data, briefcase data, etc. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a conversation control method for instant messaging is implemented.

[0194] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown. The computer device includes a processor, memory, a network interface, a display screen, and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server via a network connection. When executed by the processor, the computer program implements a session control method for instant messaging.

[0195] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the following steps are performed:

[0196] Upon receiving a conversation message sent by a client to a user via a first terminal, determining whether to enable the smart conversation mode based on the operating data of a second terminal currently used by the user;

[0197] If you need to enable the intelligent conversation mode, perform intent recognition on the conversation message to obtain the customer's conversation intent;

[0198] Obtain the customer's customer profile and the user's user profile, and generate response information for the current session based on the session intent, user profile, and customer profile;

[0199] The response information is pushed to the first terminal according to the intelligent conversation mode to conduct a conversation with the customer.

[0200] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0201] Upon receiving a conversation message sent by a client to a user via a first terminal, determining whether to enable the smart conversation mode based on the operating data of a second terminal currently used by the user;

[0202] If you need to enable the intelligent conversation mode, perform intent recognition on the conversation message to obtain the customer's conversation intent;

[0203] Obtain the customer's customer profile and the user's user profile, and generate response information for the current session based on the session intent, user profile, and customer profile;

[0204] The response information is pushed to the first terminal according to the intelligent conversation mode to conduct a conversation with the customer.

[0205] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0206] Those skilled in the art will clearly understand that for the sake of convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0207] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the scope of protection of the present invention.

Claims

1. A session control method for instant messaging, characterized in that: include: Upon receiving a conversation message sent by a client to a user via a first terminal, determining whether to enable the smart conversation mode based on the operating data of a second terminal currently used by the user; If the intelligent conversation mode needs to be enabled, the conversation message is subjected to intent recognition to obtain the conversation intent of the customer; Obtaining a customer profile of the customer and a user profile of the user, and generating response information for the current session based on the session intention, the user profile, and the customer profile; Pushing the response information to the first terminal according to the intelligent conversation mode, and conducting a conversation with the customer; The operation data includes an operation status, and determining whether to enable the smart conversation mode based on the operation data of the second terminal currently used by the user includes: Determining whether the operating state of the second terminal is a power-on state; If the running state of the second terminal is the power-on state, obtaining application data on the second terminal, the application data including an operation log of each application and a type of each application; Determining whether to enable the intelligent conversation mode according to the operation logs of each application includes: determining, based on the operation logs of the respective applications, whether the second terminal has an application currently in use; If the second terminal does not currently have an application in use, determining a target application that the user has operated within a preset time period; Determining whether the target application is a first category application, where the first category application is an instant feedback type application, including an instant messaging type application and an audio and video playback type application; If the target application is not a first-category application, it is determined that the smart conversation mode needs to be enabled.

2. The session control method according to claim 1, wherein: Generating response information for the current session according to the session intention, the user profile, and the customer profile includes: Obtaining a pre-built intent graph and pre-stored preset response template data in a database, wherein the nodes of the intent graph are the semantic content of the intent and the edges of the intent graph are the semantic associations between the nodes; Determining a target response template from the preset response template data according to the conversation intention and the intention map; The target response template is modified according to the user portrait and the customer portrait to obtain response data that matches the user portrait and the customer portrait as the response information of the current session.

3. The session control method according to any one of claims 1 to 2, wherein: After determining whether to enable the smart conversation mode based on the operating data of the second terminal currently used by the user, the method further includes: If the intelligent conversation mode does not need to be enabled, controlling the second terminal to enable a conversation assistance function, and controlling an auxiliary function menu of the conversation assistance function to be displayed in the conversation window of the second terminal; receiving an operation instruction of the user on the auxiliary function menu, and feeding back corresponding display data to the second terminal in response to the operation instruction, so that the second terminal displays the display data; receiving a data selection instruction from the user for the displayed data, and determining target data selected by the user according to the data selection instruction; Generate response information of the current session according to the target data, push the response information to the first terminal, and conduct a session with the client.

4. The session control method according to claim 3, wherein: The auxiliary function menu includes a briefcase and a quick answer, and the responding to the operation instruction and feeding back corresponding display data to the second terminal includes: If an operation instruction of the user for the quick answer is received, obtaining preset speech data from the database, and feeding back the preset speech data as display data to the second terminal; If an operation instruction of the user on the briefcase is received, obtaining briefcase data from the database and obtaining display screen data of the second terminal; A briefcase display interface is generated according to the display screen data and the briefcase data, and the briefcase display interface is fed back to the second terminal.

5. The session control method according to claim 4, wherein: The auxiliary function menu includes a customer portrait, and the responding to the operation instruction to feed back corresponding display data to the second terminal includes: If an operation instruction of the user on the customer portrait is received, determining the customer portrait data corresponding to the customer, and obtaining the display screen data of the second terminal; A customer information display interface is generated according to the customer portrait data and the display screen data, and the customer information display interface is fed back to the second terminal as the display data.

6. A conversation control device for instant messaging, characterized in that: include: A determination module is configured to, upon receiving a conversation message sent by a client to a user via a first terminal, determine whether the smart conversation mode needs to be enabled based on operating data of a second terminal currently used by the user; wherein the operating data includes an operating state, and determining whether the smart conversation mode needs to be enabled based on the operating data of the second terminal currently used by the user includes: Determining whether the operating state of the second terminal is a power-on state; If the running state of the second terminal is the power-on state, obtaining application data on the second terminal, the application data including an operation log of each application and a type of each application; Determining whether to enable the intelligent conversation mode according to the operation logs of each application includes: determining, based on the operation logs of the respective applications, whether the second terminal has an application currently in use; If the second terminal does not currently have an application in use, determining a target application that the user has operated within a preset time period; Determining whether the target application is a first category application, where the first category application is an instant feedback type application, including an instant messaging type application and an audio and video playback type application; If the target application is not a first-category application, determining that the smart conversation mode needs to be enabled; an identification module, configured to perform intent identification on the conversation message to obtain the customer's conversation intent if the intelligent conversation mode needs to be enabled; a generation module, configured to obtain a customer profile of the customer and a user profile of the user, and generate response information for the current session based on the session intention, the user profile, and the customer profile; A push module is used to push the response information to the first terminal according to the intelligent conversation mode to conduct a conversation with the customer.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the conversation control method for instant messaging according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the conversation control method for instant messaging according to any one of claims 1 to 5 are implemented.

Images