Display method and device of intelligent question and answer interface, storage medium and computer equipment

Abstract

The invention discloses a display method and device of an intelligent question-answering interface, a storage medium and computer equipment, relates to the technical field of digital medical treatment, and mainly aims to comprehensively display interaction content and improve dialogue experience and dialogue immersion of a user and an intelligent question-answering system. The method comprises the following steps: in response to an interaction request of a user, determining a target interaction stage with the user; determining an interaction type with the user in the target interaction stage, and determining an intelligent question and answer interface style and an intelligent question and answer interface display form corresponding to the interaction type; and displaying an intelligent question and answer interface interacting with the user based on the intelligent question and answer interface style and the intelligent question and answer interface display form. The method is suitable for intelligent question-answering scenes.

Description

Technical Field

[0001] The present invention relates to the field of digital medical technologies, and particularly to a method, device, storage medium, and computer device for displaying an intelligent question-and-answer interface. Background Art

[0002] With the rapid development of artificial intelligence technology, although the traditional command-based operation method still occupies the mainstream, the conversational operation method has received increasing attention. By simulating the human conversation method and using natural language processing technology and deep learning algorithms to understand and answer users' questions or provide relevant information, the learning burden of users on operations is effectively reduced. Intelligent question-and-answer products can analyze the questions input by users, understand their semantic content, and extract relevant information from a large amount of data to give answers or suggestions in the form of natural language. For example, when a doctor inputs the condition information into the intelligent question-and-answer product of a hospital, the intelligent question-and-answer product can output the medication information of the patient. Based on this, intelligent question-and-answer products based on conversational user interfaces, such as intelligent customer service and intelligent assistant functions, have begun to be piloted and launched one after another.

[0003] Currently, when a user interacts with an intelligent question-and-answer product, all interaction processes are carried out in a single interface that jumps out directly. However, if the single interface is in the form of a floating window, it is difficult for a relatively small floating window to display a large amount of interaction content at the same time, resulting in insufficient screen utilization rate and low information display density, lacking a sense of dialogue immersion, and the direct jump form makes the appearance process of the interface rigid, thus reducing the user experience. Summary of the Invention

[0004] The present invention provides a method, device, storage medium, and computer device for displaying an intelligent question-and-answer interface, mainly capable of improving the information display density of the intelligent question-and-answer interface and the naturalness of connection between interfaces, and enhancing the dialogue experience and dialogue immersion feeling of users with the intelligent question-and-answer system.

[0005] According to the first aspect of the present invention, there is provided a method for displaying an intelligent question-and-answer interface, including:

[0006] Responding to a user's interaction request to determine the target interaction stage of the user;

[0007] Determining the interaction type with the user in the target interaction stage, and determining the intelligent question-and-answer interface style and intelligent question-and-answer interface display form corresponding to the interaction type;

[0008] Displaying an intelligent question-and-answer interface for interacting with the user based on the intelligent question-and-answer interface style and the intelligent question-and-answer interface display form.

[0009] Optionally, determining the interaction type with the user in the target interaction phase, and determining the intelligent Q&A interface style and intelligent Q&A interface display form corresponding to the interaction type, includes:

[0010] In the case where the target interaction phase is the wake-up phase, determining the target wake-up type corresponding to the user;

[0011] Determining the wake-up page style and wake-up page display form corresponding to the target wake-up type, and displaying the wake-up page to the user based on the wake-up page style and the wake-up page display form.

[0012] Optionally, the determining the wake-up page style and wake-up page display form corresponding to the target wake-up type includes:

[0013] In the case where the target wake-up type is the off-screen voice wake-up type, based on the user's voice color information and wake-up content, determining that the wake-up page style corresponding to the off-screen voice wake-up type is a floating window style including a navigation function and an intelligent Q&A robot logo, and the wake-up page display form is a floating form at a preset position on the display screen of the intelligent Q&A terminal;

[0014] In the case where the target wake-up type is the in-screen wake-up type based on the current page, determining the current page including a navigation function and an intelligent Q&A robot logo as the wake-up page corresponding to the in-screen wake-up type;

[0015] In the case where the target wake-up type is the direct dialogue type, determining that the wake-up page style corresponding to the direct dialogue type is a full-screen direct dialogue interface style including an intelligent Q&A robot logo and a dialog box area, and the wake-up page display form is a dialogue interface jump display form.

[0016] Optionally, determining the interaction type with the user in the target interaction phase, and determining the intelligent Q&A interface style and intelligent Q&A interface display form corresponding to the interaction type, includes:

[0017] In the case where the target interaction phase is the dialogue phase, determining the dialogue start type corresponding to the user;

[0018] If the dialogue start type is the button trigger start type based on the full-screen dialogue start interface, determining the full-screen dialogue start interface as the intelligent Q&A interface corresponding to the button trigger start type. If the dialogue start type is a non-button trigger start type, determining the dialogue turn of the user, and determining the intelligent Q&A interface style and intelligent Q&A interface display form corresponding to the dialogue turn.

[0019] Optionally, the determining the dialogue turn of the user includes:

[0020] Obtain the conversation environment data of the user and the input conversation content, determine the conversation background noise level based on the conversation environment data, and parse the conversation content to obtain the number of interrogative words and the voice pause duration in the conversation content;

[0021] Respectively determine the weight coefficients corresponding to the conversation background noise level, the number of interrogative words, and the voice pause duration, and based on the weight coefficients, perform weighted summation on the background noise level, the number of interrogative words, and the voice pause duration to obtain the intention complexity of the conversation content;

[0022] Judge whether the intention complexity is greater than a preset threshold. If so, determine that a multi-round conversation needs to be carried out with the user; otherwise, determine that a single-round conversation needs to be carried out with the user.

[0023] Optionally, the determining the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the conversation round includes:

[0024] In the case of the single-round conversation, based on the semantic information of the conversation content, determine the target demand page corresponding to the conversation content and the jump display form corresponding to the target demand page;

[0025] In the case of the multi-round conversation, dynamically determine the full-screen conversation interface style and the full-screen conversation interface display form, wherein the method for dynamically determining the full-screen conversation interface style and the full-screen conversation interface display form includes:

[0026] Obtain the user feature data of the user and the conversation content in the current conversation round;

[0027] Based on the user feature data and the conversation content in the current conversation round, dynamically adjust the interface style and the interface display form of the current full-screen conversation interface, and use the adjusted interface style and interface display form as the full-screen conversation interface style and the full-screen conversation interface display form of the next conversation round corresponding to the current conversation round.

[0028] Optionally, the method further includes:

[0029] Preprocess the conversation content in the interaction request, extract each key entity in the preprocessed conversation content, and determine the intention feature vector corresponding to the preprocessed conversation content and the entity slot feature vector corresponding to each key entity;

[0030] Match the intention feature vector with the intention codes in the preset intention coding library, and determine the target intention code according to the matching result;

[0031] Based on the above-mentioned target intention encoding, the weight coefficients of the feature vectors of each entity slot are determined respectively, and based on the weight coefficients, the feature vectors of each entity slot are weighted and fused to obtain a reply information matching expression;

[0032] Based on the reply information matching expression, reply information is matched in a preset knowledge base, and the reply information is displayed through the intelligent Q&A interface.

[0033] According to the second aspect of the present invention, there is provided a display device for an intelligent Q&A interface, including:

[0034] A first determination unit, configured to determine a target interaction stage with the user in response to a user interaction request;

[0035] A second determination unit, configured to determine an interaction type with the user in the target interaction stage, and determine an intelligent Q&A interface style and an intelligent Q&A interface display form corresponding to the interaction type;

[0036] A display unit, configured to display an intelligent Q&A interface for interacting with the user based on the intelligent Q&A interface style and the intelligent Q&A interface display form.

[0037] According to the third aspect of the present invention, there is provided a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned display method of the intelligent Q&A interface is implemented.

[0038] According to the fourth aspect of the present invention, there is provided a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, the above-mentioned display method of the intelligent Q&A interface is implemented.

[0039] A display method, device, storage medium, and computer device for an intelligent question-and-answer interface according to the present invention, compared with the current method in which all interaction processes are carried out in a single interface when a user interacts with an intelligent question-and-answer product, the present invention determines a target interaction stage with the user in response to a user interaction request; then determines an interaction type with the user in the target interaction stage, and determines an intelligent question-and-answer interface style and an intelligent question-and-answer interface display form corresponding to the interaction type; finally, displays an intelligent question-and-answer interface for interacting with the user based on the intelligent question-and-answer interface style and the intelligent question-and-answer interface display form. Thus, according to the interaction type in different interaction stages, the intelligent question-and-answer interface style is determined, which can achieve a proper match between the interaction type and the intelligent question-and-answer interface style, that is, different interaction types correspond to different intelligent question-and-answer interface styles. When there is a large amount of interaction content, it can improve the screen utilization rate of the client, thereby ensuring the information display density of the intelligent question-and-answer interface. At the same time, different interaction types correspond to different intelligent question-and-answer interface display forms, which can avoid the blunt problem caused by using a unified jump form to display the interface, thus ensuring the naturalness of the interface connection when switching between intelligent question-and-answer interfaces of different types. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0041] Figure 1 Shows a flowchart of a display method for an intelligent question-and-answer interface provided by an embodiment of the present invention;

[0042] Figure 2 Shows a flowchart of another display method for an intelligent question-and-answer interface provided by an embodiment of the present invention;

[0043] Figure 3 Shows a schematic flowchart of an interaction process between an intelligent question-and-answer system and a user provided by an embodiment of the present invention;

[0044] Figure 4 Shows a schematic structural diagram of a display device for an intelligent question-and-answer interface provided by an embodiment of the present invention;

[0045] Figure 5 Shows a schematic structural diagram of another display device for an intelligent question-and-answer interface provided by an embodiment of the present invention;

[0046] Figure 6 Shows a schematic physical structure diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other.

[0048] Currently, when a user interacts with an intelligent Q&A product, all interaction processes are carried out in a single interface that pops up directly. If the single interface is in the form of a floating window, it is very difficult for a small floating window to display a large amount of interaction content at the same time, resulting in a low information display density, a lack of dialogue immersion, and a rigid interface appearance during the direct jump process, thus reducing the user experience.

[0049] To solve the above problems, an embodiment of the present invention provides a method for displaying an intelligent Q&A interface, as Figure 1 shown, the method includes:

[0050] 101. In response to a user's interaction request, determine the target interaction stage with the user.

[0051] Among them, the target interaction stage includes the wake-up stage of the intelligent Q&A robot and the dialogue stage, etc. For the embodiments of the present invention, when receiving an interaction request from the user and the intelligent Q&A system, it is first necessary to determine the interaction stage between the user and the intelligent Q&A system, and then determine the intelligent Q&A interface style and the intelligent Q&A interface display form based on the interaction stage.

[0052] 102. Determine the interaction type with the user in the target interaction stage, and determine the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the interaction type.

[0053] Among them, the interaction types include the wake-up type of waking up the robot by voice from outside the screen in the wake-up stage, the wake-up type of waking up the robot in the card within the interface by clicking or voice, the direct dialogue type of directly starting a dialogue by clicking the robot button, etc.; the interaction types also include the dialogue type of starting a dialogue by clicking the robot button in the dialogue stage, the dialogue type of starting a dialogue without clicking the robot button, etc.

[0054] For the embodiments of the present invention, through the voice information of the user or the operation information of the intelligent question-answering system, the interaction type in the target interaction stage can be determined. Since different interaction types correspond to different intelligent question-answering interface styles and intelligent question-answering interface display forms, the intelligent question-answering interface style and the intelligent question-answering interface display form can be matched in the intelligent question-answering interface display optimization library based on this interaction type. Finally, based on the intelligent question-answering interface style and the intelligent question-answering interface display form, an intelligent question-answering interface for interacting with the user is presented. Thus, when there is a large amount of interaction content, a larger interaction interface can be displayed, and when there is less interaction content, a smaller interaction interface can be displayed, improving the screen utilization rate and avoiding resource waste. Thereby, the information display density of the intelligent question-answering interface can be ensured. At the same time, different interaction types correspond to different intelligent question-answering interface display forms, which can avoid the rigidity problem caused by using a unified jump form to display the interface. Therefore, the naturalness of the interface connection when switching between different types of intelligent question-answering interfaces can be ensured.

[0055] 103. Display an intelligent question-answering interface for interacting with the user based on the intelligent question-answering interface style and the intelligent question-answering interface display form.

[0056] For the embodiments of the present invention, different interaction stages correspond to different styles of intelligent question-answering interfaces, and the intelligent question-answering interfaces between different interaction stages are transitionally connected through the intelligent question-answering interface display form, so as to ensure the screen utilization rate of the intelligent question-answering system and the naturalness of the connection between different intelligent question-answering interfaces, thereby enhancing the user's dialogue immersion and experience.

[0057] For example, if a doctor wakes up the robot by voice from outside the screen of the intelligent question-answering system, a floating window will appear at the bottom of the intelligent question-answering system screen. The robot is displayed in the floating window, and the robot responds to the doctor's wake-up voice. Then, the doctor inputs the patient's condition information and the demand information for the medication situation, and enters the dialogue stage. At this time, the interface jumps from the floating window interface to the full-screen dialogue interface, and at the same time, the robot in the full-screen dialogue interface outputs the patient's medication information at the corresponding position in the full-screen dialogue interface based on the patient's condition information.

[0058] A display method of an intelligent Q&A interface provided by the present invention, compared with the current way that all interaction processes are carried out in a single interface when a user interacts with an intelligent Q&A product, the present invention determines a target interaction stage with the user in response to a user interaction request; then determines an interaction type with the user in the target interaction stage, and determines an intelligent Q&A interface style and an intelligent Q&A interface display form corresponding to the interaction type; finally, displays an intelligent Q&A interface for interacting with the user based on the intelligent Q&A interface style and the intelligent Q&A interface display form. Thus, according to the interaction type in different interaction stages, the intelligent Q&A interface style is determined, which can achieve a proper match between the interaction type and the intelligent Q&A interface style, that is, different interaction types correspond to different intelligent Q&A interface styles. When there is a lot of interaction content, the screen utilization rate of the client can be improved, so as to ensure the information display density of the intelligent Q&A interface. At the same time, different interaction types correspond to different intelligent Q&A interface display forms, which can avoid the stiffness problem caused by using a unified jump form to display the interface, so as to ensure the naturalness of the interface connection when switching between intelligent Q&A interfaces of different types.

[0059] Further, in order to better illustrate the above process of displaying the intelligent Q&A interface, as a refinement and extension of the above embodiment, the embodiment of the present invention provides another display method of the intelligent Q&A interface, as Figure 2 shown, the method includes:

[0060] 201. In response to a user interaction request, determine a target interaction stage with the user.

[0061] Specifically, the target interaction stage can be determined based on the user's voice information or the operation information of the intelligent Q&A system.

[0062] 202. Determine an interaction type with the user in the target interaction stage, and determine an intelligent Q&A interface style and an intelligent Q&A interface display form corresponding to the interaction type.

[0063] For the embodiments of the present invention, when the target interaction stage is the wake-up stage for waking up the robot in the intelligent Q&A system, it is necessary to determine the interaction type with the user in the wake-up stage, as well as the intelligent Q&A interface style and intelligent Q&A interface display form corresponding to the interaction type. Based on this, step 202 specifically includes: when the target interaction stage is the wake-up stage, determining the target wake-up type corresponding to the user; determining the wake-up page style and wake-up page display form corresponding to the target wake-up type, and displaying the wake-up page to the user based on the wake-up page style and the wake-up page display form. Among them, the method for determining the wake-up page style and wake-up page display form corresponding to the target wake-up type includes: when the target wake-up type is the out-of-screen voice wake-up type, based on the user's voice timbre information and wake-up content, determining that the wake-up page style corresponding to the out-of-screen voice wake-up type is a floating window style including a navigation function and an intelligent Q&A robot logo, and the wake-up page display form is a floating form at a preset position on the display screen of the intelligent Q&A terminal; when the target wake-up type is the in-screen wake-up type based on the current page, determining the current page including the navigation function and the intelligent Q&A robot logo as the wake-up page corresponding to the in-screen wake-up type; when the target wake-up type is the direct dialogue type, determining that the wake-up page style corresponding to the direct dialogue type is a full-screen direct dialogue interface style including an intelligent Q&A robot logo and a dialog box area, and the wake-up page display form is a dialog interface jump display form.

[0064] Among them, the preset position can be determined according to actual requirements. Specifically, if the robot is woken up by voice from outside the screen to start a conversation, information such as the approximate age, gender, and wake-up requirements of the user can be determined based on the user's voice characteristics information and wake-up content. Then, based on information such as age, gender, and wake-up requirements, the page style, page format, page content, etc. are determined. Based on information such as the page style and page format, a floating window page is determined, and the floating window page is displayed in a floating form at the preset position of the intelligent question-and-answer system display screen page, such as in the middle of the bottom. If no specific requirements are extracted from the user's voice information, general navigation information is displayed in the floating window page. If specific requirements are extracted from the user's voice information, personalized navigation information corresponding to the requirements is displayed in the floating window page, such as navigation information about the locations, functions, and doctor profiles of each department in the hospital. Further, if the user clicks or wakes up the robot in the card within the intelligent question-and-answer system interface by voice to start a conversation, stay on the current interface, that is, use the current interface as the wake-up page corresponding to the wake-up type within the screen. At this time, the current page contains navigation information and a robot logo, and the robot can make corresponding responses based on the user's wake-up type to inform the user that the wake-up is successful. Further, if the user clicks the robot button to directly start a conversation, it directly jumps to the full-screen conversation interface, and at the same time, the voice reply of the robot in the full-screen conversation interface is displayed in the dialog box area of the full-screen conversation interface. The embodiments of the present invention set different intelligent question-and-answer interface styles and display forms for different interaction types in the wake-up stage, which can improve the user experience and ensure the smooth progress of the conversation and the naturalness of the interface switching.

[0065] 203. In the case where the target interaction stage is the conversation stage, determine the conversation start type corresponding to the user.

[0066] Among them, the conversation start types in the conversation stage include button-triggered start type and non-button-triggered start type. The button-triggered start type means that the user clicks the robot button to start a conversation with the robot, and the non-button-triggered start type means that the robot is controlled by voice or text to start a conversation with the robot.

[0067] Specifically, in the conversation stage, based on the voice or operation information between the user and the intelligent question-and-answer system, the conversation start type is determined.

[0068] 204. If the conversation start type is the button-triggered start type based on the full-screen conversation start interface, determine the full-screen conversation start interface as the intelligent question-and-answer interface corresponding to the button-triggered start type.

[0069] Specifically, if the user triggers the robot button in the floating window or the wake-up page to initiate a conversation request, then at this time, it jumps to the full-screen conversation startup interface, and has an official conversation with the user in the full-screen conversation startup interface (that is, determines the full-screen conversation startup interface as the intelligent Q&A interface for having a conversation with the user). If the user triggers the robot button in the current full-screen conversation startup interface to initiate a conversation request, then at this time, determines the current full-screen conversation startup interface as the intelligent Q&A interface for interacting with the user, and interacts with the user in the current full-screen conversation startup interface.

[0070] 205. If the conversation startup type is a non-button-triggered startup type, then determine the conversation turn with the user, and determine the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the conversation turn.

[0071] For the embodiments of the present invention, in the case where the conversation startup type is a non-button-triggered startup type, before determining the intelligent Q&A interface, it is first necessary to determine the conversation turn with the user. Based on this, step 205 specifically includes: obtaining the conversation environment data of the user and the input conversation content, determining the conversation background noise level based on the conversation environment data, and parsing the conversation content to obtain the number of interrogative words and the speech pause duration in the conversation content; respectively determining the weight coefficients corresponding to the conversation background noise level, the number of interrogative words, and the speech pause duration, and based on the weight coefficients, performing a weighted sum on the background noise level, the number of interrogative words, and the speech pause duration to obtain the intention complexity of the conversation content; determining whether the intention complexity is greater than a preset threshold. If so, determine that it is necessary to have multiple rounds of conversations with the user, otherwise, determine that it is necessary to have a single-round conversation with the user.

[0072] Among them, the conversation environment data refers to the location environment information where the user is located, such as noise information, voice background information, etc.; the preset threshold is set according to actual needs.

[0073] Specifically, according to the conversation environment data, determine the conversation background noise level, and at the same time determine the number of interrogative words and the speech pause duration in the conversation content input by the user. Multiply the conversation background noise level, the number of interrogative words, and the speech pause duration by their corresponding weight coefficients respectively, and add the products to obtain the intention complexity of the conversation content. If the intention complexity is greater than the preset threshold, determine that it is necessary to have multiple rounds of conversations with the user (two or more rounds of conversations). If the intention complexity is less than or equal to the preset threshold, determine that it is necessary to have a single-round conversation with the user.

[0074] Further, after determining the conversation turn, it is also necessary to determine the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the conversation turn. Based on this, step 206 specifically includes: in the case of a single-round conversation, based on the semantic information of the conversation content, determine the target demand page corresponding to the conversation content and the jump display form corresponding to the target demand page; in the case of a multi-round conversation, dynamically determine the full-screen conversation interface style and the full-screen conversation interface display form. Among them, the method for dynamically determining the full-screen conversation interface style and the full-screen conversation interface display form includes: obtaining the user's characteristic data and the conversation content in the current conversation turn; based on the user's characteristic data and the conversation content in the current conversation turn, dynamically adjust the interface style and the interface display form of the current full-screen conversation interface, and use the adjusted interface style and interface display form as the full-screen conversation interface style and the full-screen conversation interface display form of the next conversation turn corresponding to the current conversation turn.

[0075] Among them, the user's characteristic data includes information such as the user's age, gender, occupation, interests, and hobbies. The above information is the information provided by the user when registering the intelligent Q&A system and is not user privacy information.

[0076] Specifically, in the case where it is determined that the user's needs can be met through a single-round conversation, based on the conversation content input by the user, display the target demand page corresponding to the conversation content. For example, if the user inputs "Open a certain mall", then directly jump to the page of the certain mall and automatically reply by voice to end the conversation. Further, in the case where it is determined that multiple rounds of conversation are required to meet the user's needs, the conversation with the user needs to be carried out in the full-screen conversation interface at this time, and it is necessary to determine the interface style and the interface display form of the full-screen conversation interface of the next conversation turn according to the user's characteristic data and the conversation content in the current conversation turn. For example, if the current conversation content is related to animated dramas and the user is male and in his twenties, the interface style of the full-screen conversation interface can be themed on the anime style, and the display form can be in the form of frame-by-frame animation transition. If the current conversation content is related to medical information and the user is female and in her fifties, the interface style of the full-screen conversation interface can be themed on warmth and softness, and the display form can be in the form of simple horizontal or vertical cuts. Thus, by selecting the page style and display form based on the user's characteristics and conversation content, the user experience can be improved, and at the same time, the display effect of the conversation content can be enhanced.

[0077] 206. Display an intelligent Q&A interface for interacting with the user based on the intelligent Q&A interface style and the intelligent Q&A interface display form.

[0078] Specifically, after determining the intelligent Q&A interface style and the intelligent Q&A interface display form, finally, an intelligent Q&A interface for interacting with the user is displayed based on the intelligent Q&A interface style and the intelligent Q&A interface display form. Interaction with the user is implemented in the intelligent Q&A interface. During the interaction process, every time the user inputs content, it is necessary to match the reply information in the preset knowledge base. Based on this, the method includes: preprocessing the conversation content in the interaction request, extracting each key entity in the preprocessed conversation content, and determining the intent feature vector corresponding to the preprocessed conversation content and the entity slot feature vectors corresponding to each key entity; matching the intent feature vector with the intent codes in the preset intent encoding library, and determining the target intent code according to the matching result; based on the target intent code, respectively determining the weight coefficients of each entity slot feature vector, and based on the weight coefficients, performing weighted fusion on each entity slot feature vector to obtain a reply information matching expression; based on the reply information matching expression, matching the reply information in the preset knowledge base, and displaying the reply information through the intelligent Q&A interface.

[0079] Specifically, preprocessing is performed on the conversation content in the interaction request, including word segmentation, part-of-speech tagging, stop word filtering, conversation behavior classification, etc. Conversation behavior classification divides the conversation content into types such as questions, answers, confirmations, etc., so as to obtain structured conversation content, that is, the preprocessed conversation content. Then, a natural language model is used to extract the intent feature vector corresponding to the preprocessed conversation content. At the same time, each key entity is determined in the preprocessed conversation content, and the entity slot feature vector of each key entity is determined. Further, a target intent code that matches the intent feature vector is determined in the preset intent code library. Among them, the preset intent code library stores intent codes corresponding to multiple intents. The intent code is a standardized representation of the user's intent and can be a string label. For example, when the user says "I want to listen to Zhou's songs", after the intent feature vector is processed by the model, it highly matches the vector of "play music" in the preset intent code library. Therefore, the target intent code is determined to be "music_play" (play songs). Then, based on the target intent code, weights are assigned to each entity slot feature vector. For example, if the target intent code is "music_play" and the respective keyword entities are singer, time, and location, a higher weight should be assigned to the entity slot feature vector corresponding to the singer. If the target intent code is determined to be "weather_query_current" (query current weather), a higher weight should be assigned to the entity slot feature vector corresponding to the location. Then, based on the weight coefficient corresponding to each entity slot feature vector, each entity slot feature vector is weighted and fused to obtain a reply information matching expression. Finally, based on the reply information matching expression, reply information is matched in the preset knowledge base (the preset knowledge base stores a large amount of professional knowledge, factual data, and empirical knowledge), and the reply information is displayed to the user through the intelligent question-and-answer interface. In the embodiment of the present invention, by converting the conversation content into the form of a reply information matching expression to match the reply information, the structured expression can perform pattern matching faster, reduce processing time, avoid ambiguity in natural language understanding, and improve the accuracy of matching.

[0080] Furthermore, the intelligent question-answering system in the embodiments of the present invention includes a user input module, a natural language recognition module, a dialogue management module, a knowledge base module, a natural language generation module, a speech synthesis module, and a system output module. User input module: responsible for receiving user inputs, such as natural language questions, voices, pictures, etc.; Natural language recognition module: preprocesses the questions input by the user, including steps such as word segmentation, part-of-speech tagging, syntactic analysis, and semantic understanding. These processes convert natural language questions, voices, pictures into machine-understandable formats, providing a basis for subsequent reasoning and matching; Dialogue management module: performs actions such as session node selection, session state migration and maintenance according to the content input by the user. The module can be internally divided into two parts: dialogue state tracking and dialogue strategy. The "dialogue management" module of the intelligent customer service is an intelligent learning algorithm that improves the dialogue reply efficiency and quality of the machine by learning and sorting the user's historical dialogue records; Knowledge base module: the data source of the intelligent question-answering product, storing a large amount of professional knowledge, factual data, and empirical knowledge. The management of the knowledge base involves knowledge acquisition, storage, update, and maintenance; Natural language generation module: once a suitable answer is found, the natural language generation module integrates and sorts the answer and presents it in a user-friendly manner. This can include various forms such as text, voice, image, etc. to meet the needs and preferences of different users; Speech synthesis module: converts the input text information into corresponding voice waveforms to generate audible voice output; System output module: outputs the data or results processed inside the system to the user interface. As Figure 3 shows each process of the intelligent question-answering system interacting with the user.

[0081] According to another display method of the intelligent question-answering interface provided by the present invention, compared with the current way that all interaction processes are carried out in a single interface when the user interacts with the intelligent question-answering product, the present invention determines the target interaction stage with the user in response to the user's interaction request; then determines the interaction type with the user at the target interaction stage, and determines the intelligent question-answering interface style and the intelligent question-answering interface display form corresponding to the interaction type; finally, displays the intelligent question-answering interface for interacting with the user based on the intelligent question-answering interface style and the intelligent question-answering interface display form. Thus, according to the interaction type at different interaction stages, the intelligent question-answering interface style is determined, which can achieve an appropriate match between the interaction type and the intelligent question-answering interface style, that is, different interaction types correspond to different intelligent question-answering interface styles. When there is a lot of interaction content, it can improve the screen utilization rate of the client, thereby ensuring the information display density of the intelligent question-answering interface. At the same time, different interaction types correspond to different intelligent question-answering interface display forms, which can avoid the stiffness caused by using a unified jump form to display the interface, and thus ensure the naturalness of the interface connection when switching between intelligent question-answering interfaces of different types.

[0082] Further, as Figure 1 a specific implementation of Figure 4 this, an embodiment of the present invention provides a display device for an intelligent question - answering interface, as

[0083] shown in the figure, the device includes: a first determination unit 31, a second determination unit 32, and a display unit 33.

[0084] The first determination unit 31 can be used to determine the target interaction stage with the user in response to the user's interaction request.

[0085] The second determination unit 32 can be used to determine the interaction type with the user in the target interaction stage, and determine the intelligent question - answering interface style and the intelligent question - answering interface display form corresponding to the interaction type.

[0086] In a specific application scenario, in order to determine the intelligent question - answering interface style and the intelligent question - answering interface display form corresponding to the interaction type, as Figure 5 shown in the figure, the second determination unit 32 includes a determination module 321 and a display module 322.

[0087] The determination module 321 can be used to determine the target wake - up type corresponding to the user when the target interaction stage is the wake - up stage.

[0088] The display module 322 can be used to determine the wake - up page style and the wake - up page display form corresponding to the target wake - up type, and display the wake - up page to the user based on the wake - up page style and the wake - up page display form.

[0089] In a specific application scenario, to determine the wake-up page style and wake-up page display form corresponding to the target wake-up type, the determining module 321 can specifically be used to, when the target wake-up type is the out-of-screen voice wake-up type, determine, based on the user's voiceprint information and wake-up content, that the wake-up page style corresponding to the out-of-screen voice wake-up type is a floating window style including a navigation function and an intelligent Q&A robot logo, and the wake-up page display form is a floating-up form at a preset position on the display screen of the intelligent Q&A terminal; when the target wake-up type is the in-screen wake-up type based on the current page, determine the current page including the navigation function and the intelligent Q&A robot logo as the wake-up page corresponding to the in-screen wake-up type; when the target wake-up type is the direct dialogue type, determine that the wake-up page style corresponding to the direct dialogue type is a full-screen direct dialogue interface style including an intelligent Q&A robot logo and a dialogue box area, and the wake-up page display form is a dialogue interface jump display form.

[0090] In a specific application scenario, to determine the intelligent Q&A interface style and intelligent Q&A interface display form corresponding to the interaction type, the determining module 321 can also be used to, when the target interaction stage is the dialogue stage, determine the dialogue start type corresponding to the user; if the dialogue start type is the button-triggered start type based on the full-screen dialogue start interface, determine the full-screen dialogue start interface as the intelligent Q&A interface corresponding to the button-triggered start type, and if the dialogue start type is a non-button-triggered start type, determine the dialogue turn corresponding to the user, and determine the intelligent Q&A interface style and intelligent Q&A interface display form corresponding to the dialogue turn.

[0091] In a specific application scenario, to determine the dialogue turn corresponding to the user, the determining module 321 can specifically be used to obtain the user's dialogue environment data and the input dialogue content, determine the dialogue background noise level based on the dialogue environment data, and parse the dialogue content to obtain the number of interrogative words and the voice pause duration in the dialogue content; respectively determine the weight coefficients corresponding to the dialogue background noise level, the number of interrogative words, and the voice pause duration, and based on the weight coefficients, perform weighted summation on the background noise level, the number of interrogative words, and the voice pause duration to obtain the intention complexity of the dialogue content; determine whether the intention complexity is greater than a preset threshold, and if so, determine that multiple rounds of dialogue need to be carried out with the user, otherwise, determine that a single round of dialogue needs to be carried out with the user.

[0092] In a specific application scenario, to determine the intelligent question-and-answer interface style and the intelligent question-and-answer interface display form corresponding to the conversation turn, the determination module 321 may specifically be used to, in the case of a single-round conversation, based on the semantic information of the conversation content, determine the target requirement page corresponding to the conversation content and the jump display form corresponding to the target requirement page; in the case of a multi-round conversation, dynamically determine the full-screen conversation interface style and the full-screen conversation interface display form, where the method for dynamically determining the full-screen conversation interface style and the full-screen conversation interface display form includes: obtaining the user feature data of the user and the conversation content in the current conversation turn; based on the user feature data and the conversation content in the current conversation turn, dynamically adjusting the interface style and the interface display form of the current full-screen conversation interface, and using the adjusted interface style and interface display form as the full-screen conversation interface style and the full-screen conversation interface display form of the next conversation turn corresponding to the current conversation turn.

[0093] In a specific application scenario, to perform intelligent interaction with the user, the device further includes an interaction unit 34.

[0094] The interaction unit 34 may be used to preprocess the conversation content in the interaction request, extract each key entity in the preprocessed conversation content, and determine the intent feature vector corresponding to the preprocessed conversation content and the entity slot feature vectors corresponding to each key entity; match the intent feature vector with the intent codes in the preset intent coding library, and determine the target intent code according to the matching result; based on the target intent code, respectively determine the weight coefficients of each entity slot feature vector, and based on the weight coefficients, perform weighted fusion on each entity slot feature vector to obtain a reply information matching expression; based on the reply information matching expression, match the reply information in the preset knowledge base, and display the reply information through the intelligent question-and-answer interface.

[0095] It should be noted that for other corresponding descriptions of each functional module involved in the intelligent question-and-answer interface display device provided in the embodiments of the present invention, reference may be made to Figure 1 the corresponding description of the method shown, which will not be elaborated here.

[0096] Based on the above as Figure 1For the method described above, correspondingly, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the following steps are implemented: in response to a user's interaction request, determine the user's target interaction stage; determine the interaction type with the user in the target interaction stage, and determine the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the interaction type; based on the intelligent Q&A interface style and the intelligent Q&A interface display form, display an intelligent Q&A interface for interacting with the user.

[0097] Based on the above as Figure 1 shown in the method and as Figure 4 shown in the embodiment of the device, an embodiment of the present invention further provides an entity structure diagram of a computer device, as Figure 6 shown. The computer device includes: a processor 41, a memory 42, and a computer program stored on the memory 42 and executable on the processor. Both the memory 42 and the processor 41 are provided on a bus 43. When the processor 41 executes the program, the following steps are implemented: in response to a user's interaction request, determine the user's target interaction stage; determine the interaction type with the user in the target interaction stage, and determine the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the interaction type; based on the intelligent Q&A interface style and the intelligent Q&A interface display form, display an intelligent Q&A interface for interacting with the user.

[0098] Through the technical solution of the present invention, the present invention determines the user's target interaction stage in response to the user's interaction request; then determines the interaction type with the user in the target interaction stage, and determines the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the interaction type; finally, based on the intelligent Q&A interface style and the intelligent Q&A interface display form, displays an intelligent Q&A interface for interacting with the user. Thus, according to the interaction type in different interaction stages, the intelligent Q&A interface style is determined, which can achieve an appropriate match between the interaction type and the intelligent Q&A interface style, that is, different interaction types correspond to different intelligent Q&A interface styles. When there is a large amount of interaction content, the screen utilization rate of the client can be improved, thereby ensuring the information display density of the intelligent Q&A interface. At the same time, different interaction types correspond to different intelligent Q&A interface display forms, which can avoid the rigid problem caused by using a unified jump form to display the interface, thereby ensuring the naturalness of the interface connection when switching between different types of intelligent Q&A interfaces.

[0099] Obviously, those skilled in the art should understand that the various modules or steps of the present invention described above can be implemented by a general-purpose computing device. They can be concentrated on a single computing device or distributed over a network composed of multiple computing devices. Optionally, they can be implemented by program codes executable by the computing device. Thus, they can be stored in a storage device and executed by the computing device. And in some cases, the steps shown or described can be executed in a sequence different from that here, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. In this way, the present invention is not limited to any specific combination of hardware and software.

[0100] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A display method for an intelligent Q&A interface, characterized in that, including: responding to a user's interaction request to determine a target interaction stage of the user; determining an interaction type with the user in the target interaction stage, and determining an intelligent Q&A interface style and an intelligent Q&A interface display form corresponding to the interaction type; displaying an intelligent Q&A interface for interacting with the user based on the intelligent Q&A interface style and the intelligent Q&A interface display form.

2. The method according to claim 1, wherein The determining the interaction type with the user in the target interaction stage, and determining the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the interaction type includes: when the target interaction stage is a wake-up stage, determining a target wake-up type corresponding to the user; determining a wake-up page style and a wake-up page display form corresponding to the target wake-up type, and displaying a wake-up page to the user based on the wake-up page style and the wake-up page display form.

3. The method according to claim 2, wherein The determining the wake-up page style and the wake-up page display form corresponding to the target wake-up type includes: when the target wake-up type is an off-screen voice wake-up type, based on the user's voice timbre information and wake-up content, determining that the wake-up page style corresponding to the off-screen voice wake-up type is a floating window style including a navigation function and an intelligent Q&A robot logo, and the wake-up page display form is a floating form at a preset position on the display screen of the intelligent Q&A terminal; when the target wake-up type is an in-screen wake-up type based on the current page, determining the current page including a navigation function and an intelligent Q&A robot logo as the wake-up page corresponding to the in-screen wake-up type; when the target wake-up type is a direct dialogue type, determining that the wake-up page style corresponding to the direct dialogue type is a full-screen direct dialogue interface style including an intelligent Q&A robot logo and a dialog box area, and the wake-up page display form is a dialogue interface jump display form.

4. The method according to claim 1, characterized in that The determining the interaction type with the user in the target interaction stage, and determining the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the interaction type includes: when the target interaction stage is a dialogue stage, determining a dialogue start type corresponding to the user; if the dialogue start type is a button-triggered start type based on a full-screen dialogue start interface, determining the full-screen dialogue start interface as the intelligent Q&A interface corresponding to the button-triggered start type, and if the dialogue start type is a non-button-triggered start type, determining the dialogue turn of the user, and determining the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the dialogue turn.

5. The method according to claim 4, wherein The determining the dialogue turn of the user includes: obtaining the user's dialogue environment data and the input dialogue content, determining a dialogue background noise level based on the dialogue environment data, and parsing the dialogue content to obtain the number of interrogative words and the voice pause duration in the dialogue content; Determine the weight coefficients corresponding to the background noise level of the conversation, the number of interrogative words, and the speech pause duration respectively, and based on the weight coefficients, perform weighted summation on the background noise level, the number of interrogative words, and the speech pause duration to obtain the intention complexity of the conversation content; Determine whether the intention complexity is greater than a preset threshold. If so, determine that multiple rounds of conversation with the user are required; otherwise, determine that a single-round conversation with the user is required.

6. The method according to claim 5, wherein The determination of the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the conversation round includes: In the case of a single-round conversation, based on the semantic information of the conversation content, determine the target demand page corresponding to the conversation content and the jump display form corresponding to the target demand page; In the case of multiple rounds of conversation, dynamically determine the full-screen conversation interface style and the full-screen conversation interface display form. Among them, the method for dynamically determining the full-screen conversation interface style and the full-screen conversation interface display form includes: Obtain the user characteristic data of the user and the conversation content in the current round of conversation; Based on the user characteristic data and the conversation content in the current round of conversation, dynamically adjust the interface style and the interface display form of the current full-screen conversation interface, and use the adjusted interface style and the interface display form as the full-screen conversation interface style and the full-screen conversation interface display form of the next round of conversation corresponding to the current round of conversation.

7. The method according to claim 1, characterized in that, The method further includes: Preprocess the conversation content in the interaction request, extract each key entity in the preprocessed conversation content, and determine the intention feature vector corresponding to the preprocessed conversation content and the entity slot feature vector corresponding to each key entity; Match the intention feature vector with the intention codes in the preset intention coding library, and determine the target intention code according to the matching result; Based on the target intention code, determine the weight coefficients of each entity slot feature vector respectively, and based on the weight coefficients, perform weighted fusion on each entity slot feature vector to obtain a reply information matching expression; Based on the reply information matching expression, match the reply information in the preset knowledge base and display the reply information through the intelligent Q&A interface.

8. A display device for an intelligent question-and-answer interface, characterized in that, It includes: A first determination unit, configured to respond to an interaction request of a user and determine the target interaction stage with the user; A second determination unit, configured to determine the interaction type with the user in the target interaction stage and determine the intelligent Q&A interface style and the intelligent Q&A interface display form corresponding to the interaction type; A display unit, configured to display an intelligent Q&A interface for interacting with the user based on the intelligent Q&A interface style and the intelligent Q&A interface display form.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images