UI interaction method and device based on intention recognition, equipment and medium

Abstract

The invention relates to the technical field of UI interaction, and provides a UI interaction method and device based on intention recognition, equipment and a medium. The method comprises the steps of obtaining first user input; identifying a target intention matched with the first user input from a plurality of preset intentions; under the condition that the target intention is a task type intention, determining a plurality of first preset UI elements associated with the target intention and a plurality of associated preset entity types; obtaining user behaviors and environment data, and extracting target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behaviors and the environment data; constructing a first target UI interface based on the plurality of first preset UI elements, the target entity parameters and the user behavior and environment data; and displaying the first target UI interface. According to the scheme, the UI interface can be adjusted in a personalized mode, the use requirement of a user for the personalized UI is met, and the user experience is improved.

Description

Technical Field

[0001] The present application relates to the field of UI interaction technology, and in particular to a UI interaction method, apparatus, device and medium based on intent recognition. Background Art

[0002] With the popularization of smart electronic devices, users' demand for personalized and intelligent user interfaces (UI) is increasing.

[0003] However, current UI designs are mostly fixed and cannot meet users' demand for personalized UI. For example, in the ticket booking task, the UI design of the ticket booking interface is fixed, including controls such as name, contact information, departure place, destination, travel date, and travel time. Users fill in all the required booking information on the ticket booking interface so that the electronic device can perform the ticket booking task. The process of users filling in information is complicated and rigid, which greatly reduces the user experience. Summary of the Invention

[0004] Based on this, it is necessary to provide a UI interaction method, device, equipment and medium based on intent recognition to address the above technical problems.

[0005] An embodiment of the present application provides a UI interaction method based on intent recognition, the method comprising:

[0006] Get the first user input;

[0007] Identifying a target intent matching the first user input from a plurality of preset intents;

[0008] In a case where the target intent is a task-type intent, determining a plurality of first preset UI elements associated with the target intent and a plurality of preset entity types associated with the target intent;

[0009] Obtaining user behavior and environment data, and extracting target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behavior and environment data;

[0010] Constructing a first target UI interface based on the plurality of first preset UI elements, the target entity parameters, and the user behavior and environment data;

[0011] Display the first target UI interface.

[0012] An embodiment of the present application provides a UI interaction device based on intent recognition, the device comprising:

[0013] A first acquisition module, configured to acquire a first user input;

[0014] A first recognition module is configured to recognize a target intention matching the first user input from a plurality of preset intentions;

[0015] A first determining module is configured to, when the target intent is a task-type intent, determine a plurality of first preset UI elements associated with the target intent and a plurality of preset entity types associated with the target intent;

[0016] a second acquisition module, configured to acquire user behavior and environment data, and extract target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behavior and environment data;

[0017] A first construction module is configured to construct a first target UI interface based on the plurality of first preset UI elements, the target entity parameters, and the user behavior and environment data;

[0018] The first display module is used to display the first target UI interface.

[0019] An embodiment of the present application provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the UI interaction method based on intent recognition provided in any embodiment of the present application are implemented.

[0020] An embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the UI interaction method based on intent recognition provided in any embodiment of the present application are implemented.

[0021] The UI interaction method, apparatus, device and medium based on intent recognition provided by the embodiments of the present application can obtain a first user input; identify a target intent that matches the first user input from multiple preset intents; when the target intent is a task-type intent, determine multiple first preset UI elements associated with the target intent and multiple preset entity types associated with the target intent; obtain user behavior and environmental data, and extract target entity parameters corresponding to at least one preset entity type from the first user input and / or user behavior and environmental data; construct a first target UI interface based on multiple first preset UI elements, target entity parameters and user behavior and environmental data; and display the first target UI interface. It can be seen that the above technical solution can personalize the UI interface according to user input and user behavior and environmental data, meet the user's demand for personalized UI use, and improve user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 Schematic diagram of a UI interaction method based on intent recognition in one embodiment;

[0023] Figure 2 A flowchart of an intention function call in one embodiment;

[0024] Figure 3 Schematic diagram of the structure of a UI interaction system based on intent recognition in one embodiment;

[0025] Figure 4 This is a workflow diagram of a UI interaction system based on intent recognition in one embodiment;

[0026] Figure 5 A schematic diagram of a UI interaction interface in one embodiment;

[0027] Figure 6 A schematic diagram of another UI interaction interface in one embodiment;

[0028] Figure 7 A schematic diagram of yet another UI interaction interface in one embodiment;

[0029] Figure 8 A schematic diagram of another UI interaction interface in one embodiment;

[0030] Figure 9 A schematic diagram of a UI interaction interface in one embodiment;

[0031] Figure 10 A schematic diagram of another UI interaction interface in one embodiment;

[0032] Figure 11 A schematic diagram of yet another UI interaction interface in one embodiment;

[0033] Figure 12 A schematic diagram of another UI interaction interface in one embodiment;

[0034] Figure 13 A schematic diagram of a UI interaction interface in one embodiment;

[0035] Figure 14 A schematic diagram of another UI interaction interface in one embodiment;

[0036] Figure 15 Schematic diagram of the structure of a UI interaction device based on intent recognition in one embodiment;

[0037] Figure 16 FIG. 1 is a schematic structural diagram of an electronic device in an embodiment. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0039] In one embodiment, Figure 1 As shown, a UI interaction method based on intent recognition is provided. This method is applicable to UI interaction scenarios and can be performed by a UI interaction device based on intent recognition. The device can be implemented using software and / or hardware methods and can be integrated into an electronic device. The electronic device can be, but is not limited to, various personal computers, laptops, smartphones, and tablets. In this embodiment, the method includes the following steps:

[0040] S101: Obtain first user input.

[0041] Specifically, the first user input is UI interaction data input by the user to the UI interaction device based on intention recognition, wherein the types of UI interaction data include, but are not limited to, text, voice, image and / or gesture.

[0042] Optionally, the UI interaction device based on intent recognition can also perform preliminary processing on the UI interaction data, such as denoising, standardization, etc., so that it can subsequently better understand and analyze the target intent based on the first user input.

[0043] S102: Identify a target intention that matches the first user input from multiple preset intentions.

[0044] Specifically, a UI interaction device based on intent recognition is pre-set with multiple preset intents, including but not limited to booking a flight, navigating to a shopping mall, opening a first e-commerce platform, etc. Thus, based on the first user input, a matching preset intent can be identified from the multiple preset intents, and the identified preset intent can be used as the target intent.

[0045] In some embodiments, S102 includes: S1021 , determining an initial intention based on a first user input.

[0046] Specifically, NLP technology can be used to parse the first user input and identify the intent therein (i.e., the initial intent). Alternatively, a pre-trained first intent classification model can be used to map the first user input to one or more possible intents (i.e., the initial intent). However, this is not limited to this.

[0047] S1022: When the initial intention does not belong to multiple preset intentions, obtain external network data and context information corresponding to the first user input.

[0048] Specifically, if the initial intent doesn't fall within one of the predefined intents, it's ambiguous and the user's precise intent can't be directly determined. Further contextual information, external network data, and user behavior and environmental data are needed to clarify. Of course, if the initial intent falls within multiple predefined intents, the initial intent can be used as the target intent.

[0049] Specifically, external network data refers to various information obtained from the Internet through API interfaces, crawler technology, third-party services, etc. For example, it includes but is not limited to discount information obtained from various e-commerce platforms.

[0050] Specifically, context information refers to the background and associated information related to UI interaction data, such as but not limited to historical UI interaction data (i.e., UI interaction data input into the UI interaction device based on intent recognition in the past), historical user behavior data (i.e., various operating behaviors, usage habits and / or preference settings related to UI interaction data during the user's past use of electronic devices), current application status (i.e., the status of the application related to UI interaction data that the user is currently using), and / or time and location (i.e., time and geographic location information related to UI interaction data), etc.

[0051] S1023. Identify a target intent from multiple preset intentions based on the first user input, context information, external network data, and user behavior and environmental data.

[0052] Specifically, user behavior and environment data include user behavior data and environment data.

[0053] User behavior data refers to the various operational behaviors, usage habits, and / or preference settings of users when using electronic devices. Operational behaviors include, but are not limited to, clicking, scrolling, typing, and browsing pages. Usage habits include, but are not limited to, access time, frequency, duration of stay, and / or frequently used functions. Preference settings include, but are not limited to, UI layout, UI style, UI functions, font size, screen brightness, volume, language selection, notification preferences, etc.

[0054] Environmental data includes physical environment data and device environment data. Physical environment information refers to information about the user's objective real-world location, including, but not limited to, geographic location, time, and / or weather. Device environment information refers to the various hardware and software conditions of the electronic device itself, including, but not limited to, hardware device information, operating system, and / or software application information.

[0055] Specifically, for each preset intent, the UI interaction device based on intent recognition is pre-set with its corresponding preset application scenario. In this way, user behavior data and environmental data can be analyzed to extract user behavior characteristics and patterns. User behavior characteristics refer to specific attributes or indicators extracted from user behavior data, which can help better understand user needs and preferences. For example, user behavior data and environmental data can be analyzed through statistical, cluster analysis and / or machine learning to obtain user behavior characteristics, but are not limited to these. User behavior patterns refer to regular behaviors discovered based on user behavior characteristics. For example, user behavior data and environmental data can be analyzed through machine learning to obtain user behavior patterns. Furthermore, based on the analysis of user behavior characteristics and patterns, first user input, contextual information, and external network data, the current application scenario can be determined, and then a target intent can be identified from multiple preset intents based on the current application scenario. Alternatively, a pre-trained second intent classification model can be used to map the first user input, contextual information, external network data, and user behavior and environmental data to an intent (i.e., the target intent). However, this is not limited to these.

[0056] For example, Figure 2 As shown, the first user input is "I want to buy a mobile phone", and the large model generates the answer "OK, we will provide you with mobile phone buying services", identifying the initial intention as "buy a mobile phone", and "buy a mobile phone" does not belong to multiple preset intentions. The multiple preset intentions include "navigation" and "open the first e-commerce platform". Among them, the preset application scenario corresponding to "navigation to the mall" is expressed by an intent function description, for example, including the user's preference to buy electronic products offline. The preset application scenario corresponding to "open the first e-commerce platform" is expressed by another intent function description, for example, including the first e-commerce platform's recent launch of discount information (for example, in the past 3 days, etc.). If the intent function is predicted by the large model based on the first user input, contextual information, external network data, and user behavior and environmental data, and the current application scenario is that the first e-commerce platform recently launched discount information and the predicted intent function is "navigation to the mall", the function name and other function information of the predicted intent function are extracted, and then the function name is checked to see if it matches. If it matches, it can be determined that the target intent is "navigation".

[0057] It is understandable that existing technologies typically perform intent recognition based solely on user input data, resulting in an inaccurate understanding of the user's intent. This inevitably requires multiple rounds of dialogue and, when necessary, further questioning of the user to obtain an accurate understanding of the intent, significantly reducing the user experience. This application sets multiple preset intents and analyzes the first user input, contextual information, external network data, and user behavior and environmental data. This allows for an accurate understanding of the user's intent through multi-dimensional analysis, thereby improving the accuracy and efficiency of intent recognition.

[0058] S103. When the target intent is a task-type intent, determine a plurality of first preset UI elements associated with the target intent and a plurality of preset entity types associated with the target intent.

[0059] Specifically, a task-type intent refers to a user's desire to complete a specific task by interacting with a UI interaction device based on intent recognition. Task-type intents include, but are not limited to, booking a flight. Of course, if the target intent is a conversation-type intent (a conversation-type intent refers to a user's desire to engage in general conversational communication through interaction with the system, rather than to complete a specific task), it is sufficient to determine a reply message for the first user input and display the reply message.

[0060] Specifically, for each preset intent, a plurality of first preset UI elements associated with it are pre-set in the UI interaction device based on intent recognition. For a preset intent, "the plurality of first preset UI elements associated with it" refers to all UI elements required to complete the specific task related to the preset intent, wherein UI elements refer to various components and controls in the UI interface for interacting with users, helping users to complete the specific tasks associated with the preset intent. For example, if the target intent is "booking a flight ticket", the plurality of first preset UI elements associated with "booking a flight ticket" refer to all UI elements required to complete the task of booking a flight ticket.

[0061] Specifically, for each preset intent, the intent-recognition-based UI interaction device is pre-set with multiple preset entity types associated with it. For a preset intent, the "multiple preset entity types associated with it" refer to all entity types required to complete the specific task related to the preset intent. For example, if the target intent is "book a flight", the multiple preset entity types associated with "book a flight" refer to all entity types required to complete the flight booking task, such as but not limited to name, contact information, departure place, destination, travel date, travel time, etc.

[0062] S104: Obtain user behavior and environment data, and extract target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behavior and environment data.

[0063] Specifically, the target entity parameter is the entity parameter corresponding to the preset entity type. For example, if the first user input is "book a flight from Beijing to Shanghai", the entity parameter corresponding to the departure place "Beijing" and the entity parameter corresponding to the destination "Shanghai" can be extracted from it; if the user behavior and environment data includes the user's personal data entered in the user's history, and the user's personal data includes the name "Zhao Moumou", the entity parameter corresponding to the name "Zhao Moumou" can be extracted from it.

[0064] Specifically, NLP technology can be used to extract target entity parameters corresponding to a preset entity type from the first user input and / or user behavior and environmental data, but is not limited thereto.

[0065] It is understandable that extracting target entity parameters corresponding to a preset entity type from the first user input and / or user behavior and environmental data reduces the user's input burden and is conducive to improving user experience.

[0066] S105: Construct a first target UI interface based on multiple first preset UI elements, target entity parameters, user behavior and environmental data.

[0067] Specifically, user behavior data and environmental data can be analyzed to extract user behavior characteristics and patterns, wherein the user behavior characteristics and patterns include but are not limited to UI design preferences, such as but not limited to UI layout preferences (e.g., preference setting areas for various UI elements, etc.), UI style preferences (e.g., preferred font color and font size, etc.), and UI function preferences (e.g., preferences for functions to be hidden or folded, etc.). Furthermore, at least some of the multiple first preset UI elements and target entity parameters can be arranged in a first target UI interface according to the user's UI design preferences, thereby constructing a first target UI interface.

[0068] S106: Display the first target UI interface.

[0069] In the above-mentioned UI interaction method based on intent recognition, by analyzing the first user input, context information, external network data, and user behavior and environmental data, an accurate understanding of the user's intention is achieved, the accuracy and efficiency of intent recognition are improved, and the UI interface is personalized according to the user input and user behavior and environmental data to meet the user's demand for personalized UI and improve the user experience.

[0070] In one embodiment, S105 includes: S1051. Select at least one first target UI element associated with the target entity parameter from a plurality of first preset UI elements.

[0071] Specifically, the "first target UI element associated with the target entity parameter" includes: a first preset UI element for displaying or editing the target entity parameter, and a first preset UI element for displaying or editing the preset entity type corresponding to the target entity parameter, but is not limited thereto.

[0072] S1052. In the case where there is an unresolved entity type among multiple preset entity types, based on the unresolved entity type, at least one second target UI element associated with the unresolved entity type is selected from multiple first preset UI elements, wherein the unresolved entity type is a preset entity type for which the corresponding target entity parameters have not been extracted.

[0073] Specifically, when there are unresolved entity types among multiple preset entity types, it indicates that some entity parameters corresponding to some preset entity types are missing and the user needs to be guided to supplement them. Figure 2 As shown, after extracting the target entity parameters corresponding to the preset entity type, it is necessary to check whether the entity parameters are missing, that is, whether there are some entity parameters corresponding to the preset entity type that have not been extracted.

[0074] Specifically, the "second target UI element associated with an unresolved entity type" includes, but is not limited to, a first preset UI element for editing entity parameters corresponding to the unresolved entity type and a first preset UI element for displaying the unresolved preset entity type. The second target UI element is used to guide the user to supplement the entity parameters corresponding to the unresolved entity type.

[0075] S1053: Construct a first target UI interface based on the first target UI element, the second target UI element, the target entity parameter, and the user behavior and environment data.

[0076] Specifically, the first target UI element, the second target UI element, and the target entity parameters may be arranged in the first target UI interface according to the user's UI design preference, thereby constructing the first target UI interface.

[0077] Optionally, after S106, the method further includes: S107, obtaining a second user input.

[0078] Specifically, the second user input is UI interaction data that the user further inputs to the UI interaction device based on intent recognition in order to supplement the entity parameters corresponding to the unresolved entity type.

[0079] S108. Extract target entity parameters corresponding to at least one unresolved entity type from the second user input.

[0080] Specifically, NLP technology can be used to extract the target entity parameters corresponding to the unresolved entity type from the second user input, but it is not limited to this. Figure 2 As shown, after extracting the target entity parameters corresponding to the preset entity type, it is also possible to check whether the target entity parameters match the corresponding preset entity type and / or check whether the extracted target entity parameters are valid, but the present invention is not limited thereto.

[0081] S109: Return to the step of selecting at least one first target UI element associated with the target entity parameter from the plurality of first preset UI elements, until no unresolved entity type exists in the plurality of preset entity types.

[0082] Specifically, after extracting the target entity parameters corresponding to at least one unresolved entity type, the currently extracted target entity parameters are updated (at least one target entity parameter corresponding to an unresolved entity type is added). At this time, the process returns to execution S1052, i.e., at least one first target UI element associated with the updated target entity parameter is selected from a plurality of first preset UI elements, thereby updating the first target UI element. Then, if there are still unresolved entity types among the plurality of preset entity types, at least one second target UI element associated with the unresolved entity type is selected from a plurality of first preset UI elements based on the unresolved entity type, thereby updating the second target UI element. Then, based on the updated first target UI element, the updated second target UI element, the updated target entity parameter, and the user behavior and environment data, the first target UI interface is updated, thereby displaying the updated first target UI interface. This process is repeated until the target entity parameters corresponding to each preset entity type associated with the target intent are obtained. Of course, in the process of dynamically refreshing the first target UI interface, personalized animation effects can also be added according to the user's animation effect preferences.

[0083] In this solution, by dynamically refreshing the first target UI interface, only the first target UI element, second target UI element, and target entity parameters relevant to the current task stage are displayed during each stage of executing the specific task associated with the target intent. This reduces the amount of information the user needs to process, making it easier for them to understand and operate the UI interface. Furthermore, by displaying the second target UI element, the user is provided with clear information about the entity parameters that need to be supplemented in the next step, helping them to smoothly provide the target entity parameters corresponding to the preset entity type.

[0084] In one embodiment, the method further includes: S110, when there is no unresolved entity type among multiple preset entity types, based on user behavior and environmental data, determining user behavior characteristics and patterns associated with the target intention and detecting whether there are to-do tasks to obtain to-do task detection results.

[0085] Specifically, user behavior data and environmental data can be analyzed to extract user behavior characteristics and patterns. From all analyzed user behavior characteristics and patterns, those associated with the target intent can be found. Furthermore, user behavior data and environmental data can be analyzed to detect whether there are currently pending tasks. For example, user behavior data can be examined to identify uncompleted tasks, but this is not limited to this.

[0086] S111. Generate candidate recommendation options corresponding to the target intent based on target entity parameters corresponding to multiple preset entity types, user behavior characteristics and patterns, and to-do task detection results.

[0087] Specifically, candidate recommendation options refer to recommended options that are recommended to users for selection in order to complete specific tasks associated with the target intent. For example, the user behavior characteristics and patterns associated with the target intent include task preferences regarding "specific tasks associated with the target intent." Candidate recommendation options are generated based on the task preferences, target entity parameters corresponding to multiple preset entity types, and to-do task detection results. The candidate recommendation options meet the task preferences and target entity parameters corresponding to multiple preset entity types, and do not conflict with the to-do tasks in time. For example, if the target intent is "booking a flight," after obtaining the name, contact information, departure place, destination, travel date, travel time, and other information required for booking a flight, multiple candidate recommended flights are recommended to the user based on the above information, flight booking preferences, and to-do tasks.

[0088] S112: Update the first target UI interface based on the candidate recommendation options.

[0089] Specifically, after determining the candidate recommendation options, the candidate recommendation options are added to the first target UI interface to update the first target UI interface. For example, according to the user's UI design preferences, multiple first target UI elements, target entity parameters corresponding to multiple preset entity types, and candidate recommendation options can be arranged in the first target UI interface to construct the first target UI interface.

[0090] S113: Display the updated first target UI interface.

[0091] In the above scheme, candidate recommendation options are generated based on target entity parameters, user behavior characteristics and patterns, and to-do task detection results, which can make the candidate recommendation options more in line with user preferences and actual needs, that is, it can achieve personalized and accurate recommendations for users, reduce the time users spend searching in massive information, and improve user experience.

[0092] Optionally, after S113, the method further includes: S114, obtaining a third user input.

[0093] Specifically, the third user input is UI interaction data that the user further inputs into the UI interaction device based on intention recognition for the candidate recommendation option.

[0094] S115 . Modify target entity parameters corresponding to at least one preset entity type among the plurality of preset entity types based on the third user input.

[0095] Specifically, new entity parameters corresponding to at least one preset entity type can be extracted from the third user input, thereby modifying the entity parameters corresponding to the preset entity type. Alternatively, although the third user input does not include entity parameters corresponding to the preset entity type, the third user input describes restrictions on the entity parameters corresponding to at least one preset entity type. In this case, the entity parameters corresponding to the preset entity type can be modified based on the restrictions. For example, if the third user input is "I have a one-hour meeting at 4 pm tomorrow," the departure time associated with the "book a flight" intent can be modified so that it is not within the range of 4 pm to 5 pm. However, this is not limited to this.

[0096] S116. Based on the updated target entity parameters, user behavior characteristics, and to-do task detection results corresponding to multiple preset entity types, update the candidate recommendation options corresponding to the target intent.

[0097] Specifically, after the target entity parameters corresponding to the plurality of preset entity types are updated, candidate recommendation options may be regenerated based on the updated target entity parameters, user behavior characteristics, and to-do task detection results.

[0098] S117: Update the first target UI interface based on the updated candidate recommendation options.

[0099] Specifically, the updated candidate recommendation options replace the original candidate recommendation options, thereby updating the first target UI interface.

[0100] S118: Display the updated first target UI interface.

[0101] In the above solution, by timely updating the candidate recommendation options after the target entity parameters are updated, the updated candidate recommendation options are made more in line with user preferences and actual needs.

[0102] Optionally, after S113 or S118, the method further includes: S119, in response to the selection operation, determining a target recommended option corresponding to the selection operation from the candidate recommended options.

[0103] Specifically, the selection operation may be any operation capable of selecting a target recommendation option from candidate recommendation options. For example, the selection operation may include triggering the target recommendation option by means of a mouse, touch, etc., but is not limited thereto.

[0104] S120: Based on the target recommendation options, complete the target task associated with the target intention.

[0105] Specifically, the system automatically completes the target task based on the target recommendation options and target entity parameters corresponding to multiple preset entity types. For example, if the target intent is "book a flight," the system automatically books a flight for the user based on the target recommended flight and the required booking information, such as name, contact information, departure location, destination, travel date, and travel time.

[0106] S121: Determine a plurality of second preset UI elements associated with continuation tasks of the target task.

[0107] Specifically, a continuation task refers to a subsequent task that is further performed after the target task. A continuation task usually relies on the data or results generated by the previous task to be completed, and its purpose is to advance the user's goals or meet the user's further needs. At least one continuation task corresponding to each target task can be pre-set in the UI interaction device based on intent recognition. Moreover, for each continuation task, a plurality of second preset UI elements associated with it are pre-set in the UI interaction device based on intent recognition. For a continuation task, "a plurality of second preset UI elements associated with it" refers to the UI elements required to complete the continuation task.

[0108] Specifically, there are many specific implementation methods for triggering "determining multiple second preset UI elements associated with the continuation task of the target task". Typical examples are described below, but they do not constitute a limitation of the present application. In some examples, specific trigger conditions can be set in advance, and S121 is automatically executed when the specific trigger conditions are met. For example, the target task is to book a plane ticket, the continuation task includes self-service check-in, and the specific trigger condition may include 1 hour before the departure time. In other examples, S121 is automatically executed when the fifth user input is received and the intention associated with the continuation task is extracted from the fifth user input. For example, the target task is to book a plane ticket, and the continuation task includes self-service check-in. When the fifth user input "I want to check in by myself now" is received, S121 is automatically executed. But it is not limited to this.

[0109] S122: Construct a second target UI interface based on multiple second preset UI elements, target recommendation options, target entity parameters corresponding to multiple preset entity types, and user behavior and environment data.

[0110] Specifically, according to the user's UI design preference, at least some of the multiple second preset UI elements, target recommendation options and at least some of the target entity parameters can be arranged in the first target UI interface to construct the second target UI interface.

[0111] S123: Display the second target UI interface.

[0112] In the above solution, by automatically displaying the second target UI interface corresponding to the continued task, the various tasks are not only interrelated but also form a complete service process, thereby improving efficiency and user experience.

[0113] In order to clearly illustrate the UI interaction method based on intent recognition provided in an embodiment of the present application, a specific example is provided below for illustration.

[0114] For example, Figure 3 As shown, the UI interaction device based on intent recognition is integrated with a UI interaction system based on intent recognition. The UI interaction system based on intent recognition includes a user input module, a data analysis module, an intent recognition module and a UI interaction module, wherein the user input module can receive UI interaction data of types such as text input, voice input, image input and gesture input. The data analysis module can obtain contextual information, external network data, environmental data and user behavior data, analyze the collected environmental data and user behavior data, extract user behavior characteristics and patterns, and use the collected data to train and optimize the intent recognition module. The intent recognition module can identify application scenarios, intent recognition, extract entity parameters and fill entity slots. The UI interaction module can display the UI interface. In addition, after the intent recognition module recognizes the user's true target intent and extracts the target entity parameters involved in the target intent, it can analyze whether the target task involved in the target intent conflicts with other to-do tasks, such as time conflicts, so as to recommend suitable candidate recommendation options to the user. The UI interaction module automatically generates a UI interface based on the target intent by combining different target entity parameters. Users can interact with the UI interface multiple times and continuously adjust the input. The data analysis module and intent recognition module further optimize the intent based on the new user input, forming a closed-loop feedback and UI interface change process.

[0115] For example, Figure 4 As shown in the figure, the workflow of the UI interaction system based on intent recognition is as follows: (1) Identify the user's target intention: Figure 5 and Figure 6 As shown in the figure, the user inputs "Help me book a flight ticket from Shenzhen to Beijing tomorrow". The intent recognition module identifies the current scenario as "travel" and determines that the user's target intention is "booking a flight ticket". (2) Extracting entity parameters: Figures 7 to 9As shown, the UI interface for booking air tickets is automatically generated to determine whether all necessary entity parameters related to the target intention have been extracted. For example, the "name" is automatically supplemented through the user's personal data, but the necessary "departure time" is missing. The user is guided to conduct secondary interaction through the UI interface, and the "departure time" can be supplemented by various methods such as inputting text or voice. After the necessary entity parameter "departure time" is supplemented, the UI interaction module immediately completes the update and change of the user interface. (3) Generate candidate recommendation options: As shown Figure 10 As shown in the figure, after confirming the entity parameters required for booking a flight, the system intelligently analyzes the information of each flight that meets the user's requirements, such as meals, delay rate, reputation, price, etc., and simultaneously obtains to-do tasks to analyze whether there are conflicting factors such as time and location. Based on the flight recommendation data, the UI interface is updated and changed again. As shown in the figure above, the UI interface for flight recommendation is changed to facilitate user interaction and confirmation. (4) User intention modification: Figure 11 and Figure 12 As shown in the figure, the UI interface that changes in real time supports multiple rounds of user interaction input. The UI interface guides users to modify their intentions, such as "There is a one-hour meeting at 16:00 tomorrow afternoon." The intention recognition module combines context information to re-identify the user's target intention, recommends a recommended flight that best matches the user's combined input, and updates the flight recommendation UI interface in real time. (5) Tasks involved in executing the target intention: Figure 13 and Figure 14 As shown in the figure, after confirming the recommended flight, the intent recognition module automatically books the flight for the user. Users can interact through various methods, such as voice or text input, to update the UI interface in real time, such as the self-service check-in UI. After completing the check-in process, the UI interaction module automatically updates the user to the boarding UI, allowing them to easily check and board the flight at any time.

[0116] In summary, the UI interaction method based on intent recognition in this application can personalize the UI interface in real time based on user input (compatible with various forms of input such as text, voice, image, gesture, etc.), and can change the displayed UI interface in real time based on further user input. The UI interface update process can also include animation effects to enhance the user experience. It can also accurately predict the user's intention by analyzing user input, and can store and compare past user input to improve the accuracy of intent recognition.

[0117] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 1At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0118] In one embodiment, Figure 15 As shown, a UI interaction device based on intent recognition is provided, including:

[0119] A first acquisition module 1510 is configured to acquire a first user input;

[0120] A first recognition module 1520 is configured to recognize a target intent matching the first user input from a plurality of preset intents;

[0121] A first determining module 1530 is configured to, when the target intent is a task-type intent, determine a plurality of first preset UI elements associated with the target intent and a plurality of preset entity types associated with the target intent;

[0122] A second acquisition module 1540 is configured to acquire user behavior and environment data, and extract target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behavior and environment data;

[0123] A first construction module 1550 is configured to construct a first target UI interface based on the plurality of first preset UI elements, the target entity parameters, and the user behavior and environment data;

[0124] The first display module 1560 is configured to display the first target UI interface.

[0125] Optionally, the first recognition module 1520 is specifically configured to determine an initial intention based on the first user input;

[0126] When the initial intention does not belong to the multiple preset intentions, obtaining external network data and context information corresponding to the first user input;

[0127] The target intent is identified from the multiple preset intents based on the first user input, the context information, the external network data, and the user behavior and environment data.

[0128] Optionally, the first construction module 1550 includes: a first selection submodule, configured to select at least one first target UI element associated with the target entity parameter from the plurality of first preset UI elements;

[0129] a second selection submodule configured to, if an unresolved entity type exists among the plurality of preset entity types, select, based on the unresolved entity type, at least one second target UI element associated with the unresolved entity type from the plurality of first preset UI elements, wherein the unresolved entity type is a preset entity type for which no corresponding target entity parameter has been extracted;

[0130] The first construction submodule is used to construct the first target UI interface based on the first target UI element, the second target UI element, the target entity parameters, and the user behavior and environment data.

[0131] Optionally, the device further includes: a third acquisition module, configured to acquire a second user input;

[0132] A fourth acquisition module, configured to extract at least one target entity parameter corresponding to the unresolved entity type from the second user input;

[0133] The first returning module is configured to return to the step of selecting at least one first target UI element associated with the target entity parameter from the plurality of first preset UI elements until no unresolved entity type exists in the plurality of preset entity types.

[0134] Optionally, the apparatus further comprises: a second determining module configured to, when no unresolved entity type exists among the plurality of preset entity types, determine user behavior characteristics and patterns associated with the target intent based on the user behavior and environment data and detect whether there is a to-do task to obtain a to-do task detection result;

[0135] A first generating module, configured to generate candidate recommendation options corresponding to the target intent based on the target entity parameters corresponding to the multiple preset entity types, the user behavior characteristics and patterns, and the to-do task detection results;

[0136] A first updating module, configured to update the first target UI interface based on the candidate recommendation options;

[0137] The second updating module is used to display the updated first target UI interface.

[0138] Optionally, the device further includes: a fifth acquisition module configured to, after displaying the updated first target UI interface, further include: acquiring a third user input;

[0139] A first modification module, configured to modify target entity parameters corresponding to at least one preset entity type among the plurality of preset entity types based on the third user input;

[0140] A third updating module is configured to update the candidate recommendation options corresponding to the target intent based on the updated target entity parameters corresponding to the multiple preset entity types, the user behavior characteristics and patterns, and the to-do task detection results;

[0141] A fourth updating module, configured to update the first target UI interface based on the updated candidate recommendation options;

[0142] The fifth updating module is used to display the updated first target UI interface.

[0143] Optionally, the device further comprises: a third determining module, configured to determine, in response to a selection operation, a target recommended option corresponding to the selection operation from the candidate recommended options;

[0144] A first execution module is configured to complete a target task associated with the target intention based on the target recommendation option;

[0145] a fourth determining module, configured to determine a plurality of second preset UI elements associated with the continuation tasks of the target task;

[0146] A second construction module is configured to construct a second target UI interface based on the plurality of second preset UI elements, the target recommendation options, the target entity parameters corresponding to the plurality of preset entity types, and the user behavior and environment data;

[0147] The second display module is used to display the second target UI interface.

[0148] In the above-mentioned UI interaction device based on intention recognition, by analyzing the first user input, context information, external network data, and user behavior and environmental data, an accurate understanding of the user's intention is achieved, the accuracy and efficiency of intention recognition are improved, and the UI interface is personalized according to the user input and user behavior and environmental data to meet the user's demand for personalized UI and improve the user experience.

[0149] For the specific limitations of the UI interaction device based on intent recognition, please refer to the limitations of the UI interaction method based on intent recognition above, which will not be repeated here. The various modules in the above-mentioned UI interaction device based on intent recognition can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0150] In one embodiment, an electronic device is provided. The electronic device may be a terminal, and its internal structure diagram may be as follows: Figure 16As shown. The electronic device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the electronic device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a UI interaction method based on intent recognition is implemented. The display screen of the electronic device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the electronic device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the electronic device, or an external keyboard, touchpad or mouse.

[0151] Those skilled in the art will understand that Figure 16 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the electronic device to which the solution of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0152] In one embodiment, the UI interaction device based on intention recognition provided by the present application can be implemented in the form of a computer program, which can be used in Figure 16 The memory of the electronic device may store various program modules constituting the UI interaction device based on intent recognition, such as: Figure 15 The first acquisition module 1510, the first recognition module 1520, the first determination module 1530, the second acquisition module 1540, the first construction module 1550, and the first display module 1560 are shown. The computer program composed of various program modules enables the processor to execute the steps of the UI interaction method based on intent recognition in various embodiments of the present application described in this specification.

[0153] For example, Figure 16 The electronic device shown can be Figure 15The first acquisition module 1510 in the device shown is used to obtain the first user input; the electronic device can use the first identification module 1520 to identify the target intention that matches the first user input from multiple preset intentions; the electronic device can use the first determination module 1530 to determine multiple first preset UI elements associated with the target intention and multiple preset entity types associated with the target intention when the target intention is a task-type intention; the electronic device can use the second acquisition module 1540 to obtain user behavior and environmental data, and extract target entity parameters corresponding to at least one preset entity type from the first user input and / or user behavior and environmental data; the electronic device can use the first construction module 1550 to construct a first target UI interface based on multiple first preset UI elements, target entity parameters and user behavior and environmental data; the electronic device can use the first display module 1560 to display the first target UI interface.

[0154] In one embodiment, an electronic device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the following steps when executing the computer program: obtaining a first user input; identifying a target intent that matches the first user input from multiple preset intentions; when the target intent is a task-type intention, determining multiple first preset UI elements associated with the target intent and multiple preset entity types associated with the target intent; obtaining user behavior and environmental data, and extracting target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behavior and environmental data; constructing a first target UI interface based on the multiple first preset UI elements, target entity parameters and user behavior and environmental data; and displaying the first target UI interface.

[0155] In one embodiment, when the processor executes the computer program, the processor further implements the following steps: identifying a target intent that matches the first user input from a plurality of preset intents, including:

[0156] Based on the first user input, determining an initial intent;

[0157] When the initial intention does not belong to the multiple preset intentions, obtaining external network data and context information corresponding to the first user input;

[0158] From multiple preset intentions, a target intention is identified based on the first user input, context information, external network data, and user behavior and environmental data.

[0159] In one embodiment, when the processor executes the computer program, it further implements the following steps: constructing a first target UI interface based on a plurality of first preset UI elements, target entity parameters, and user behavior and environment data, including:

[0160] Selecting at least one first target UI element associated with the target entity parameter from a plurality of first preset UI elements;

[0161] In the case where there is an unresolved entity type among the multiple preset entity types, selecting at least one second target UI element associated with the unresolved entity type from the multiple first preset UI elements based on the unresolved entity type, wherein the unresolved entity type is a preset entity type for which no corresponding target entity parameters are extracted;

[0162] A first target UI interface is constructed based on the first target UI element, the second target UI element, the target entity parameters, and the user behavior and environment data.

[0163] In one embodiment, when the processor executes the computer program, the processor further implements the following steps: obtaining a second user input;

[0164] extracting a target entity parameter corresponding to at least one unresolved entity type from the second user input;

[0165] Return to the step of selecting at least one first target UI element associated with the target entity parameter from the plurality of first preset UI elements until no unresolved entity type exists in the plurality of preset entity types.

[0166] In one embodiment, when the processor executes the computer program, the processor further implements the following steps: if there is no unresolved entity type among the plurality of preset entity types, determining user behavior characteristics and patterns associated with the target intent based on the user behavior and environment data and detecting whether there is a to-do task to obtain a to-do task detection result;

[0167] Generate candidate recommendation options corresponding to the target intent based on target entity parameters corresponding to multiple preset entity types, user behavior characteristics and patterns, and to-do task detection results;

[0168] Based on the candidate recommendation options, update the first target UI interface;

[0169] Display the updated first target UI interface.

[0170] In one embodiment, when the processor executes the computer program, the following steps are further implemented: after displaying the updated first target UI interface, the following steps are further included:

[0171] Get third user input;

[0172] Based on a third user input, modify a target entity parameter corresponding to at least one preset entity type among a plurality of preset entity types;

[0173] Based on the updated target entity parameters corresponding to multiple preset entity types, user behavior characteristics and patterns, and to-do task detection results, update the candidate recommendation options corresponding to the target intent;

[0174] Based on the updated candidate recommendation options, update the first target UI interface;

[0175] Display the updated first target UI interface.

[0176] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0177] In response to a selection operation, determining a target recommended option corresponding to the selection operation from the candidate recommended options;

[0178] Recommend options based on the target and complete the target task associated with the target intention;

[0179] Determining a plurality of second preset UI elements associated with continuation tasks of the target task;

[0180] Constructing a second target UI interface based on multiple second preset UI elements, target recommendation options, target entity parameters corresponding to multiple preset entity types, and user behavior and environment data;

[0181] Display the second target UI interface.

[0182] In the above-mentioned electronic device, by analyzing the first user input, context information, external network data, and user behavior and environmental data, an accurate understanding of the user's intention is achieved, the accuracy and efficiency of intention recognition are improved, and the UI interface is personalized according to the user input and user behavior and environmental data to meet the user's demand for personalized UI and improve the user experience.

[0183] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented: obtaining a first user input; identifying a target intent that matches the first user input from multiple preset intentions; when the target intent is a task-type intention, determining multiple first preset UI elements associated with the target intent and multiple preset entity types associated with the target intent; obtaining user behavior and environmental data, and extracting target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behavior and environmental data; constructing a first target UI interface based on multiple first preset UI elements, target entity parameters and user behavior and environmental data; and displaying the first target UI interface.

[0184] In one embodiment, when the processor executes the computer program, the processor further implements the following steps: identifying a target intent that matches the first user input from a plurality of preset intents, including:

[0185] Based on the first user input, determining an initial intent;

[0186] When the initial intention does not belong to the multiple preset intentions, obtaining external network data and context information corresponding to the first user input;

[0187] From multiple preset intentions, a target intention is identified based on the first user input, context information, external network data, and user behavior and environmental data.

[0188] In one embodiment, when the processor executes the computer program, it further implements the following steps: constructing a first target UI interface based on a plurality of first preset UI elements, target entity parameters, and user behavior and environment data, including:

[0189] Selecting at least one first target UI element associated with the target entity parameter from a plurality of first preset UI elements;

[0190] In the case where there is an unresolved entity type among the multiple preset entity types, selecting at least one second target UI element associated with the unresolved entity type from the multiple first preset UI elements based on the unresolved entity type, wherein the unresolved entity type is a preset entity type for which no corresponding target entity parameters are extracted;

[0191] A first target UI interface is constructed based on the first target UI element, the second target UI element, target entity parameters, and user behavior and environmental data.

[0192] In one embodiment, when the processor executes the computer program, the processor further implements the following steps: obtaining a second user input;

[0193] extracting a target entity parameter corresponding to at least one unresolved entity type from the second user input;

[0194] Return to the step of selecting at least one first target UI element associated with the target entity parameter from the plurality of first preset UI elements until no unresolved entity type exists in the plurality of preset entity types.

[0195] In one embodiment, when the processor executes the computer program, the processor further implements the following steps: if there is no unresolved entity type among the plurality of preset entity types, determining user behavior characteristics and patterns associated with the target intent based on the user behavior and environment data and detecting whether there is a to-do task to obtain a to-do task detection result;

[0196] Generate candidate recommendation options corresponding to the target intent based on target entity parameters corresponding to multiple preset entity types, user behavior characteristics and patterns, and to-do task detection results;

[0197] Based on the candidate recommendation options, update the first target UI interface;

[0198] Display the updated first target UI interface.

[0199] In one embodiment, when the processor executes the computer program, the following steps are further implemented: after displaying the updated first target UI interface, the following steps are further included:

[0200] Get third user input;

[0201] Based on a third user input, modify a target entity parameter corresponding to at least one preset entity type among a plurality of preset entity types;

[0202] Based on the updated target entity parameters corresponding to multiple preset entity types, user behavior characteristics and patterns, and to-do task detection results, update the candidate recommendation options corresponding to the target intent;

[0203] Based on the updated candidate recommendation options, update the first target UI interface;

[0204] Display the updated first target UI interface.

[0205] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0206] In response to a selection operation, determining a target recommended option corresponding to the selection operation from the candidate recommended options;

[0207] Recommend options based on the target and complete the target task associated with the target intention;

[0208] Determining a plurality of second preset UI elements associated with continuation tasks of the target task;

[0209] Constructing a second target UI interface based on multiple second preset UI elements, target recommendation options, target entity parameters corresponding to multiple preset entity types, and user behavior and environment data;

[0210] Display the second target UI interface.

[0211] In the above-mentioned storable medium, by analyzing the first user input, context information, external network data, and user behavior and environmental data, an accurate understanding of the user's intention is achieved, the accuracy and efficiency of intention recognition are improved, and the UI interface is personalized according to the user input and user behavior and environmental data to meet the user's demand for personalized UI and improve the user experience.

[0212] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static random access memory (SRAM) and dynamic random access memory (DRAM).

[0213] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0214] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A UI interaction method based on intent recognition, characterized in that: The method comprises: Get the first user input; Identifying a target intent that matches the first user input from a plurality of preset intents; In a case where the target intent is a task-type intent, determining a plurality of first preset UI elements associated with the target intent and a plurality of preset entity types associated with the target intent; Obtaining user behavior and environment data, and extracting target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behavior and environment data; Constructing a first target UI interface based on the plurality of first preset UI elements, the target entity parameters, and the user behavior and environment data; Display the first target UI interface.

2. The method according to claim 1, characterized in that The step of identifying a target intent matching the first user input from a plurality of preset intents includes: determining an initial intent based on the first user input; When the initial intention does not belong to the multiple preset intentions, obtaining external network data and context information corresponding to the first user input; The target intent is identified from the multiple preset intents based on the first user input, the context information, the external network data, and the user behavior and environment data.

3. The method according to claim 1, characterized in that The constructing a first target UI interface based on the plurality of first preset UI elements, the target entity parameters, and the user behavior and environment data includes: Selecting at least one first target UI element associated with the target entity parameter from the plurality of first preset UI elements; In a case where there is an unresolved entity type among the plurality of preset entity types, selecting, based on the unresolved entity type, at least one second target UI element associated with the unresolved entity type from the plurality of first preset UI elements, wherein the unresolved entity type is a preset entity type for which no corresponding target entity parameter has been extracted; The first target UI interface is constructed based on the first target UI element, the second target UI element, the target entity parameters, and the user behavior and environment data.

4. The method according to claim 3, characterized in that Also includes: Get the second user input; extracting at least one target entity parameter corresponding to the unresolved entity type from the second user input; Return to the step of selecting at least one first target UI element associated with the target entity parameter from the plurality of first preset UI elements until no unresolved entity type exists in the plurality of preset entity types.

5. The method according to claim 4, characterized in that Also includes: If no unresolved entity type exists among the plurality of preset entity types, determining user behavior characteristics and patterns associated with the target intent based on the user behavior and environment data and detecting whether there is a to-do task to obtain a to-do task detection result; Generate candidate recommendation options corresponding to the target intent based on the target entity parameters corresponding to the multiple preset entity types, the user behavior characteristics and patterns, and the to-do task detection results; Based on the candidate recommendation options, updating the first target UI interface; The updated first target UI interface is displayed.

6. The method according to claim 5, characterized in that After displaying the updated first target UI interface, the method further includes: Get third user input; Based on the third user input, modify the target entity parameters corresponding to at least one preset entity type among the plurality of preset entity types; Based on the updated target entity parameters corresponding to the multiple preset entity types, the user behavior characteristics and patterns, and the to-do task detection results, updating the candidate recommendation options corresponding to the target intent; Based on the updated candidate recommendation options, updating the first target UI interface; The updated first target UI interface is displayed.

7. The method according to claim 5 or 6, characterized in that Also includes: In response to a selection operation, determining a target recommended option corresponding to the selection operation from the candidate recommended options; Based on the target recommendation options, completing the target task associated with the target intention; Determining a plurality of second preset UI elements associated with continuation tasks of the target task; Constructing a second target UI interface based on the plurality of second preset UI elements, the target recommendation options, the target entity parameters corresponding to the plurality of preset entity types, and the user behavior and environment data; Display the second target UI interface.

8. A UI interaction device based on intention recognition, characterized in that: The device comprises: A first acquisition module, configured to acquire a first user input; A first recognition module is configured to recognize a target intention matching the first user input from a plurality of preset intentions; A first determining module is configured to, when the target intent is a task-type intent, determine a plurality of first preset UI elements associated with the target intent and a plurality of preset entity types associated with the target intent; a second acquisition module, configured to acquire user behavior and environment data, and extract target entity parameters corresponding to at least one preset entity type from the first user input and / or the user behavior and environment data; A first construction module is configured to construct a first target UI interface based on the plurality of first preset UI elements, the target entity parameters, and the user behavior and environment data; The first display module is used to display the first target UI interface.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. 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.

Images