Message processing method, computing device, storage medium and program product

Abstract

The embodiment of the invention provides a message processing method, computing equipment, a storage medium and a program product. The method comprises the following steps: detecting a session request sent by a first user side; wherein the session request is generated according to a session starting operation triggered by the first user for the second user; performing multiple rounds of sessions with the first user side by using the intelligent session robot, and detecting whether the first user meets a first quality requirement or not according to at least one session message sent by the first user side; obtaining a first detection result; and when the first detection result is yes, sending the historical message record generated by the multi-round session and the session message currently sent by the first user side to the second user side. According to the technical scheme provided by the embodiment of the invention, the session quality is improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of computer, and particularly relate to a message processing method, a computing device, a storage medium and a program product. BACKGROUND

[0002] In some instant communication scenarios, there is a case that one user needs to have a conversation with multiple users respectively, for example, some online systems providing objects for customers to perform interactive behaviors, in order to provide better interactive experience for customers, instant communication services can be provided, and customer service personnel of object providers or online systems can provide customer service for customers. In practical applications, one customer service personnel can need to have a conversation with multiple customers simultaneously, and some customers can be abnormal users lacking of clear demands or malicious harassment, thereby affecting the conversation quality. SUMMARY

[0003] Embodiments of the present application provide a message processing method, a computing device, a storage medium and a program product, to solve the problem of low conversation quality in the prior art.

[0004] In a first aspect, a message processing method is provided in the embodiments of the present application, comprising: detecting a conversation request sent by a first user end; the conversation request is generated according to a conversation start operation triggered by a first user for a second user; in response to the conversation request, performing multi-round conversation with the first user end by using an intelligent conversation robot, and detecting whether the first user meets a first quality requirement according to at least one conversation message sent by the first user end, to obtain a first detection result; in a case that the first detection result is yes, sending historical message records generated by the multi-round conversation and a conversation message currently sent by the first user end to the second user end.

[0005] Optionally, the response to the conversation request, performing multi-round conversation with the first user end by using an intelligent conversation robot, and detecting whether the first user meets a first quality requirement according to at least one conversation message sent by the first user end comprises: in response to the conversation request, determining a user feature of the first user, and detecting whether the first user meets a second quality requirement according to the user feature, to obtain a second detection result; in a case that the second detection result is no, performing multi-round conversation with the first user end by using an intelligent conversation robot, and detecting whether the first user meets a first quality requirement according to at least one conversation message sent by the first user end; The method further comprises: If the first detection result is yes, the session message currently sent by the first user terminal is sent to the second user terminal corresponding to the second user.

[0006] Optionally, if the second detection result is negative, the step of using an intelligent chatbot to conduct multiple rounds of conversations with the first user terminal, and detecting whether the first user meets the first quality requirement based on at least one conversation message sent by the first user terminal includes: If the first detection result is negative, obtain the first session message sent by the first user terminal for the first time; Based on the first session message, detect whether the first user meets the third quality requirement and obtain the third detection result; If the third detection result is yes, the first session message is sent to the second user terminal; If the third detection result is negative, the intelligent conversational robot conducts multiple rounds of conversations with the first user terminal based on the first conversation message, and detects whether the first user meets the first quality requirement based on at least one conversation message sent by the first user terminal.

[0007] Optionally, after detecting whether the first user meets the third quality requirement and obtaining the third detection result, the method further includes: Store the third detection result; The step of detecting whether the first user meets the third quality requirement and obtaining the third detection result based on the first session message includes: Check the stored data to see if there is a third detection result corresponding to the first user; If so, obtain the third detection result; If not, based on the first session message, detect whether the first user meets the third quality requirement and obtain the third detection result.

[0008] Optionally, detecting whether the first user meets the first quality requirement based on the user characteristics includes: Detect whether there are any historical human-to-human conversation records between the first user and the second user; If so, send the session message currently sent by the first user terminal to the second user terminal corresponding to the second user; If not, based on the user characteristics, detect whether the first user meets the first quality requirement.

[0009] Optionally, determining the user characteristics of the first user includes: User features are extracted from the first user's first user behavior data; the user behavior data includes source channel data, historical order data, and / or account attribute data.

[0010] Optionally, the step of detecting whether the first user meets the third quality requirement and obtaining the third detection result based on the first session message includes: Based on at least one of the following: the message content of the first session message, the message type of the first session message, the second user behavior data of the first user, the first detection result corresponding to the first quality requirement, the device information corresponding to the first user terminal, the historical session detection data of the first user, and the historical risk control data of the first user, a target feature is constructed. Based on the target characteristics, it is determined whether the first user meets the third quality requirement, and a third detection result is obtained.

[0011] Optionally, detecting whether the first user meets the second quality requirement based on the user characteristics includes: Based on the user characteristics, the first detection model is used to detect whether the first user meets the second quality requirement; The step of detecting whether the first user meets the third quality requirement based on the first session message includes: Based on the first session message, the second detection model is used to detect whether the first user meets the third quality requirement.

[0012] Optionally, the step of detecting whether the first user meets the second quality requirement using the first detection model based on the user characteristics includes: Based on the user characteristics, the first detection model is used to generate a first detection score; Determine whether the first detection score is greater than a first detection threshold; wherein, if the first detection score is greater than the first detection threshold, determine that the first user meets the first quality requirement, otherwise determine that the first user does not meet the first quality requirement; the first detection model is trained based on the sample user features and the training labels corresponding to the sample user features that meet or do not meet the first quality requirement. The step of detecting whether the first user meets the third quality requirement using the second detection model based on the first session message includes: A second detection score is generated based on the first session message using a second detection model; Determine whether the second detection score is greater than the second detection threshold; wherein, if the second detection score is greater than the second detection threshold, determine that the first user meets the third quality requirement, otherwise determine that the first user does not meet the third quality requirement; the second detection model is trained based on sample session messages and training labels corresponding to the sample session messages that meet or do not meet the third quality requirement.

[0013] Optionally, detecting whether the first user meets the first quality requirement based on at least one session message sent by the first user terminal includes: The intelligent detection model is invoked to parse at least one session message sent by the first user terminal to identify the user's intent, and based on the user's intent, to detect whether the first user meets the first quality requirement.

[0014] Optionally, the second user is the object provider; the multi-round conversation with the first user using an intelligent chatbot includes: Using an intelligent chatbot, multiple rounds of conversations are conducted with the first user terminal according to conversation constraints; the conversation constraints include object transaction requirement guidance requirements.

[0015] Optionally, determining the user characteristics of the first user in response to the session request includes: In response to the session request, if the first user is a first type of test user, the user characteristics of the first user are determined; The method also includes If the first user is a second type of test user, the session message sent by the first user terminal will be sent to the second user terminal.

[0016] Optionally, determining the user characteristics of the first user in response to the session request includes: In response to the session request, it is detected whether the second user meets the detection requirements; If the second user meets the detection requirements, the user characteristics of the first user are determined; The method further includes: If the second user fails to meet the detection requirements, the session message sent by the first user terminal will be sent to the second user terminal.

[0017] Optionally, it also includes: Based on feedback data from multiple test users, perform one of the following processing operations: Adjust the first detection threshold; Adjust the second detection threshold; Retrain the first detection model; Retrain the second detection model.

[0018] Secondly, this application provides a computing device, including a processing component and a storage component; The storage component stores a computer program; the computer program is invoked and executed by the processing component to implement the message processing method as described in the first aspect above.

[0019] Thirdly, this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processing component, implements the message processing method described in the first aspect above.

[0020] Fourthly, this application provides a computer program product, including a computer program or instructions, which, when executed by a processing component, implement the message processing method described in the first aspect above.

[0021] This application embodiment detects a session request sent by a first user terminal; wherein the session request is generated based on a session initiation operation triggered by the first user for the second user; a smart chatbot conducts multiple rounds of conversations with the first user terminal, and detects whether the first user meets a first quality requirement based on at least one session message sent by the first user terminal; obtains a first detection result; if the first detection result is yes, the historical message record generated by the multiple rounds of conversations and the session message currently sent by the first user terminal are sent to the second user terminal.

[0022] By utilizing an intelligent chatbot to handle conversation messages from the first user and engaging in multi-turn dialogues with it, the system can detect in real-time whether the first user meets the primary quality requirements, thus improving the efficiency of conversation quality detection. Furthermore, using an intelligent chatbot to handle conversation messages from the first user allows for quality verification without the second user's awareness, preventing the second user from conversing with abnormal users and improving conversation quality and user experience. In addition, if the first user meets the secondary quality requirements, indicating that the first user is not abnormal, historical message records and current conversation messages can be sent to the second user. The second user can then obtain the complete conversation context and communicate with the first user, achieving a seamless switch from the intelligent chatbot to the second user and ensuring service continuity.

[0023] These or other aspects of this application will become more apparent in the following description of the embodiments. Attached Figure Description

[0024] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 This application provides a schematic diagram illustrating the structure of one embodiment of a system. Figure 2 A flowchart of one embodiment of a message processing method provided in this application is shown; Figure 3 This illustrates a flowchart of message processing in a practical application. Figure 4 This illustrates a signaling flowchart for message processing in a practical application. Figure 5 A schematic diagram of the structure of one embodiment of a message processing device provided in this application; Figure 6 A schematic diagram of one embodiment of a computing device provided in this application is shown. Detailed Implementation

[0025] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0026] It should be noted that, in the cases involving user information in the embodiments of this application, the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in the embodiments of this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse. In addition, the various models involved in this application (including but not limited to language models or generative models) comply with relevant laws and standards.

[0027] Additionally, it should be noted that when user interaction operations or triggering operations are involved in the embodiments of this application, these operations include, but are not limited to, various interaction methods such as touch operations, gesture operations, voice operations, head movement operations, and eye movement operations. Touch operations include, but are not limited to, click operations, double-click operations, long-press operations, swipe operations, pinch operations, or mouse hover operations. Swipe operations include, but are not limited to, straight-line swipes and curved-line swipes.

[0028] The technical solutions of this application can be applied to instant messaging scenarios. As described in the background section, in some instant messaging scenarios, a user needs to have conversations with multiple users separately. For example, some online systems that provide objects for customers to perform interactive behaviors can provide instant messaging services to provide a better interactive experience for customers. Customer service is provided by the object provider or the online system's customer service personnel. In practical applications, a customer service representative may need to have conversations with multiple customers simultaneously. Some customers may be abnormal users who lack clear needs or are maliciously harassing, which may affect the quality of the conversation.

[0029] To improve conversation quality, the inventors conceived of first filtering out abnormal users who lack clear needs or are maliciously harassing them, and then intercepting these abnormal users. However, filtering based on user conversation messages using keywords or rules has a high false positive rate, and manual filtering is inefficient. After a series of studies, the inventors proposed the technical solution of this application. In the embodiments of this application, the present application detects conversation requests sent by a first user terminal. The conversation request is generated based on a conversation initiation operation triggered by the first user against a second user. In response to the conversation request, an intelligent conversation robot conducts multiple rounds of conversations with the first user terminal, and based on at least one conversation message sent by the first user terminal, it detects whether the first user meets the first quality requirement and obtains a first detection result. If the first detection result is yes, the historical message records generated by the multiple rounds of conversations and the conversation message currently sent by the first user terminal are sent to the second user terminal.

[0030] By utilizing an intelligent chatbot to handle conversation messages from the first user and engaging in multi-turn dialogues with it, the system can detect in real-time whether the first user meets the primary quality requirements, thus improving the efficiency of conversation quality detection. Furthermore, using an intelligent chatbot to handle conversation messages from the first user allows for quality verification without the second user's awareness, preventing the second user from conversing with abnormal users and improving conversation quality and user experience. In addition, if the first user meets the secondary quality requirements, indicating that the first user is not abnormal, historical message records and current conversation messages can be sent to the second user. The second user can then obtain the complete conversation context and communicate with the first user, achieving a seamless switch from the intelligent chatbot to the second user and ensuring service continuity.

[0031] 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.

[0032] It should be noted that the technical solutions in this application are applicable to virtual network environments, and the users described generally refer to "virtual users." Real users can register user accounts on the server through registration to obtain user identities in the network environment. The same user account can log in to the server through different types of client terminals, enabling the server to identify the same user.

[0033] Interactions between the server and the user can be based on user accounts. The data received or sent by the server to the user is also based on the user account; in reality, the user's client, corresponding to the user account, receives or sends data to the server. Furthermore, users can also communicate with each other through their user accounts. Here, "user" can refer to an individual or an organization, such as a company; this application does not impose specific restrictions.

[0034] Figure 1 A system architecture diagram of a technical solution of an embodiment of this application is shown, which can be applied thereto. The system architecture may include a first user terminal 101, a server terminal 102, and a second user terminal 103.

[0035] The first user terminal 101 and the second user terminal 103 can establish a network connection with the server 102 respectively. The first user terminal 101 and the second user terminal 103 can establish a session connection based on the server, and the first user can interact with messages through the first user terminal 101 and the second user can interact through the second user terminal 103 based on the session connection.

[0036] The first user terminal 101 and the second user terminal 103 can be browsers, apps, web applications such as H5 (HyperText Markup Language 5) applications, lightweight applications (also known as mini-programs), or cloud applications. The first user terminal 101 and the second user terminal 103 can be deployed on electronic devices and require the device to run or certain apps on the device to function. Electronic devices can have displays and support information browsing, such as personal mobile terminals like mobile phones, tablets, personal computers, desktop computers, smart speakers, smartwatches, etc. For ease of understanding... Figure 1 The first user terminal 101 and the second user terminal 103 are primarily represented by the image of devices. Various other types of applications can also be configured in electronic devices, such as human-computer interaction applications, model training applications, text processing applications, web browser applications, shopping applications, search applications, instant messaging tools, email clients, social media platform software, etc. Electronic devices can refer to devices used by users that have the computing, internet access, and communication functions required by the user, such as mobile phones, tablets, personal computers, wearable devices, etc. Electronic devices typically include at least one processing component and at least one storage component. Electronic devices may also include basic configurations such as network interface cards (NICs), I / O buses, and audio / video components; this application does not limit this. Optionally, depending on the implementation of the electronic device, it may also include some peripheral devices, such as keyboards, mice, input pens, printers, etc.; this application does not limit this.

[0037] Server 102 may include servers providing various services, such as servers supporting model training or servers processing information sent by users. It should be noted that server 102 can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. The server can also be a server in a distributed system, or a server integrated with blockchain. The server can also be a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology.

[0038] Server 102 can obtain session messages sent by the first user from the first user terminal 101 based on a session connection. Server 102 can use an intelligent chatbot to generate session messages to conduct multiple rounds of sessions with the first user terminal, and can also send the session messages sent by the first user terminal to the second user terminal 103, etc.

[0039] It should be understood that Figure 1 The number of client and server instances shown is merely illustrative. Depending on implementation needs, there can be any number of client and server instances.

[0040] The implementation details of the technical solutions in the embodiments of this application are described in detail below.

[0041] Figure 2 This is a flowchart illustrating an embodiment of a message processing method provided in this application. The technical solution of this embodiment can be executed by a server, which is deployed in an online system. In practical applications, the online system typically consists of a first user terminal, a second user terminal, and a server. The first user terminal and the second user terminal establish connections with the server via a network. The network provides a medium for communication links between the first user terminal, the second user terminal, and the server. The network can include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0042] The first user terminal can be for a first user, and the second user terminal can be for a second user. In some online systems that provide objects for customers to interact with, the second user can publish objects through the second user terminal, while the first user can consume objects through the first user terminal. The first user can be a customer, and the second user can be the object provider offering customer service or the online system's customer service personnel. In object transaction scenarios, the first user can refer to the buyer.

[0043] This method may include the following steps: 201: Detect the session request sent by the first user client.

[0044] The session request can be generated based on the session initiation operation triggered by the first user against the second user.

[0045] The server can respond to a session request, establishing a first session connection between the server and the first user. The server can then send session messages to the first user based on this connection. The server can also respond to a session request by displaying the first session interface corresponding to the first user and the second user on the first user's client.

[0046] In practical applications, in online systems that provide objects for customers to interact with, such as e-commerce platforms, the first user can be a customer, and the second user can be the object provider offering customer service or the online system's customer service personnel. A session initiation operation can be triggered by the first user clicking the second user's customer service session prompt control. Clicking the customer service session prompt control could be done by clicking the prompt control on the object details page of any object provided by the second user, or by clicking the prompt control on the second user's store homepage, and so on. The session initiation operation triggered by the first user clicking the second user's customer service session prompt control can generate a session request, and the server can respond to the session request by displaying the first session interface on the first user's end.

[0047] 202: In response to a session request, the system uses an intelligent chatbot to conduct multiple rounds of conversations with the first user client, and detects whether the first user meets the first quality requirement based on at least one session message sent by the first user client, and obtains the first detection result.

[0048] The session message can be determined by the first user's response to the first user's message input operation in the first session interface. At least one session message can be sent to the server through the first session connection.

[0049] At least one session message may include the session message currently sent based on the first session connection, and may also include previously sent session messages.

[0050] Intelligent chatbots are intelligent processing modules based on artificial intelligence technology. They possess natural language processing and contextual understanding capabilities, dynamically maintaining the conversation state during dialogue and providing coherent and accurate responses based on the conversation history, thereby improving interaction efficiency and the user experience for the first user. In practical applications, intelligent chatbots can invoke one or more intelligent processing models to perform conversational and detection operations.

[0051] Among them, multi-turn conversations can refer to the continuous, context-related interaction process between an intelligent chatbot and the first user.

[0052] 203: If the first detection result is yes, send the historical message records generated by the multi-round session and the session message currently sent by the first user terminal to the second user terminal.

[0053] If the first user meets the first quality requirement, it indicates that the first user is a high-quality user, rather than an abnormal user.

[0054] Sending historical message records and currently sent session messages to the second user allows the second user to obtain the complete session communication context.

[0055] If the first user does not meet the first quality requirement, the intelligent chatbot takes over the chat messages from the first user, conducts multiple rounds of chat with the first user, and detects in real time whether the first user meets the first quality requirement based on at least one chat message.

[0056] In this embodiment, an intelligent chatbot is used to receive conversation messages from the first user and engage in multi-round dialogues with it. This allows for real-time detection of whether the first user meets the first quality requirement, improving the efficiency of conversation quality detection. Furthermore, using the intelligent chatbot to receive conversation messages from the first user allows for quality verification without the second user's awareness, preventing the second user from conversing with abnormal users and improving conversation quality and user experience. Additionally, if the first user meets the second quality requirement, indicating they are not an abnormal user, historical message records and current conversation messages can be sent to the second user. The second user can then obtain the complete conversation context and communicate with the first user, achieving a seamless switch from the intelligent chatbot to the second user and ensuring service continuity.

[0057] In some embodiments, responding to a session request, using an intelligent chatbot to conduct multiple rounds of conversations with a first user terminal, and detecting whether the first user meets a first quality requirement based on at least one session message sent by the first user terminal may include: responding to a session request, determining the user characteristics of the first user, and detecting whether the first user meets a second quality requirement based on the user characteristics, and obtaining a second detection result; if the second detection result is negative, using an intelligent chatbot to conduct multiple rounds of conversations with the first user terminal, and detecting whether the first user meets the first quality requirement based on at least one session message sent by the first user terminal.

[0058] The method may further include: if the second detection result is yes, sending the session message currently sent by the first user terminal to the second user terminal corresponding to the second user.

[0059] This embodiment can perform preliminary screening by detecting whether the first user meets the second quality requirement based on user characteristics before using the intelligent chatbot to detect whether the first user meets the first quality requirement based on the chat messages. If the first user meets the first quality requirement, the chat messages of the first user are not intercepted, and the chat messages currently sent by the first user are sent to the second user, and the first user can directly chat with the second user. If the first user does not meet the first quality requirement, the first user is initially determined to be an abnormal user, and the chat messages of the first user are intercepted. The intelligent chatbot takes over the chat messages of the first user and detects whether the first user meets the second quality requirement, which can ensure the accuracy of interception and avoid misjudging the chat quality of the first user.

[0060] In some embodiments, determining the user characteristics of the first user may include: extracting user characteristics from the first user's first user behavior data; the user behavior data includes source channel data, historical order data, and / or account attribute data.

[0061] Source channel data refers to information about how the first user discovered the second user, such as social media, search engines, or in-system promotions. Historical order data can include information about past orders, such as order time, frequency, amount, products, and / or payment methods. Account attribute data can include basic information such as account registration time, the first user's age, gender, and region, and / or account activity.

[0062] Therefore, based on the user characteristics, it can be determined whether the first user meets the second quality requirement, and a second detection result can be obtained.

[0063] In some embodiments, when the second detection result is negative, using an intelligent chatbot to conduct multiple rounds of conversations with the first user terminal, and detecting whether the first user meets the first quality requirement based on at least one conversation message sent by the first user terminal, may include: when the second detection result is negative, obtaining the first conversation message sent by the first user terminal for the first time; detecting whether the first user meets the third quality requirement based on the first conversation message, and obtaining a third detection result; when the third detection result is positive, sending the first conversation message to the second user terminal; when the third detection result is negative, using an intelligent chatbot to conduct multiple rounds of conversations with the first user terminal based on the first conversation message, and detecting whether the first user meets the first quality requirement based on at least one conversation message sent by the first user terminal. For ease of understanding, Figure 3 This diagram illustrates the message processing flow in a real-world application.

[0064] like Figure 3 As shown, the first client can generate a session request in response to the session initiation operation triggered by the first user for the second user, and send the session request to the server (step 301).

[0065] The server 102 can respond to the session request, determine the user characteristics of the first user, and generate a first detection score based on the user characteristics using the first detection model (step 302).

[0066] Determine whether the first detection score is greater than the first detection threshold to obtain a second detection result indicating whether the first user meets the second quality requirement (step 303); wherein, if the first detection score is greater than the first detection threshold, it is determined that the first user meets the second quality requirement; otherwise, it is determined that the first user does not meet the second quality requirement.

[0067] If the second detection result is yes, the first user's session messages are not intercepted, and the session messages currently sent by the first user are sent to the second user's corresponding second user terminal, so that the first user terminal can directly communicate with the second user terminal (step 304).

[0068] If the second detection result is negative, obtain the first session message sent by the first user terminal for the first time (step 305). Based on the first session message, a second detection score is generated using the second detection model, and it is determined whether the second detection score is greater than the second detection threshold.

[0069] If the second detection score is greater than the second detection threshold, the first user is determined to meet the third quality requirement; otherwise, the first user is determined not to meet the third quality requirement, thus obtaining the third detection result of whether the first user meets the third quality requirement (step 306).

[0070] If the third detection result is yes, the first session message is sent to the second user terminal (step 307).

[0071] If the third detection result is negative, the intelligent chatbot conducts multiple rounds of conversation with the first user based on the first conversation message (step 308), and detects whether the first user meets the first quality requirement based on at least one conversation message sent by the first user (step 309), thereby realizing dynamic evaluation in the conversation.

[0072] If the first user meets the first quality requirement, the historical message records generated by the multi-round session and the session message currently sent by the first user terminal are sent to the second user terminal (step 310).

[0073] If the first user fails to meet the first quality requirement, the intelligent chatbot will continue the conversation with the first user.

[0074] This embodiment can intercept the first session message sent by the first user after detecting that the first user does not meet the second quality requirement. Based on the first session message, it can detect whether the first user meets the third quality requirement. If the first user does not meet the third quality requirement, the intelligent conversational robot takes over the first user's session message, conducts multiple rounds of dialogue with the first user, and detects whether the first user meets the first quality requirement, realizing dynamic evaluation during the dialogue. Through multiple detections, the accuracy of the session quality judgment can be further guaranteed, avoiding misjudgments.

[0075] In some embodiments, detecting whether a first user meets a third quality requirement based on a first session message and obtaining a third detection result may include: constructing a target feature based on at least one of the message content and message type of the first session message, the second user behavior data of the first user, the second detection result corresponding to the second quality requirement, the device information corresponding to the first user's terminal, the historical session detection data of the first user, and the historical risk control data of the first user; and detecting whether the first user meets the third quality requirement based on the target feature and obtaining a third detection result.

[0076] The message type of the first session message can include message format such as text, image, or card (cards can refer to product cards), or message function such as order message or address confirmation message. The second user behavior data can include source channel data, historical order data, and / or account attribute data, and may be the same as or different from the first user behavior data.

[0077] The device information corresponding to the first user terminal can refer to the device information of the terminal device corresponding to the first user terminal, such as device type and device model. Device type can include tablet computers and mobile phones, etc.

[0078] The historical session detection data of the first user can refer to the detection data of historical sessions that are detected as abnormal or of high quality, such as the number of historical sessions that are detected as abnormal or of high quality. This historical session detection data can reflect the session quality of the first user's historical sessions.

[0079] The first user's historical risk control data may include information such as risk level, risk behavior records, and / or risk management records.

[0080] By utilizing multi-dimensional data to construct target features, the accuracy of detection can be improved by detecting whether the first user meets the third quality requirement.

[0081] In some embodiments, after detecting whether the first user meets the third quality requirement and obtaining the third detection result, the method may further include: storing the third detection result.

[0082] Based on the first session message, detecting whether the first user meets the third quality requirement and obtaining the third detection result may include: checking whether there is a third detection result corresponding to the first user in the stored data; if yes, obtaining the third detection result; if no, based on the first session message, detecting whether the first user meets the third quality requirement and obtaining the third detection result.

[0083] The third-party detection results can be stored in a cache or persistent storage medium such as a database. Storing in a cache facilitates fast retrieval, while persistent storage allows for long-term preservation.

[0084] In this embodiment, the third detection result is stored after detecting whether the first user meets the third quality requirement based on the first session message. In the scenario where the first user has sent the first session message but has not continued the session, and the first user subsequently initiates a session request for the second user, the third detection result can be directly retrieved from the stored data without having to detect whether the first user meets the third quality requirement again, which can improve response speed and data processing efficiency.

[0085] In addition, in some embodiments, the method may also store a second detection result. Determining the user characteristics of the first user and, based on the user characteristics, detecting whether the first user meets the second quality requirement, and obtaining the second detection result may include: searching the stored data to see if a second detection result corresponding to the first user exists; if yes, obtaining the second detection result; if no, determining the user characteristics of the first user and, based on the user characteristics, detecting whether the first user meets the second quality requirement, and obtaining the second detection result.

[0086] The second detection result can be stored in a cache. In scenarios where the first user has initiated a session request to the second user without sending a session message, and the first user subsequently initiates another session request to the second user, the second detection result can be retrieved directly from the stored data without needing to re-detect whether the first user meets the second quality requirement, thus improving response speed and data processing efficiency.

[0087] In some embodiments, detecting whether the first user meets the second quality requirement based on user characteristics may include: detecting whether there is a historical manual conversation record between the first user and the second user; if so, sending the conversation message currently sent by the first user terminal to the second user terminal corresponding to the second user; if not, detecting whether the first user meets the second quality requirement based on user characteristics.

[0088] If the first user and the second user have a prior recorded conversation, the current conversation messages sent by the first user will not be intercepted; instead, they can be directly sent to the second user, allowing the first user to communicate directly with the second user. If the first user and the second user have no prior conversation, the system checks whether the first user meets the second quality requirement.

[0089] In some embodiments, detecting whether a first user meets a second quality requirement based on user characteristics includes: Based on user characteristics, the first detection model is used to detect whether the first user meets the second quality requirement.

[0090] The step of detecting whether a first user meets a first quality requirement using a first detection model based on user characteristics may include: generating a first detection score using the first detection model based on user characteristics; determining whether the first detection score is greater than a first detection threshold; wherein, if the first detection score is greater than the first detection threshold, the first user is determined to meet a second quality requirement; otherwise, the first user is determined not to meet the second quality requirement. The first detection model is trained based on sample user characteristics and corresponding training labels indicating whether the second quality requirement is met or not.

[0091] In this system, the training label corresponding to the sample user features that meets the second quality requirement is 1, and the training label corresponding to the sample user features that does not meet the first quality requirement is 0. This first detection score can be the probability value predicted by the first detection model based on the user features, indicating that the first user meets the second quality requirement.

[0092] In addition, as another alternative implementation method, predetermined rules can be used to detect whether the first user meets the first quality requirement.

[0093] In some embodiments, detecting whether a first user meets the third quality requirement based on the first session message may include: detecting whether the first user meets the third quality requirement using a second detection model based on the first session message.

[0094] The process of detecting whether the first user meets the third quality requirement using the second detection model based on the first session message may include: generating a second detection score using the second detection model based on the first session message; determining whether the second detection score is greater than a second detection threshold; wherein, if the second detection score is greater than the second detection threshold, it is determined that the first user meets the third quality requirement, otherwise it is determined that the first user does not meet the second quality requirement.

[0095] The second detection model can be trained based on sample session messages and training labels corresponding to the sample session messages that meet or do not meet the third quality requirement.

[0096] In this model, the training label corresponding to a sample session message that meets the third quality requirement is 1, and the training label corresponding to a sample session message that does not meet the third quality requirement is 0. This second detection score can be the probability value predicted by the second detection model based on the first session message that the first user meets the third quality requirement.

[0097] The first detection threshold can be lower than the second detection threshold. The lower threshold makes the initial screening more lenient, retaining more possibilities and leaving room for subsequent detection. When using the second detection model, the second detection threshold is higher, making the detection more refined. This hierarchical detection can avoid misjudgment and improve detection accuracy.

[0098] Both the second detection model and the first detection model mentioned above can be deep learning models. The number of parameters in the first detection model can be less than that in the second detection model. Using the first detection model can consume fewer computing resources, resulting in high computational efficiency and a low-cost, fast way to filter out low-quality users. More computing resources can then be used for subsequent, more precise detection, maximizing computational efficiency while ensuring detection accuracy. Of course, there is no limit to the number of parameters in the first and second detection models. For example, the first detection model can be a DNN (Deep Neural Network) model, and the second detection model can incorporate a fasttext model (a word vector model). Incorporating a word vector model helps improve the detection accuracy of text-based conversational messages.

[0099] In addition, as another alternative implementation method, predefined rules can be used to detect whether the first user meets the third quality requirement.

[0100] In some embodiments, detecting whether a first user meets a first quality requirement based on at least one session message sent by a first user terminal may include: invoking an intelligent detection model, parsing at least one session message sent by the first user terminal to identify user intent, and detecting whether the first user meets the first quality requirement based on user intent.

[0101] Identifying user intent can include understanding the first user's transaction needs, such as whether they provide the specifications of the goods they want to trade, whether they mention a delivery time, and whether they provide a purchase budget. Identifying user intent can also include assessing the first user's level of expertise, such as whether they use industry terminology and understand basic trade terms. Furthermore, it can include assessing the first user's responsiveness, such as whether they reply promptly and in detail.

[0102] This intelligent detection model can be a deep learning model.

[0103] In some embodiments, the second user can be the object provider.

[0104] Utilizing an intelligent chatbot to conduct multiple rounds of conversations with the first user client can include: using the intelligent chatbot to conduct multiple rounds of conversations with the first user client according to conversation constraints. These conversation constraints can include requirements for guiding the user's transaction needs.

[0105] The requirements for guiding transaction needs may include, for example, requiring the intelligent chatbot to respond based on the product-related information and conversation context of the conversation message from the first user, and guiding the first user to express their transaction needs in accordance with specifications.

[0106] For example, the requirements for guiding object transaction needs may include the following: 1. Answer the buyer's questions and ask them further questions to gather their requirements. The conversation history is in <history>< / history>.

[0107] 2. Don't tell the buyer you're a bot. Respond in a human tone, and you can use emojis. Pay attention to the conversation history in `<history>< / history>` to avoid using duplicate emojis or replying with the same content repeatedly.

[0108] 3. Buyers can send a gift card, but don't say thank you for the gift card.

[0109] 4. You should refer to the information in `<Knowledge>< / Knowledge>`, which contains knowledge about the seller and the product. Do not fabricate information or make any promises, and do not provide contact information not included in `<Knowledge>< / Knowledge>`. You can tell the buyer that you will check the information later.

[0110] 5. The question you want to ask the buyer is in `<question>< / question>` (this is not the question the buyer wants to know). If `<question>< / question>` is empty, you should not ask the question. You should only ask the first question in this round. Even if the buyer doesn't reply, don't ask questions you've already asked in `<History>< / History>`.

[0111] In some embodiments, after sending the current session message from the first user terminal to the second user terminal, the method may further include: if the second user meets the hosting conditions, generating a reply message corresponding to the session message using an intelligent customer service robot. The second user meeting the hosting conditions may mean that the second user is offline or online but has not replied within a predetermined time. This intelligent customer service robot and the aforementioned intelligent session robot may be different intelligent robots. The intelligent session robot is used to have multiple rounds of conversations with the first user and detect whether the first user meets the first quality requirements, determining whether the first user is an abnormal user. The intelligent customer service robot, after determining that the first user is not an abnormal user, and the first user can directly communicate with the second user, provides customer service and engages in conversation with the first user due to the second user's hosting status, thus responding to the first user in a timely manner.

[0112] In some embodiments, determining the user characteristics of the first user in response to a session request may include: determining the user characteristics of the first user if the first user is a first type of test user in response to a session request.

[0113] The method may further include: if the first user is a second type of test user, sending the session message sent by the first user terminal to the second user terminal.

[0114] This embodiment divides users into two categories. The first category of test users undergoes session quality testing. If they fail to meet quality requirements, their messages are blocked; otherwise, they are allowed to communicate directly with the second user. The second category of test users is not subject to session blocking and can communicate directly with the second user. This allows for session quality testing to be implemented on a small scale, avoiding large-scale failures caused by deploying session quality testing to a full-scale deployment. The feasibility of session quality testing can also be verified by comparing the two categories of test users.

[0115] In some embodiments, determining the user characteristics of the first user in response to a session request may include: detecting whether a second user meets the detection requirements in response to a session request; and determining the user characteristics of the first user if the second user meets the detection requirements.

[0116] The method may further include: if the second user fails to meet the detection requirements, sending the session message sent by the first user terminal to the second user terminal.

[0117] If the second user meets the detection requirements, the session quality detection function can be used to perform quality detection on the first user who requests a session with the second user. If the first user does not meet the quality requirements, the message will be intercepted. If the quality requirements are met, the first user can communicate directly with the second user.

[0118] If the second user does not meet the detection requirements, the first user's request to have a session with the second user will not be blocked, and the first user can communicate directly with the second user.

[0119] Meeting the detection requirements for a second user can mean that the second user has enabled the session quality detection function or that the second user has permission to perform session quality detection. The second user can enable or disable the session quality detection function, which helps meet their flexible needs. Granting session quality detection permission only to a select group of users helps implement session quality detection within a small user base, avoiding large-scale failures caused by deploying session quality detection to all users.

[0120] Specifically, if the first session message sent by the first user is not directly delivered to the second user's end, but is responded to by the intelligent chatbot, it indicates that the system has activated the session quality detection function.

[0121] In some embodiments, the method may further include: performing a processing operation based on feedback data from multiple test users, namely: adjusting a first detection threshold; adjusting a second detection threshold; retraining a first detection model; and retraining a second detection model.

[0122] Actual feedback from test users can reveal the applicability of the current detection threshold and detection model in real-world scenarios, thereby allowing for adjustments to the threshold or retraining of the model. This avoids situations where the detection threshold is too high, leading to a large number of users being judged as abnormal, or where the threshold is too low, making it difficult to filter out abnormal users. It also helps prevent inaccurate detection models and enhances the acceptability and fairness of the system.

[0123] For ease of understanding, Figure 4 The signaling flowchart for message processing in a practical application is shown.

[0124] like Figure 4 As shown, the first user terminal 101 can be directed to the first user, and the second user terminal 103 can be directed to the second user. In the object transaction scenario, the first user can be the buyer, and the second user can be the object provider, i.e., the merchant. The first user terminal 101 generates a session request in response to the session initiation operation triggered by the first user for the second user, and sends the session request to the server 102 (step 401).

[0125] Server 102 can respond to a session request, determine the user characteristics of the first user, and generate a first detection score based on the user characteristics using a first detection model. It then determines whether the first detection score is greater than a first detection threshold to obtain a second detection result indicating whether the first user meets the second quality requirement (step 402). If the first detection score is greater than the first detection threshold, the first user is determined to meet the second quality requirement; otherwise, the first user is determined not to meet the second quality requirement. If the second detection result is yes, the server does not intercept the first user's session messages, directly allowing the session messages sent by the first user terminal 101 to pass through, and directly sends the session messages currently sent by the first user terminal 101 to the second user's corresponding second user terminal 103. The first user terminal 101 can then directly communicate with the second user terminal 103 (step 403).

[0126] If the second detection result is negative, the first session message sent by the first user terminal 101 is obtained (step 404). Based on the first session message, a second detection score is generated using the second detection model, and it is determined whether the second detection score is greater than the second detection threshold. If the second detection score is greater than the second detection threshold, it is determined that the first user meets the third quality requirement; otherwise, it is determined that the first user does not meet the third quality requirement, thus obtaining the third detection result of whether the first user meets the third quality requirement (step 405).

[0127] If the third detection result is yes, the first session message will be sent to the second user terminal 103.

[0128] If the third detection result is negative, server 102 uses intelligent chatbot 104 to receive the conversation message sent by first user terminal 101, generates a reply message based on the first conversation message, and conducts multiple rounds of conversation with first user terminal 101 (step 406). Based on at least one conversation message sent by first user terminal 101, server 102 detects whether the first user meets the first quality requirement (step 407). If the first user meets the first quality requirement, server 102 releases the conversation message sent by first user terminal 101 (step 408), and sends the historical message record generated from the multiple rounds of conversation and the conversation message currently sent by first user terminal 101 to second user terminal 103. First user terminal 101 can then directly communicate with second user terminal 103.

[0129] If the first user does not meet the first quality requirement, the intelligent callback robot will continue to communicate with the first user terminal 101.

[0130] This embodiment can intercept the first session message sent by the first user after detecting that the first user does not meet the second quality requirement. Based on the first session message, it can determine whether the first user meets the third quality requirement. If the first user does not meet the third quality requirement, an intelligent chatbot takes over the first user's session message and checks whether the first user meets the first quality requirement. Multiple checks can further ensure the accuracy of session quality judgment and avoid misjudgments.

[0131] Specifically, utilizing an intelligent chatbot to handle conversation messages from the first user and engage in multi-round dialogues with it, while simultaneously detecting whether the first user meets the first quality requirement, can improve the efficiency of conversation quality detection. Furthermore, using an intelligent chatbot to handle conversation messages from the first user allows for quality verification without the second user's awareness, preventing the second user from conversing with abnormal users and thus improving conversation quality and user experience. In addition, if the first user meets the second quality requirement, indicating that the first user is not abnormal, historical message records and current conversation messages can be sent to the second user. The second user can then obtain the complete conversation context and communicate with the first user, achieving a seamless switch from the intelligent chatbot to the second user and ensuring service continuity.

[0132] The detailed implementation methods and beneficial effects of each step in this embodiment have been described in detail in the foregoing embodiments, and will not be elaborated here.

[0133] It should be noted that some processes described in the above embodiments and accompanying drawings include multiple operations appearing in a specific order. However, it should be clearly understood that these operations may not be executed in the order they appear in this document, or they may be executed in parallel. The operation numbers, such as 201, 202, etc., are merely used to distinguish different operations and do not represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should also be noted that the descriptions such as "first" and "second" in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0134] Figure 5 This application provides a schematic diagram of the structure of a message processing apparatus according to one embodiment. The apparatus includes: The first detection module 501 is used to detect the session request sent by the first user terminal; wherein the session request can be generated based on the session initiation operation triggered by the first user against the second user.

[0135] The second detection module 502 is used to respond to a session request, conduct multiple rounds of conversations with the first user terminal using an intelligent chatbot, and detect whether the first user meets the first quality requirement based on at least one session message sent by the first user terminal, and obtain the first detection result.

[0136] The sending module 503 is used to send the historical message records generated by the multi-round session and the session message currently sent by the first user terminal to the second user terminal when the first detection result is yes.

[0137] In some embodiments, the second detection module, in response to a session request, conducts multiple rounds of conversations with the first user terminal using an intelligent chatbot, and detects whether the first user meets the first quality requirement based on at least one session message sent by the first user terminal. This may include: in response to a session request, determining the user characteristics of the first user, and based on the user characteristics, detecting whether the first user meets the second quality requirement, and obtaining a second detection result; if the second detection result is negative, conducting multiple rounds of conversations with the first user terminal using an intelligent chatbot, and detecting whether the first user meets the first quality requirement based on at least one session message sent by the first user terminal.

[0138] The sending module can also be used to: if the second detection result is yes, send the session message currently sent by the first user terminal to the second user terminal corresponding to the second user.

[0139] In some embodiments, the second detection module may determine the user characteristics of the first user by: extracting user characteristics from the first user's first user behavior data; the user behavior data includes source channel data, historical order data, and / or account attribute data.

[0140] In some embodiments, if the second detection module is negative, it may use an intelligent chatbot to conduct multiple rounds of conversations with the first user terminal and detect whether the first user meets the first quality requirement based on at least one conversation message sent by the first user terminal. This may include: if the second detection result is negative, obtaining the first conversation message sent by the first user terminal for the first time; detecting whether the first user meets the third quality requirement based on the first conversation message and obtaining a third detection result; if the third detection result is positive, sending the first conversation message to the second user terminal; if the third detection result is negative, using an intelligent chatbot to conduct multiple rounds of conversations with the first user terminal based on the first conversation message and detecting whether the first user meets the first quality requirement based on at least one conversation message sent by the first user terminal.

[0141] In some embodiments, the second detection module detects whether the first user meets the third quality requirement based on the first session message, and obtaining the third detection result may include: constructing a target feature based on at least one of the message content and message type of the first session message, the second user behavior data of the first user, the second detection result corresponding to the second quality requirement, the device information corresponding to the first user terminal, the historical session detection data of the first user, and the historical risk control data of the first user; and detecting whether the first user meets the third quality requirement based on the target feature to obtain the third detection result.

[0142] In some embodiments, the device may further include a storage module for detecting whether the first user meets the third quality requirement, and storing the third detection result after obtaining the third detection result.

[0143] The second detection module detects whether the first user meets the third quality requirement based on the first session message, and obtains the third detection result by: searching the stored data to see if there is a third detection result corresponding to the first user; if yes, obtaining the third detection result; if no, detecting whether the first user meets the third quality requirement based on the first session message and obtaining the third detection result.

[0144] In addition, in some embodiments, the storage module may also store a second detection result. The second detection module determines the user characteristics of the first user and, based on the user characteristics, detects whether the first user meets the second quality requirements. Obtaining the second detection result may include: searching the stored data to see if a second detection result corresponding to the first user exists; if yes, obtaining the second detection result; if no, determining the user characteristics of the first user and, based on the user characteristics, detecting whether the first user meets the second quality requirements, obtaining the second detection result.

[0145] In some embodiments, the second detection module may detect whether the first user meets the second quality requirement based on user characteristics by: detecting whether there is a historical manual conversation record between the first user and the second user; if so, sending the conversation message currently sent by the first user terminal to the second user terminal corresponding to the second user; if not, detecting whether the first user meets the second quality requirement based on user characteristics.

[0146] In some embodiments, the second detection module detects whether the first user meets the second quality requirement based on user characteristics by: using the first detection model to detect whether the first user meets the second quality requirement based on user characteristics.

[0147] The step of detecting whether a first user meets a first quality requirement using a first detection model based on user characteristics may include: generating a first detection score using the first detection model based on user characteristics; determining whether the first detection score is greater than a first detection threshold; wherein, if the first detection score is greater than the first detection threshold, the first user is determined to meet a second quality requirement; otherwise, the first user is determined not to meet the second quality requirement. The first detection model is trained based on sample user characteristics and corresponding training labels indicating whether the second quality requirement is met or not.

[0148] In some embodiments, the second detection module detecting whether the first user meets the third quality requirement based on the first session message may include: using a second detection model to detect whether the first user meets the third quality requirement based on the first session message.

[0149] The process of detecting whether the first user meets the third quality requirement using the second detection model based on the first session message may include: generating a second detection score using the second detection model based on the first session message; determining whether the second detection score is greater than a second detection threshold; wherein, if the second detection score is greater than the second detection threshold, it is determined that the first user meets the third quality requirement, otherwise it is determined that the first user does not meet the second quality requirement.

[0150] The second detection model can be trained based on sample session messages and training labels corresponding to the sample session messages that meet or do not meet the third quality requirement.

[0151] In this model, the training label corresponding to a sample session message that meets the third quality requirement is 1, and the training label corresponding to a sample session message that does not meet the third quality requirement is 0. This second detection score can be the probability value predicted by the second detection model based on the first session message that the first user meets the third quality requirement.

[0152] In some embodiments, the second detection module may detect whether the first user meets the first quality requirement based on at least one session message sent by the first user terminal, including: calling an intelligent detection model to parse at least one session message sent by the first user terminal to identify the user intent, and detecting whether the first user meets the first quality requirement based on the user intent.

[0153] In some embodiments, the second user can be the object provider. The second detection module's use of an intelligent chatbot to conduct multiple rounds of conversations with the first user client may include: using the intelligent chatbot to conduct multiple rounds of conversations with the first user client according to conversation constraints. These conversation constraints may include object transaction requirement guidance requirements.

[0154] In some embodiments, after the sending module sends the session message currently sent by the first user terminal to the second user terminal corresponding to the second user, the device may further include: an intelligent customer service module, used to generate a reply message corresponding to the session message by using an intelligent customer service robot when the second user meets the hosting conditions.

[0155] In some embodiments, the second detection module, in response to a session request, determining the user characteristics of the first user may include: in response to a session request, determining the user characteristics of the first user if the first user is a first type of test user.

[0156] The sending module can also be used to send session messages sent by the first user terminal to the second user terminal when the first user is a second type of test user.

[0157] In some embodiments, the second detection module, in response to a session request, determines the user characteristics of the first user, which may include: in response to a session request, detecting whether the second user meets the detection requirements; and if the second user meets the detection requirements, determining the user characteristics of the first user.

[0158] The sending module can also be used to send the session message sent by the first user terminal to the second user terminal if the second user fails to meet the detection requirements.

[0159] In some embodiments, the apparatus may further include: an adjustment module for performing a processing operation based on feedback data from multiple test users, namely: adjusting a first detection threshold; adjusting a second detection threshold; retraining a first detection model; and retraining a second detection model.

[0160] Figure 5 The message processing device can perform Figure 2 The implementation principle and technical effects of the message processing method described in the illustrated embodiments will not be repeated here. The specific methods by which each module and unit of the message processing device in the above embodiments performs its operations have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0161] Figure 6 This is a schematic diagram of the structure of one embodiment of a computing device provided in this application. Figure 6 As shown, in practice, the computing device may include a storage component 601 and a processing component 602.

[0162] Storage component 601 is used to store computer programs and can be configured to store various other data to support operation on a computing device. Examples of this data include instructions for any application or method used to operate on the computing device, data structures, contact data, phone book data, messages, pictures, videos, etc.

[0163] Processing component 602, coupled to storage component 601, is used to execute computer programs in storage component 601 for implementing, etc. Figure 2 The message processing method described in the illustrated embodiment.

[0164] Furthermore, such as Figure 6 As shown, the computing device may also include other components such as a communication component 603, a display component 604, a power supply component 605, and an audio component 606. Figure 6 The diagram only shows some components and does not mean that the computing device includes only these components. Figure 6 The components shown. Additionally... Figure 6 The components within the dashed box are optional, not mandatory, and their specific requirements depend on the product form of the computing device. The computing device in this embodiment can be a terminal device such as a desktop computer, laptop computer, smartphone, or IoT (Internet of Things) device, or a server-side device such as a conventional server, cloud server, or server array. If the computing device in this embodiment is implemented as a terminal device such as a desktop computer, laptop computer, or smartphone, it may include... Figure 6 The components within the dashed box; if the computing device in this embodiment is implemented as a conventional server, cloud server, or server array, etc., then it may not include... Figure 6 The component within the dashed box.

[0165] The processing component described above includes one or more processors to execute computer instructions to complete all or part of the steps in the method described above. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the method described above.

[0166] The aforementioned storage components can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0167] The aforementioned communication component is configured to facilitate wired or wireless communication between the device housing the communication component and other devices. The device housing the communication component can access wireless networks based on communication standards, such as mobile communication networks, or combinations thereof. In one exemplary embodiment, the communication component receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel.

[0168] The aforementioned display components may include a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of touch or swipe actions but also the duration and pressure associated with the touch or swipe operation.

[0169] The aforementioned power supply components provide power to various components within the device in which they reside. These power supply components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power to the device in which they reside.

[0170] The aforementioned audio component can be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) configured to receive external audio signals when the device containing the audio component is in an operating mode, such as call mode, recording mode, or voice recognition mode. The received audio signals can be further stored in memory or transmitted via a communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0171] Accordingly, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, enables the processor to implement the steps in the above-described method embodiments. The computer-readable storage medium includes volatile or non-volatile components, or a combination thereof, and can be removable or non-removable. Examples of computer-readable storage media include, but are not limited to, phase-change random access memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random-access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), flash memory or other memory technologies, CD-ROM, Digital Video Disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium. Accordingly, this application also provides a computer program product, which includes a computer program or instructions that, when executed by a processor, cause the processor to implement the steps in the above method embodiments. It should be understood that each step or combination of steps in the above method flow can be implemented by the computer program or instructions. Furthermore, these computer programs or instructions can be applied to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device, enabling the processor of the general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing device to function as an apparatus for implementing the corresponding functions in the above method embodiments.

[0172] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0173] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, or apparatus that includes that element.

[0174] Finally, it should be noted that the above are merely embodiments of this application and are not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A message processing method, characterized in that, include: Detect the session request sent by the first user client; The session request is generated based on a session initiation operation triggered by the first user against the second user; In response to the session request, the intelligent chatbot conducts multiple rounds of conversations with the first user terminal, and based on at least one session message sent by the first user terminal, detects whether the first user meets the first quality requirement and obtains the first detection result; If the first detection result is yes, the historical message records generated by the multi-round session and the session message currently sent by the first user terminal are sent to the second user terminal.

2. The method according to claim 1, characterized in that, The step of responding to the session request by engaging in multiple rounds of conversation with the first user terminal using an intelligent chatbot, and detecting whether the first user meets the first quality requirement based on at least one session message sent by the first user terminal, includes: In response to the session request, the user characteristics of the first user are determined, and based on the user characteristics, it is detected whether the first user meets the second quality requirement, and a second detection result is obtained; If the second detection result is negative, the intelligent chatbot will conduct multiple rounds of conversations with the first user terminal, and based on at least one conversation message sent by the first user terminal, it will be detected whether the first user meets the first quality requirement. The method further includes: If the first detection result is yes, the session message currently sent by the first user terminal is sent to the second user terminal corresponding to the second user.

3. The method according to claim 2, characterized in that, If the second detection result is negative, the step of using an intelligent chatbot to conduct multiple rounds of conversations with the first user terminal, and detecting whether the first user meets the first quality requirement based on at least one conversation message sent by the first user terminal includes: If the first detection result is negative, obtain the first session message sent by the first user terminal for the first time; Based on the first session message, detect whether the first user meets the third quality requirement and obtain the third detection result; If the third detection result is yes, the first session message is sent to the second user terminal; If the third detection result is negative, the intelligent conversational robot conducts multiple rounds of conversations with the first user terminal based on the first conversation message, and detects whether the first user meets the first quality requirement based on at least one conversation message sent by the first user terminal.

4. The method according to claim 3, characterized in that, After detecting whether the first user meets the third quality requirement and obtaining the third detection result, the method further includes: Store the third detection result; The step of detecting whether the first user meets the third quality requirement and obtaining the third detection result based on the first session message includes: Check the stored data to see if there is a third detection result corresponding to the first user; If so, obtain the third detection result; If not, based on the first session message, detect whether the first user meets the third quality requirement and obtain the third detection result.

5. The method according to claim 2, characterized in that, The step of detecting whether the first user meets the second quality requirement based on the user characteristics includes: Detect whether there are any historical human-to-human conversation records between the first user and the second user; If so, send the session message currently sent by the first user terminal to the second user terminal corresponding to the second user; If not, based on the user characteristics, detect whether the first user meets the second quality requirement.

6. The method according to claim 2, characterized in that, The user characteristics of the first user are determined as follows: User features are extracted from the first user's first user behavior data; the user behavior data includes source channel data, historical order data, and / or account attribute data.

7. The method according to claim 3, characterized in that, The step of detecting whether the first user meets the third quality requirement and obtaining the third detection result based on the first session message includes: Based on at least one of the following: the message content of the first session message, the message type of the first session message, the second user behavior data of the first user, the first detection result corresponding to the first quality requirement, the device information corresponding to the first user terminal, the historical session detection data of the first user, and the historical risk control data of the first user, a target feature is constructed. Based on the target characteristics, it is determined whether the first user meets the third quality requirement, and a third detection result is obtained.

8. The method according to claim 3, characterized in that, The step of detecting whether the first user meets the second quality requirement based on the user characteristics includes: Based on the user characteristics, the first detection model is used to detect whether the first user meets the second quality requirement; The step of detecting whether the first user meets the third quality requirement based on the first session message includes: Based on the first session message, the second detection model is used to detect whether the first user meets the third quality requirement.

9. The method according to claim 8, characterized in that, The step of detecting whether the first user meets the second quality requirement using the first detection model based on the user characteristics includes: Based on the user characteristics, the first detection model is used to generate a first detection score; Determine whether the first detection score is greater than a first detection threshold; wherein, if the first detection score is greater than the first detection threshold, determine that the first user meets the first quality requirement, otherwise determine that the first user does not meet the first quality requirement; the first detection model is trained based on the sample user features and the training labels corresponding to the sample user features that meet or do not meet the first quality requirement. The step of detecting whether the first user meets the third quality requirement using the second detection model based on the first session message includes: A second detection score is generated based on the first session message using a second detection model; Determine whether the second detection score is greater than the second detection threshold; wherein, if the second detection score is greater than the second detection threshold, determine that the first user meets the third quality requirement, otherwise determine that the first user does not meet the third quality requirement; the second detection model is trained based on sample session messages and training labels corresponding to the sample session messages that meet or do not meet the third quality requirement.

10. The method according to claim 1, characterized in that, The step of detecting whether the first user meets the first quality requirement based on at least one session message sent by the first user terminal includes: The intelligent detection model is invoked to parse at least one session message sent by the first user terminal to identify the user's intent, and based on the user's intent, to detect whether the first user meets the first quality requirement.

11. The method according to claim 1, characterized in that, The second user is the object provider; the multi-round conversation between the intelligent chatbot and the first user includes: Using an intelligent chatbot, multiple rounds of conversations are conducted with the first user terminal according to conversation constraints; the conversation constraints include object transaction requirement guidance requirements.

12. The method according to claim 2, characterized in that, In response to the session request, determining the user characteristics of the first user includes: In response to the session request, if the first user is a first type of test user, the user characteristics of the first user are determined; The method also includes If the first user is a second type of test user, the session message sent by the first user terminal will be sent to the second user terminal.

13. The method according to claim 2, characterized in that, In response to the session request, determining the user characteristics of the first user includes: In response to the session request, it is detected whether the second user meets the detection requirements; If the second user meets the detection requirements, the user characteristics of the first user are determined; The method further includes: If the second user fails to meet the detection requirements, the session message sent by the first user terminal will be sent to the second user terminal.

14. The method according to claim 9, characterized in that, Also includes: Based on feedback data from multiple test users, perform one of the following processing operations: Adjust the first detection threshold; Adjust the second detection threshold; Retrain the first detection model; Retrain the second detection model.

15. A computing device, characterized in that, This includes processing components and storage components; The storage component stores a computer program; the computer program is invoked and executed by the processing component to implement the message processing method as described in any one of claims 1 to 14.

16. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by the processing component, implements the message processing method as described in any one of claims 1 to 14.

17. A computer program product, characterized in that, It includes a computer program or instructions that, when executed by a processing component, implement the message processing method as described in any one of claims 1 to 14.

Images