Task execution methods, devices, electronic equipment and storage media

Abstract

This application discloses a task execution method, apparatus, electronic device, and storage medium, relating to the field of electronic device technology. The specific technical solution is as follows: receiving a user's first input on first information in a first interface; responding to the first input, displaying the task execution result corresponding to a first task intent, wherein the first task intent is determined based on first information and second information, and the second information includes at least one of the following: interface context information of the first interface and user historical behavior data.

Description

Technical Field

[0001] This application belongs to the field of electronic equipment technology, and specifically relates to a task execution method, apparatus, electronic device and storage medium. Background Technology

[0002] Currently, as the functions of electronic devices continue to improve, users' demand for services provided by these devices is also increasing. For example, users can use electronic devices to handle tasks such as filling out office forms, cross-language communication, and scheduling.

[0003] In related technologies, users typically need to perform a series of operations on electronic devices to trigger them to execute corresponding tasks. For example, after receiving a message from colleague A saying "Meeting at 9 AM tomorrow" through a social media app on their phone, if a user wants to send the message to colleague B, they must first copy the message, then find colleague B in the social media app's contact list, open the chat interface with colleague B, and finally copy the message into the chat's input box and click the send button to trigger the phone to send the message to colleague B. This results in low task execution efficiency for electronic devices. Summary of the Invention

[0004] The purpose of this application is to provide a task execution method, apparatus, electronic device, and storage medium that can improve the task execution efficiency of electronic devices.

[0005] In a first aspect, embodiments of this application provide a task execution method, the method comprising:

[0006] Receive the user's first input on the first information displayed on the first interface;

[0007] In response to the first input, the task execution result corresponding to the first task intent is displayed. The first task intent is determined based on first information and second information. The second information includes at least one of the following: interface context information of the first interface and user historical behavior data.

[0008] Secondly, embodiments of this application provide a task execution apparatus, the apparatus comprising:

[0009] The receiving module is used to receive the user's first input on the first information in the first interface;

[0010] The display module is used to respond to the first input and display the task execution result corresponding to the first task intent. The first task intent is determined based on first information and second information. The second information includes at least one of the following: interface context information of the first interface and user historical behavior data.

[0011] Thirdly, embodiments of this application provide an electronic device including a processor and a memory, wherein the memory stores programs or instructions executable on the processor, and the programs or instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0012] Fourthly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method described in the first aspect.

[0013] Fifthly, embodiments of this application provide a chip, the chip including a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the method as described in the first aspect.

[0014] In a sixth aspect, embodiments of this application provide a computer program product stored in a storage medium, which is executed by at least one processor to implement the method described in the first aspect.

[0015] In this embodiment, the system receives a first input from a user regarding first information on a first interface; in response to the first input, it displays the task execution result corresponding to a first task intent. The first task intent is determined based on first information and second information, whereby the second information includes at least one of the following: interface context information of the first interface and user historical behavior data. Thus, when the electronic device receives user input regarding the first information, it can quickly and accurately identify the user's task intent based on at least one of the first information, the interface context information of the first interface, and user historical behavior data, and then display the task execution result corresponding to the user's task intent, thereby improving the task execution efficiency of the electronic device. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a task execution method provided in some embodiments of this application;

[0017] Figure 2 This is a flowchart illustrating a task execution method provided in some embodiments of this application;

[0018] Figure 3 This is a flowchart illustrating a task execution method provided in some embodiments of this application;

[0019] Figure 4A This is a schematic diagram of the first interface provided in some embodiments of this application;

[0020] Figure 4B These are schematic diagrams illustrating task execution results provided by some embodiments of this application;

[0021] Figure 4C This is a schematic diagram of a control wheel provided in some embodiments of this application;

[0022] Figure 5 This is a flowchart illustrating a task execution method provided in some embodiments of this application;

[0023] Figure 6 These are schematic diagrams of the structure of a task execution device provided in some embodiments of this application;

[0024] Figure 7 These are schematic diagrams of the structure of a task execution device provided in some embodiments of this application;

[0025] Figure 8 These are schematic diagrams of the hardware structure of electronic devices provided in some embodiments of this application;

[0026] Figure 9 These are schematic diagrams of the structure of electronic devices provided in some embodiments of this application. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0028] The terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and not to describe a specific order or sequence. It should be understood that such use of data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class and the number of objects is not limited; for example, a first object can be one or more. Furthermore, in the specification and claims, "and / or" indicates at least one of the connected objects, and the character " / " generally indicates that the preceding and following objects are in an "or" relationship.

[0029] The terms "at least one," "at least one of," etc., used in the specification and claims of this application refer to any one, any two, or a combination of two or more of the included items. For example, at least one of a, b, and c can mean: "a," "b," "c," "a and b," "a and c," "b and c," and "a, b, and c," where a, b, and c can be single or multiple. Similarly, "at least two" refers to two or more items, and its meaning is similar to that of "at least one."

[0030] The following will explain the terminology used in the embodiments of this application.

[0031] Application (APP): An application, or simply an application, is a computer program designed to perform one or more specific tasks. It runs in user mode, can interact with the user, and has a visual user interface.

[0032] In this embodiment, the electronic device may have multiple applications (APPs) installed, and icons of these applications may be displayed on the home screen of the electronic device. The applications in this embodiment may be embedded applications (i.e., system applications of the electronic device) or downloadable applications. Embedded applications are applications provided as part of the implementation of the electronic device. Downloadable applications are applications that can provide their own Internet Protocol Multimedia Subsystem (IMS) connectivity. These downloadable applications may be applications pre-installed on the electronic device or third-party applications downloaded and installed by the user. As an example, applications may include: social networking applications, image management applications (e.g., photo albums), map applications (e.g., maps), browser applications, music applications, etc.

[0033] The main screen of an electronic device can also be called the desktop or main interface. This main screen may include one or more sub-screens, which can be used to display control elements. Alternatively, the desktop of the electronic device may include one or more pages. Users can switch between different sub-screens by swiping left or right. A control is a graphical user interface (GUI) element; it is a software component contained within an application that controls all the data processed by the application and the interactive operations related to that data. Users can interact with controls through direct manipulation to read or edit information related to the application. Generally, controls can include icons (such as application icons or folder icons on an electronic device), buttons, menus, tabs, text boxes, dialog boxes, status bars, navigation bars, widgets, and other visual interface elements.

[0034] Interface: Refers to the medium through which humans interact with electronic devices. The interface allows users to send commands to the system via input devices and receive feedback information via output devices. Input devices can be keyboards, mice, touchscreens, etc.; monitors, speakers, etc.

[0035] User Interface (UI): The user interface (UI) is the bridge for interaction and communication between the system and the user. User interface design refers to the style setting of the graphical user interface, including color scheme, content arrangement, and layout. It includes not only the graphical interfaces of commonly seen desktop and mobile systems, but also new interaction methods such as touchscreens, motion sensing, and voice control. The three main principles of user interface design are: consistency, flexibility, and simplicity.

[0036] Controls: A control (also called a part, component, widget, or control) is a graphical user interface element. It is a basic building block of the user interface, such as a window or text box, and is displayed in the program interface of any application. Controls can be buttons, text boxes, labels, etc., and are used to control all the data processed by each application and the interactive operations on that data.

[0037] A control is an element in a graphical user interface that receives user input to perform corresponding processing or display relevant data. Controls can include, but are not limited to, virtual buttons, sliders, progress bars, and checkboxes.

[0038] Model: A model is a simulation or abstraction of certain characteristics and inherent relationships of objective reality. A model is a role. The concept of a model can be defined as follows: a thing is called a "model" because of its role or purpose in a specific situation—in that situation, it directly or indirectly carries certain attributes of another thing, and based on these attributes, it acts as a substitute or representation of that thing; thus, by using the attributes obtained from the model, the correlation between operations and the corresponding attributes of that thing can be achieved.

[0039] Model training: Model training refers to adjusting model parameters by learning from a large amount of data, enabling the model to accurately predict unknown data. Model training is a continuous process of adjusting model parameters, which requires making full use of the dataset to evaluate model performance in order to obtain a model with good performance and strong generalization ability.

[0040] The task processing method provided in this application embodiment can identify the user's intent and execute the task corresponding to the user's intent during the interaction between the user and the electronic device.

[0041] In related technologies, during user interaction with electronic devices, if the device needs to perform a task, the user typically needs to perform a series of operations on the content displayed on the screen before triggering the device to execute the task related to that content. For example, if a user receives a message from colleague A saying "Meeting at 9 AM tomorrow" through a social media app on their phone, and wants to send the message to colleague B, the user must first copy the message, then find colleague B in the social media app's contact list, open the chat interface with colleague B, and finally copy the message into the chat's input box and click the send button before the phone can send the message to colleague B. This results in low task execution efficiency for electronic devices.

[0042] Therefore, this application provides a task processing method that receives a user's first input on first information; receives the user's first input on first information in a first interface; and, in response to the first input, displays the task execution result corresponding to a first task intent. The first task intent is determined based on first information and second information, whereby the second information includes at least one of the following: interface context information of the first interface and user historical behavior data. Thus, when an electronic device receives user input on first information, it can quickly and accurately identify the user's task intent based on at least one of the first information, the interface context information of the first interface, and user historical behavior data, and then display the task execution result corresponding to the user's task intent, thereby improving the task execution efficiency of the electronic device.

[0043] The task execution method provided in this application can be executed by an electronic device, or at least one of the functional modules and physical modules within the electronic device capable of implementing the task execution method. The specific implementation can be determined based on actual usage requirements, and this application does not impose any limitations. The following explanation uses an electronic device executing the task execution method as an example to illustrate the task execution method provided in this application.

[0044] The task execution method provided in this application will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios. Figure 1 This is a flowchart illustrating the task processing method provided in the embodiments of this application, as shown below. Figure 1 As shown, the task processing method provided in this application embodiment may include the following steps 101 and 102:

[0045] Step 101: The electronic device receives the user's first input on the first information in the first interface.

[0046] In some embodiments of this application, the first interface described above may be the application interface of the first application.

[0047] In some embodiments of this application, the first application described above may include, but is not limited to, any of the following: social networking applications, payment applications, navigation applications, shopping applications, video applications, photo album applications, music applications, etc. The specific application can be determined based on actual usage needs, and this application embodiment does not impose any limitations.

[0048] In some embodiments of this application, the first information described above may include at least one of the following:

[0049] Text, images, and files.

[0050] In some embodiments of this application, the first input is used to instruct the electronic device to identify the user's first task intention regarding the first information and to display the task execution result corresponding to the first task intention.

[0051] In some embodiments of this application, the first input mentioned above can be a user's click input, swipe input, press input, voice input, gesture input, or other feasible inputs, and this application does not limit this.

[0052] In some embodiments of this application, the above-mentioned gesture input may include, but is not limited to, at least one of the following: click gesture, swipe gesture, drag gesture, pressure recognition gesture, long press gesture, area change gesture, double press gesture, double tap gesture, specific gesture input or other possible gesture inputs. The specific gesture input form can be determined according to actual needs, and is not limited in some embodiments.

[0053] In some embodiments of this application, the above-mentioned click input can be single-click input, double-click input, or any number of clicks, or it can be long-press input or short-press input. In some embodiments, this is not limited.

[0054] In some embodiments of this application, the above-mentioned sliding input can be a sliding input in any direction, such as sliding up, sliding down, sliding left, or sliding right, etc., and in some embodiments, this is not limited.

[0055] For example, the first interface mentioned above could be the interface of a certain social media platform, and the first piece of information could be information about products shared by bloggers that the user follows on that social media platform.

[0056] For example, the first interface mentioned above could be the interface of Taobao, and the first piece of information could be detailed information about the good products shared by bloggers that users find on Taobao.

[0057] Step 102: The electronic device responds to the first input and displays the task execution result corresponding to the first task intent.

[0058] In some embodiments of this application, the first task intent is determined based on the first information and the second information, wherein the second information includes at least one of the following: the interface context information of the first interface and user historical behavior data.

[0059] In some embodiments of this application, the aforementioned interface context information may include at least one of the following: the interface content of the aforementioned first interface; the application attribute information of the aforementioned first application; the geographical location information of the electronic device; and second reference information associated with the aforementioned first information;

[0060] The application attribute information mentioned above includes at least one of the following: application name; application type; application function; the second reference information mentioned above is the interface content of the second interface, the second interface mentioned above is the second application interface of the second application accessed by the user in the past, and the second application interface mentioned above includes the first information.

[0061] In some embodiments of this application, the interface content of the first interface may be interface content other than the first information in the first interface.

[0062] In some embodiments of this application, the second application may include, but is not limited to, any of the following: social networking applications, payment applications, navigation applications, shopping applications, video applications, photo album applications, music applications, etc. The specific application can be determined based on actual usage needs, and this application does not impose any limitations.

[0063] In some embodiments of this application, the content of the second interface can be any content in the second interface other than the first information.

[0064] For example, such as Figure 4A As shown, the first interface can be Taobao's interface 40, the first information can be product information on Taobao's interface 40, such as shoe information 411, and the second interface can be a bookstore interface. The product information displayed on Taobao's interface is copied by the user from the bookstore's history and searched on Taobao. The product information is also information shared by bloggers on the bookstore, meaning the bookstore interface includes the first information. Since the phone determines the user's current location is at home based on its geolocation data, after receiving the user's first input of product information 411 on Taobao's interface 40, the phone can determine the user's intention to share product information found on Taobao with friends based on the user's current location and their history of sharing Taobao-found product information with friends from home. Furthermore, since the product is also a recommended item shared by a blogger the user follows on the bookstore. Therefore, the phone can send the product information to the user's friends via WeChat, and combine this with... Figure 4A ,like Figure 4BAs shown, the chat interface 42 between the user and a friend is displayed. This interface includes a message 421 sent by the user to a friend, which includes product information found by the user on Taobao.

[0065] For example, the first interface mentioned above could be a chat interface between the user and colleague A, and the first piece of information mentioned above could be the message "Meeting at 9 AM tomorrow" sent by colleague A, displayed on the first interface. Since the phone determines the user is at the company based on its location data, after receiving the user's initial input of the first piece of information, the phone can determine, based on the user's current location and historical behavior data of forwarding messages sent by colleague A to colleague B while at the company, that the user intends to forward the "Meeting at 9 AM tomorrow" message to colleague B. Therefore, the phone can forward the "Meeting at 9 AM tomorrow" message to colleague B.

[0066] In the task execution method provided in this application embodiment, the electronic device receives a first input from a user on first information in a first interface; responds to the first input and displays the task execution result corresponding to a first task intent. The first task intent is determined based on first information and second information, whereby the second information includes at least one of the following: interface context information of the first interface and user historical behavior data. Thus, when the electronic device receives input from a user on the first information, it can quickly and accurately identify the user's task intent based on at least one of the first information, the interface context information of the first interface, and the user's historical behavior data, and then display the task execution result corresponding to the user's task intent, thereby improving the task execution efficiency of the electronic device.

[0067] In some embodiments of this application, combined with Figure 1 ,like Figure 2 As shown, prior to step 102 above, the task processing method provided in this application embodiment may further include steps 103 and 104, and step 102 above can be implemented through step 102a below:

[0068] Step 103: The electronic device displays at least one task control corresponding to the first task intent.

[0069] In some embodiments of this application, each of the at least one task control described above indicates a task.

[0070] In some embodiments of this application, the at least one task indicated by the at least one task control can be the task corresponding to the first task intent.

[0071] In some embodiments of this application, when the number of tasks indicated by the at least one task control is one, the first task intent can be a single task intent. When the number of tasks indicated by the at least one task control is multiple, the first task intent can be multiple task intents.

[0072] In some embodiments of this application, the electronic device may display at least one task control on the interface of the first application described above.

[0073] In some embodiments of this application, the shape of the task control can be any possible shape such as a circle, rectangle, triangle, rhombus, ring, or polygon, which can be determined according to actual usage requirements. This embodiment of the invention does not limit the shape.

[0074] Step 104: The electronic device receives a second input from the user to a first task control in at least one task control.

[0075] In some embodiments of this application, the first task control described above can be any one of the at least one task control described above.

[0076] In some embodiments of this application, the second input is used to instruct the processing of the task indicated by the first task control and to display the task execution result of the task indicated by the first task control.

[0077] In some embodiments of this application, the second input can be a user's click input, swipe input, press input, voice input, gesture input, or other feasible inputs, and this application does not limit this.

[0078] Step 102a: The electronic device responds to the second input and displays the task execution result of the task indicated by the first task control.

[0079] The first task control can be any one of the at least one task controls.

[0080] In some embodiments of this application, the electronic device may respond to the second input described above, process the task indicated by the first task control, and display the task execution result of the task indicated by the first task control as the task execution result corresponding to the first task intent.

[0081] Thus, by displaying at least one task control corresponding to the first task intent, receiving a second input from the user to the first task control among the at least one task controls, and responding to the second input, displaying the task execution result corresponding to the first task intent, the electronic device is able to display the processing result of the task indicated by the task control selected by the user, thereby improving the flexibility of the electronic device in performing tasks.

[0082] In some embodiments of this application, the at least one task control can form a control wheel; each task control can be a sector control in the control wheel, and the angle of the sector control is determined based on the confidence level of the task intent corresponding to the sector control.

[0083] In some embodiments of this application, the greater the confidence level of the task intent corresponding to the aforementioned sector control, the larger the angle of the sector control.

[0084] In some embodiments of this application, when the electronic device receives a second input from the user to the first sector control, it can also display the sector control in at least one of the following ways:

[0085] Highlight mode, zoom mode.

[0086] For example, combining Figure 4A As shown, when the mobile phone receives the first input 411 regarding product A on the Taobao interface 40, it can respond to the first input and combine it with... Figure 4A ,like Figure 4C As shown, a control wheel 412 is displayed on the Taobao interface 40. The control wheel 412 consists of three fan-shaped controls. The first fan-shaped control can be a task control 4121 for comparing prices of product A across the entire network; the second fan-shaped control can be a task control 4122 for sharing the link of product A with friends; and the third fan-shaped control can be a task control 4123 for purchasing product A. Since the task intent of sharing the link of product A with friends has the highest confidence, the second fan-shaped control has the largest angle. Furthermore, when the mobile phone receives a second input for the task control 4122 to share the link of product A with friends, it can respond to the second input by sending the product information to the user's friends via WeChat, and combine this with... Figure 4C ,like Figure 4B As shown, the chat interface 42 between the user and a friend is displayed. This interface includes a message 421 sent by the user to a friend, which includes product information found by the user on Taobao.

[0087] In this way, the electronic device displays a control wheel composed of at least one task control. Each task control is a sector control in the control wheel, and the angle of the sector control is determined based on the confidence level of the task intent corresponding to each sector control, thereby improving the flexibility of the electronic device in displaying task controls.

[0088] In some embodiments of this application, the aforementioned user historical behavior data is at least one; combined with Figure 1 ,like Figure 3As shown, the task execution method provided in this application embodiment may further include the following steps 105 to 107:

[0089] Step 105: The electronic device acquires the first feature information of each user's historical behavior data.

[0090] In some embodiments of this application, the first input is an input that is dragged from a first position coordinate to a second position coordinate, the first position coordinate is located within the display area of ​​the first information, the second position coordinate is located within a third interface, and the third interface is the application interface of a third application.

[0091] In some embodiments of this application, after the electronic device receives the input dragged from the first position coordinate to the second position coordinate, it starts timing. If the timing duration reaches the second duration and the input position coordinate of the first input is still at the second position coordinate, the electronic device can determine that the user has a task intent recognition requirement and a requirement to execute the task corresponding to the task intent. Then the electronic device can execute the above steps 105 to 107.

[0092] In some embodiments of this application, the electronic device can perform screen recognition operations on the screen of the electronic device at preset time intervals within a first time period to obtain the aforementioned user historical behavior data.

[0093] In some embodiments of this application, a user's historical behavior data may include at least one of the following: the application identifier of the application used by the user, the data type of the target data operated by the user in the application, and the type of operation behavior performed by the user on the target information.

[0094] Step 106: The electronic device acquires the first fused feature information corresponding to the first information and the interface context information.

[0095] In some embodiments of this application, the electronic device can extract features from the first information and the interface context information respectively to obtain feature information of the first information and feature information of the interface context information. Then, the feature information of the first information and the feature information of the interface context information are fused to obtain the first fused feature information corresponding to the first information and the interface context information.

[0096] Step 107: The electronic device determines the first task intent based on the historical behavior data of each user, the first feature information, and the first fused feature information.

[0097] In some embodiments of this application, step 107 described above can be implemented by the following steps 107a to 107d:

[0098] Step 107a: The electronic device acquires the first time decay value of each historical behavior data.

[0099] In some embodiments of this application, the aforementioned first time decay value is used to indicate the importance of user historical behavior data in the time dimension.

[0100] In some embodiments of this application, the electronic device can calculate the time difference between the current time and the generation time of each historical behavior data, and then determine the first time decay value of each historical behavior data based on the time difference corresponding to each historical behavior data.

[0101] In some embodiments of this application, the larger the time difference of a historical behavior data, the greater the first time decay value of the historical data, and the lower the importance of the user's historical behavior data in the time dimension.

[0102] In some embodiments of this application, the electronic device may determine the aforementioned first time attenuation value using the following formula:

[0103] W_time1 = exp(−λ1×Δt); (1)

[0104] Where W_time1 represents the first time decay value mentioned above, Δt represents the time difference mentioned above, and λ1 represents the time decay coefficient.

[0105] Step 107b: The electronic device calculates the first similarity between each first feature information and the first fused feature information.

[0106] In some embodiments of this application, the first feature information of a historical behavior data can be a feature vector of the historical behavior data, and the aforementioned fused feature information can be a fused feature vector corresponding to the first information and the interface context information.

[0107] In some embodiments of this application, the first similarity can be the cosine similarity between the feature vector of each historical behavioral feature data and the fused feature vector.

[0108] In some embodiments of this application, the electronic device can calculate the first similarity using the following formula:

[0109] Sim_i = cosine(V_content, V_memory_i); (2)

[0110] Where Sim_i represents the first similarity mentioned above. V_content represents the first fused feature information mentioned above, and V_memory_i represents the first feature information mentioned above, i∈[1,L]. Where L represents the number of at least one first feature information mentioned above, and i represents the i-th first feature information.

[0111] Step 107c: Based on the historical behavior data of each user on the electronic device, determine at least one historical behavior intent feature.

[0112] In some embodiments of this application, each user's historical behavior data corresponds to historical behavior intent feature information.

[0113] In some embodiments of this application, the electronic device can determine the historical behavior intent feature information corresponding to each user's historical behavior data based on the first feature information of each user's historical behavior data.

[0114] In some embodiments of this application, the electronic device can perform intent recognition on the first feature information of each user's historical behavior data to determine the historical behavior intent feature information corresponding to each user's historical behavior data.

[0115] In some embodiments of this application, the electronic device performs intent recognition on the first feature information of each user's historical behavior data to determine the historical behavior intent feature information corresponding to each user's historical behavior data.

[0116] In some embodiments of this application, the electronic device can extract the semantics of each first feature information to obtain the semantic feature information of each first feature information, which is the historical behavior intention feature information.

[0117] Step 107d: The electronic device determines the first task intent based on each first time decay value, each first similarity, each historical behavioral intent feature information, and the first fusion feature information.

[0118] In some embodiments of this application, step 107d above can be implemented by the following steps 107d1 to 107d4:

[0119] Step 107d1: Based on each first similarity, the electronic device determines N first feature information whose similarity with the first fused feature information satisfies the similarity condition.

[0120] In some embodiments of this application, N is a positive integer.

[0121] In some embodiments of this application, the above similarity criteria may include any one of the following:

[0122] The similarity between the first feature information and the first fused feature information is greater than or equal to the similarity threshold;

[0123] After sorting at least one primary feature information in descending order of primary similarity, the top N primary feature information with the highest similarity are selected.

[0124] In some embodiments of this application, the electronic device can sort at least one first feature information in descending order of the first similarity to obtain a sequence of first feature information, and then determine the first N first feature information from the first feature information sequence. These N first feature information are the N first feature information that satisfy the above similarity conditions.

[0125] Step 107d2: The electronic device performs a weighted operation on the N first feature information based on the N first similarities corresponding to the N first feature information and the N time decay values ​​corresponding to the N first feature information, respectively, to obtain the second feature information.

[0126] In some embodiments of this application, the electronic device may calculate the aforementioned second feature information using the following formula:

[0127] V_ST = Σ(W_time_j × Sim_j ×V_memory_j); (3)

[0128] Where V_ST represents the second feature information mentioned above, j∈[1,N].

[0129] Step 107d3: The electronic device performs weighted fusion of the first fused feature information, the second feature information, and the third fused feature information to obtain the third fused feature information.

[0130] In some embodiments of this application, the second fusion feature information is the fusion feature information of at least one historical behavioral intent feature information.

[0131] In some embodiments of this application, the electronic device can perform feature fusion on the above-mentioned at least one historical behavioral intent feature information to obtain the above-mentioned second fused feature information.

[0132] In some embodiments of this application, since the second fusion feature information is a fusion feature information of at least one historical behavioral intention feature information, the second fusion feature information fuses at least one feature of historical behavioral intention, and thus the second fusion feature information can be used to characterize the user's long-term behavioral characteristics.

[0133] In some embodiments of this application, the electronic device can calculate the aforementioned third fusion feature information using the following formula:

[0134] V_input = V_content +β1×V_ST +β2×V_habit; (4)

[0135] Wherein, V_input represents the third fusion feature information mentioned above, V_habit represents the second fusion feature information mentioned above, and β1 and β2 represent the weight parameters of the second feature information and the second fusion feature information mentioned above, respectively.

[0136] Step 107d4: The electronic device determines the first task intent based on the third fused feature information.

[0137] In some embodiments of this application, step 107d4 described above can be implemented by the following steps A1 and A2:

[0138] Step A1: The electronic device inputs the third fusion feature information into the trained task intent prediction model and outputs at least one task intent.

[0139] In some embodiments of this application, the model structure of the above-mentioned task intent prediction model can be a large language model, and electronic devices can use open-source large language models as the above-mentioned task intent prediction model.

[0140] In some embodiments of this application, the electronic device can first perform intent recognition on the third fused feature information through the above-mentioned task intent prediction model to obtain at least one task intent feature information, and then decode each of the above-mentioned at least one task intent feature information to output the above-mentioned at least one task intent.

[0141] In some embodiments of this application, each of the at least one task intent feature information can be a feature information of one of the at least one task intents. In other words, each of the at least one task intent feature information can correspond to one of the at least one task intents.

[0142] Step A2: The electronic device determines the task intent with the highest confidence among at least one task intent as the first task intent.

[0143] In some embodiments of this application, the electronic device may sort the above-mentioned at least one task intent in descending order of confidence level, and then determine one or more task intents with the highest confidence level as the above-mentioned first task intent.

[0144] In this way, the electronic device inputs the third fusion feature information into the task intent prediction model and outputs at least one task intent; the task intent with the highest confidence among the at least one task intent is determined as the first task intent, which can accurately and quickly determine the user's task intent for the first information.

[0145] In some embodiments of this application, prior to step A2 described above, the task processing method provided in this application may further include steps A31 to A35:

[0146] Step A31: The electronic device calculates the second similarity between the second task intent feature information and the third fused feature information.

[0147] In some embodiments of this application, the aforementioned second task intent feature information may be task intent feature information corresponding to the second task intent from at least one historical task intent feature information, and the aforementioned second task intent may be any one of the aforementioned at least one task intent.

[0148] It should be noted that the specific implementation process of step A31 can be referred to the description of step 107b above. To avoid repetition, this embodiment will not repeat it here.

[0149] In some embodiments of this application, the electronic device can calculate the second similarity using the following formula:

[0150] S_semantic = cosine(V_input, V_intent_k); (5)

[0151] Where S_semantic represents the second similarity mentioned above, V_input represents the third fusion feature information mentioned above, and V_intent_k represents the second task intent feature information mentioned above.

[0152] Step A32: The electronic device obtains the matching degree between the second task intent and the third application.

[0153] In some embodiments of this application, the matching degree is used to indicate whether the third application can perform the task corresponding to the second task intent.

[0154] In some embodiments of this application, if the third application can execute the task corresponding to the second task intent, it indicates that the second task intent matches the third application, and the matching degree between the second task intent and the third application is 1.

[0155] For example, the third application mentioned above is a navigation application, and the second task intent mentioned above is a navigation intent. Since the navigation application can perform navigation tasks, the second task intent mentioned above matches the third application mentioned above, and the matching degree between the second task intent mentioned above and the third application mentioned above is 1.

[0156] In some embodiments of this application, if the third application cannot execute the task corresponding to the second task intent, it indicates that the second task intent and the third application do not match, and the matching degree between the second task intent and the third application is 0.

[0157] For example, the third application mentioned above is a navigation application, and the second task intent mentioned above is a product sharing intent. Since the navigation application can not execute the product sharing intent, the second task intent mentioned above does not match the third application mentioned above, and the matching degree between the second task intent mentioned above and the third application mentioned above is 0.

[0158] Step A33: The electronic device determines the first confidence score corresponding to the second task intent based on the second similarity and the second time decay value.

[0159] In some embodiments of this application, the first confidence score can be used to indicate the confidence level of the second task intent in the time dimension.

[0160] In some embodiments of this application, the second time decay value can be the time decay value of the first historical behavior data in the at least one historical behavior data, and the first historical behavior data can be the historical behavior data corresponding to the historical behavior intention feature information that matches the second task intention feature information in the at least one historical behavior data.

[0161] In some embodiments of this application, the historical behavioral intent feature information matched by the second task intent feature information can be understood as the target historical behavioral intent feature information among the at least one historical behavioral intent feature information whose similarity to the second task intent feature information is greater than or equal to a first similarity threshold.

[0162] In some embodiments of this application, the aforementioned target historical behavior intent feature information may be the historical behavior intent feature information corresponding to the aforementioned first historical behavior data.

[0163] In some embodiments of this application, the second task intent corresponding to the second task intent feature information may be the same as the historical behavior intent corresponding to the target historical behavior intent feature information.

[0164] In some embodiments of this application, the electronic device may calculate the aforementioned first confidence score using the following formula:

[0165] S_memory_bias = W_time 2× S_semantic; (6)

[0166] Where S_memory_bias represents the first confidence score mentioned above, and W_time 2 represents the second time decay value mentioned above.

[0167] Step A34: The electronic device obtains the second confidence score corresponding to the second task intent.

[0168] In some embodiments of this application, the aforementioned second confidence score is used to indicate the likelihood of performing the task corresponding to the aforementioned second task intent in the aforementioned third application.

[0169] In some embodiments of this application, step A34 above can be implemented by the following steps A341 to A345:

[0170] Step A341: The electronic device determines M historical behavior data that match the third application from at least one historical behavior data.

[0171] In some embodiments of this application, M is a positive integer.

[0172] In some embodiments of this application, among the at least one historical behavior data mentioned above, if the historical behavior data includes the application identifier of the application used by the user, and the application identifier indicates the application mentioned above as the third application, then these historical behavior data can be M historical behavior data that match the third application mentioned above.

[0173] In some embodiments of this application, the aforementioned third application may include, but is not limited to, any of the following: social networking applications, payment applications, navigation applications, shopping applications, video applications, photo album applications, music applications, etc. The specific application can be determined based on actual usage needs, and this application embodiment does not impose any limitations.

[0174] Step A342: The electronic device determines the intentions of M historical behaviors based on M historical behavioral data.

[0175] In some embodiments of this application, the electronic device can obtain M historical behavior intention feature information corresponding to the above M historical behavior data, and then decode each of the above M historical behavior intention feature information to obtain the above M historical behavior intentions.

[0176] Step A343: The electronic device determines the first number of historical behavioral intent categories corresponding to the M historical behavioral intents.

[0177] In some embodiments of this application, the electronic device can classify the above-mentioned M historical behavioral intentions and determine a first number of historical behavioral intention categories corresponding to the M historical behavioral intentions.

[0178] For example, if M is 6, the electronic device can categorize 6 historical behavioral intentions. If the 6 historical behavioral intentions can be divided into 2 categories, then the first number of the 6 historical behavioral intention categories is 2.

[0179] Step A344: The electronic device determines the second number of the second task intent from M historical behavioral intents.

[0180] In some embodiments of this application, the electronic device can determine the number of historical behavioral intentions that are the same as the second task intention from the M historical behavioral intentions, and determine the number of historical behavioral intentions that are the same as the second task intention from the M historical behavioral intentions as the second number.

[0181] Step A345: The electronic device determines the second confidence score based on the second quantity, the first quantity, and M.

[0182] In some embodiments of this application, the electronic device may calculate the aforementioned second confidence score using the following formula:

[0183] P_Intent_k = (M + α1) / (N_total + α1 × K); (7)

[0184] Where P_Intent_k represents the second confidence level mentioned above, N_total represents the first quantity mentioned above, K represents the second quantity mentioned above, and α1 is the smoothing parameter.

[0185] Thus, the electronic device can accurately determine the second confidence score of the second task intention by identifying M historical behavior data that match the third application from at least one historical behavior data; determining M historical behavior intentions based on the M historical behavior data; determining a first number of historical behavior intention categories corresponding to the M historical behavior intentions; determining a second number of second task intentions from the M historical behavior intentions; and determining a second confidence score based on the second number, the first number, and M.

[0186] Step A35: The electronic device determines the third confidence score of the second task intent based on the second similarity, matching degree, first confidence score, and second confidence score.

[0187] In some embodiments of this application, step A35 above can be implemented by the following steps A351 to A353:

[0188] Step A351: The electronic device performs a weighted calculation on the second similarity, matching degree, first confidence score and second confidence score to obtain the confidence weighted value corresponding to the second task intent.

[0189] In some embodiments of this application, the electronic device may calculate the confidence weighted value using the following formula:

[0190] Score_k =α2×S_semantic +β3×S_target_match +γ2×S_memory_bias+δ×P_Intent_k; (8)

[0191] Where Score_k represents the aforementioned confidence weighted value, S_target_match represents the aforementioned matching degree, and α2, β3, γ2, and δ represent the weight parameters of the second similarity, matching degree, first confidence score, and second confidence score, respectively.

[0192] Step A352: The electronic device calculates the exponential mapping value of the confidence weighted value corresponding to the second task intent.

[0193] In some embodiments of this application, the electronic device may use the following formula to determine the above-mentioned exponential mapping value:

[0194] Value_k = exp(Score_k); (9)

[0195] Step A353: The electronic device determines the third confidence score of the second task intent based on the index mapping value corresponding to the second task intent and the sum of all index mapping values ​​corresponding to at least one task intent.

[0196] In some embodiments of this application, the electronic device may calculate the third confidence score of the second task intent using the following formula:

[0197] Confidence_k =Value_k / Σexp(Score_m); (10)

[0198] Where m∈[1,K], K represents the number of at least one task intent, Σexp(Score_m) represents the sum of all exponential mapping values ​​corresponding to at least one task intent, Score_m represents the confidence weighted value corresponding to the m-th task intent among the at least one task intent, and exp(Score_m) represents the exponential mapping value of the confidence weighted value corresponding to the m-th task intent.

[0199] Thus, the electronic device obtains the confidence weighted value corresponding to the second task intent by weighting the second similarity, matching degree, first confidence score and second confidence score; calculates the exponential mapping value of the confidence weighted value corresponding to the second task intent; and determines the third confidence score of the second task intent based on the confidence weighted value corresponding to the second task intent and the sum of all exponential mapping values ​​corresponding to at least one task intent, thereby accurately determining the confidence of the second task intent.

[0200] Thus, by calculating the second similarity between the second task intent feature information and the third fused feature information; the matching degree between the second task intent and the third application; determining the first confidence score corresponding to the second task intent based on the second similarity and the second time decay value; obtaining the second confidence score corresponding to the second task intent; and determining the third confidence score of the second task intent based on the second similarity, matching degree, first confidence score, and second confidence score, the electronic device can accurately determine the confidence of the second task intent.

[0201] Thus, the electronic device determines N first feature information that satisfy the similarity condition with the first fused feature information based on each first similarity; it then performs a weighted operation on the N first feature information based on the first similarity and the first time decay value corresponding to each of the N first feature information to obtain the second feature information; it then performs a weighted fusion operation on the first fused feature information, the second feature information, and the second fused feature information to obtain the third fused feature information, where the second fused feature information is a fused feature information of at least one historical behavioral intent feature information; and finally, it determines the first task intent based on the third fused feature information.

[0202] Thus, the electronic device acquires the first time decay value of each historical behavior data; calculates the first similarity between each first feature information and the fused feature information; determines at least one historical behavior intention feature information based on each of the user's historical behavior data, with each user's historical behavior data corresponding to historical behavior intention feature information; and determines the first task intention based on each first time decay value, each first similarity, each historical behavior intention feature information, and the first fused feature information, thereby accurately and quickly determining the user's task intention for the first information.

[0203] In this way, the electronic device can accurately and quickly determine the user's task intent for the first information by acquiring the first feature information of each user's historical behavior data; acquiring the first fused feature information corresponding to the first information and the interface context information; and determining the first task intent based on each user's historical behavior data, each first feature information and the first fused feature information.

[0204] The task execution method provided in this application embodiment will be further described in detail below with reference to specific implementation methods, taking a mobile phone as an example.

[0205] This implementation aims to provide an intuitive, accurate, and efficient cross-application intelligent interaction solution by introducing the novel interaction form of "intent wheel" and combining it with "memory context injection" technology.

[0206] Summary of improvements in this implementation method:

[0207] The first "Intent Wheel" interactive design:

[0208] The core interactive innovation of this implementation lies in the introduction of a "hover-triggered" mechanism. Unlike existing technologies, users are not immediately interrupted when dragging content. Only when the user drags the content above the target application and pauses briefly or for a specific duration (hover), the phone combines the three-dimensional information of "source content + memory + target application attributes" to activate the "intent wheel" around the finger. This not only avoids accidental touches but also ensures that "the right option is provided in the right place."

[0209] Memory-based context injection mechanism:

[0210] While analyzing dragged content, "short-term memory" (such as recent browsing history and clipboard stack) and "long-term memory" (such as user habits and hierarchical relationships) from the local vector database are injected to achieve a qualitative change from "guessing content" to "guessing intent".

[0211] Dynamic intent generation and sorting:

[0212] The options on the intention wheel are not fixed, but are dynamically generated, sorted, and the sector size is adjusted based on the confidence level of "content + memory".

[0213] Figure 5 This is a schematic diagram of the task execution method provided in this embodiment. For example... Figure 5 As shown, the task execution method provided in this embodiment may include the following steps 501 to 506:

[0214] Step 501: Drag and drop the phone to move it around the entire area.

[0215] Action: The user long-presses the content in the source application (such as WeChat) and begins to drag.

[0216] For example, the content that the user long-presses is the first information mentioned above, and the source application mentioned above is the first application mentioned above.

[0217] For example, a user can long-press the "vivo phone desktop display" card to extract the card's content to their fingertip.

[0218] Feedback: Instead of displaying the intent wheel, a lightweight "semantic capsule" is shown to keep the user's view clear and allow them to focus on finding the target application.

[0219] For example, a semantic capsule is a thumbnail of the displayed content.

[0220] For example, the target application is the third application mentioned above.

[0221] Data preparation: The feature vectors of the source content and the context of the source application are silently extracted in the background of the mobile phone and are in standby mode.

[0222] Step 502: The mobile phone performs target application recognition and hover detection.

[0223] Entry detection: When a user drags a capsule with their finger into the interactive area (window, icon, or floating window) of the target application, the system identifies the target application ID (such as com.android.calendar).

[0224] Hover detection: The system detects the displacement speed and dwell time of the finger within the target area.

[0225] If the time is greater than the threshold T (e.g., 300ms) and the displacement is less than the threshold D (i.e., the finger is steady), it is determined as "hovering intention".

[0226] It should be noted that if the user releases their finger directly, the default normal Drop operation will be performed, which is compatible with traditional habits.

[0227] Step 503: The mobile phone performs memory context injection.

[0228] For example, the memory context is the interface context information mentioned above.

[0229] This step is the core algorithm processing stage of this implementation method. While analyzing the dragged content, the mobile phone combines historical user behavior data and the current environmental state to perform contextual enhancement, thereby improving the accuracy of intent prediction.

[0230] I. Construction of Memory Data Structure

[0231] The mobile phone constructs a hierarchical memory index model, including short-term memory index and long-term behavior modeling modules.

[0232] (a) Short-term memory index construction

[0233] The mobile phone uses a time window mechanism to maintain the short-term memory data set.

[0234] The time window T is set to the most recent 30 minutes (which can be dynamically adjusted according to system resources).

[0235] Within this time window, the system records the following data types:

[0236] 1. Application switching history;

[0237] 2. Search keywords;

[0238] 3. Clipboard history;

[0239] 4. Current screen OCR-recognized text;

[0240] 5. The category of the most recently performed operation.

[0241] Each short-term memory data structure includes:

[0242] 1. Timestamp;

[0243] 2. Application identifier app_id;

[0244] 3. Data type: data_type;

[0245] 4. Semantic vector embedding_vector;

[0246] 5. Action tag.

[0247] For example, the operation label is used to indicate the type of operation behavior, and the semantic vector mentioned above is the first feature information mentioned above.

[0248] The semantic vector embedding_vector is generated using a local lightweight semantic encoding model.

[0249] To reduce historical interference, the mobile phone performs time decay weighting on the short-term memory data, as detailed in the formula (1) above.

[0250] It should be noted that short-term memory data refers to the aforementioned user historical behavior data.

[0251] (ii) Long-term behavioral modeling

[0252] The long-term behavior model is built on a statistical behavior modeling mechanism.

[0253] The historical frequency of mobile phone users selecting a certain type of intent for a specific data type in the target application environment is used to form a conditional probability model. For details, please refer to the above formula (7), which will not be elaborated here.

[0254] This long-term model does not rely on fixed rules and can be adapted to different applications and scenarios.

[0255] II. Memory Retrieval and Fusion Mechanism

[0256] The currently dragged content and its memory context are first encoded as a semantic vector: V_content

[0257] Short-term semantic similarity retrieval

[0258] The phone calculates the cosine similarity in the short-term memory vector library, which can be referred to in the above formula (2), and will not be repeated here. The top N memory records with the highest similarity are selected to form a short-term context set, and time and similarity weighted fusion is performed on them, which can be referred to in the above formula (3), and will not be repeated here.

[0259] III. Construction of Integrated Inputs

[0260] The mobile phone determines the final input vector used for the intent prediction model, which can be referred to in formula (4) above, and will not be repeated here. It should be noted that the above input vector is the third fusion feature information mentioned above.

[0261] For example, this fusion mechanism enables comprehensive modeling of "content semantics + short-term environment + long-term habits".

[0262] Step 504: The mobile phone performs intent prediction and generates a dynamic wheel.

[0263] For example, based on the fused input generated in step 503, the system performs candidate intent generation and confidence calculation.

[0264] I. Generation of Candidate Intent Set

[0265] The sources of the candidate intent set include:

[0266] 1. The set of operations supported by the target application;

[0267] 2. A set of system-level global operations;

[0268] 3. Collection of users' historical high-frequency operations.

[0269] Let the candidate set be: Intent_Set = {I1, I2, …, In}

[0270] II. Multifactor scoring model

[0271] For each candidate intent Ii, the system calculates a comprehensive score, which can be referred to in the above formula (8), and will not be repeated here.

[0272] (a) Semantic matching score calculation

[0273] The calculation process of semantic matching score can be referred to the above formula (5), and will not be repeated here.

[0274] For example, the semantic matching score is the second similarity mentioned above.

[0275] (ii) Target application matching score

[0276] If the intent falls within the scope of native support of the target application: S_target_match = 1

[0277] Otherwise: S_target_match = 0

[0278] Alternatively, it can be set to a continuous value in the interval 0–1.

[0279] For example, the target application matching score is the matching degree mentioned above.

[0280] (iii) Short-term memory enhancement score

[0281] For example, the short-term memory enhancement score is used to reinforce recent-related operational tendencies, and the calculation process of the short-term memory enhancement score can refer to the above formula (6).

[0282] For example, the short-term memory enhancement score is the first confidence score mentioned above.

[0283] (iv) Prior scores of long-term behavior

[0284] For example, long-term behavioral prior scores are used to enhance personalized matching capabilities, and the calculation process of long-term behavioral prior scores can refer to the above formula (7).

[0285] For example, the long-term behavioral prior score is the second confidence score mentioned above.

[0286] III. Confidence Normalization Calculation

[0287] To avoid multiple candidate intentions from scoring high at the same time, the system uses Softmax normalization, which can be referred to in formula (9) and formula (10) above. It will not be repeated here.

[0288] IV. Calculation of sector angles on the roulette wheel

[0289] The sector angle of the intent wheel is directly related to the confidence level:

[0290] Angle_i = 360° × Confidence_i; (11)

[0291] For example, to avoid the interactive experience being affected by excessively small sectors, a minimum angle threshold is set: Angle_i ≥ θ_min.

[0292] For example, a candidate intent may not be displayed if its confidence level is below a threshold.

[0293] Step 505: The mobile phone performs intent wheel rendering.

[0294] Visual presentation: Instantly render a semi-transparent circular UI (intent wheel) around the finger touch point.

[0295] Adsorption and feedback:

[0296] When a finger slides into a sector, that sector is highlighted and slightly magnified.

[0297] Haptic feedback: Triggers a "click" vibration from the linear motor, simulating the feel of a physical turntable.

[0298] Release trigger: The user releases their finger on the target sector, which is considered confirmation of the intention. If the user drags the finger back to the center of the wheel or the cancel area at the edge of the screen, the operation is canceled.

[0299] Step 506: The mobile phone executes structured instructions.

[0300] Once the user confirms their intention on the wheel, the system will generate a structured action payload and distribute it to the target application or system service.

[0301] Example of a data packet structure (JSON):

[0302] JSON

[0303] {

[0304] "trace_id": "drag-20251209-xa98s",

[0305] "source_context": {

[0306] "app_pkg": "com.chat.app",

[0307] "data_type": "text / plain",

[0308] "raw_content": "Patent review will be held next Friday at 3 PM in Conference Room 202".

[0309] "injected_memory": {

[0310] "related_entity": "Patent Review Project Team", / / Association from long-term memory

[0311] "recent_activity": "Checked the booking status of meeting room 202" / / From short-term memory

[0312] }

[0313] },

[0314] "selected_intent": {

[0315] "intent_code": "CREATE_SCHEDULE",

[0316] "confidence": 0.92,

[0317] "target_app": "com.calendar.app"

[0318] },

[0319] "execution_params": {

[0320] "title": "Patent Review",

[0321] "start_time": "2025-12-19T15:00:00",

[0322] "location": "Meeting Room 202",

[0323] "attendees": ["Patent Review Project Team"] / / Automatically fills in attendees based on memory.

[0324] }

[0325] }

[0326] 4. Examples of specific implementation methods

[0327] Scenario 1: Intelligent schedule creation based on social chat content

[0328] User behavior: In a WeChat chat window, a message was received from a colleague: "General Manager Wang said that we need to see the quarterly report at 10 a.m. next Monday. Remember to prepare." The user long-pressed and dragged the message.

[0329] System processing (memory injection):

[0330] Extracted content: "Next Monday at 10:00 AM", "Quarterly report".

[0331] Injecting long-term memory: The system retrieves the user's habit of storing "quarterly reports" in the "Work / Docs" folder.

[0332] Inject short-term memory: The system detected that the user opened the PPT software 5 minutes ago.

[0333] Intention Wheel Pops Up: A wheel containing three sectors appears around your finger.

[0334] Sector A (concept view, 150 degrees): "Create Schedule" (automatically recognizes time).

[0335] Sector B (sub-intention, 105 degrees): "Send to..." (based on memory recommendation "General Manager Wang" or "Finance Department Group").

[0336] Sector C (sub-intention, 105 degrees): "Find File" (searches local documents based on the keyword "quarterly report").

[0337] Interaction: Users do not need to search for the calendar app icon; they can simply swipe "up" (hitting sector A) and release.

[0338] Result: The system opens the calendar in a semi-floating window with the title "Preparing Quarterly Report" pre-filled and the time pre-filled as 10:00 AM next Monday. The notes automatically link to the local "Quarterly Report.ppt" file (based on memory association).

[0339] Scenario 2: Real-time processing mode within the source application

[0340] User behavior: When browsing pictures in the photo album application, a user long-presses a picture and drags it, but does not move it to other applications. The picture only moves slightly within the original area and meets the hovering judgment condition.

[0341] System processing (memory injection):

[0342] Extracted content: Image semantic features (such as category tags for people, documents, products, etc.).

[0343] Injecting long-term memory: The system retrieved data showing that the user had performed the "extract text" operation multiple times on "document-type images" in the past.

[0344] Inject short-term memory: The system detected that the user recently edited an ID photo.

[0345] Intent Cart Pop-up: Since the target application ID was not recognized, the system calls the source application's intent set and the system's general intent set to generate a cart. The cart contains the following sectors:

[0346] Sector A (Plan): "Extract Text";

[0347] Sector B: "Remove Background";

[0348] Sector C: "Hidden";

[0349] Sector D: "Add to Favorites".

[0350] The angle of each sector is dynamically allocated based on the confidence level.

[0351] Interaction: The user slides to the right to the "Extract Text" sector and releases their finger.

[0352] Result: The system displays the text recognition results panel on the current interface and automatically enters the copyable state, without the need for cross-application operations.

[0353] Scenario 3: Cross-device air delivery mode

[0354] User behavior: The user long-presses a video file in the phone's photo album and drags it to the right edge of the screen and hovers over it.

[0355] System processing (device detection and memory injection):

[0356] The system detected the following online devices under the same account: tablet, PC, and TV.

[0357] Extracted content: The file type is video.

[0358] Injecting long-term memory: The user has cast the video to TV multiple times in the past.

[0359] Implanting short-term memory: The most recent delivery target was the living room TV.

[0360] Intent wheel pop-up: Device mapping mode is triggered at the edge of the screen, and the wheel sector corresponds to the device orientation.

[0361] Sector A (concept diagram): "Cast to TV";

[0362] Sector B: "Drop to Flat Panel";

[0363] Sector C: "Drop to PC".

[0364] The sector angle is dynamically allocated based on the device matching probability and historical behavior statistics.

[0365] Interaction: The user swipes towards the TV and releases their finger.

[0366] Result: The system establishes a transmission channel between devices and automatically plays video content on the TV.

[0367] Scenario 4: Multimodal Fusion Interaction Mode

[0368] User behavior: A user drags a meeting notification text from the chat application to the calendar application icon and hovers over it.

[0369] System processing (memory injection):

[0370] Content extraction: Identify text containing conference keywords but without a specific time.

[0371] Injecting long-term memory: Users typically set project review meetings to 1 hour.

[0372] Injecting short-term memory: The user has been frequently creating meeting schedules recently.

[0373] Intention wheel pops up:

[0374] Sector A (Concept Diagram): “Creating a Schedule”;

[0375] Sector B: "Send Alert";

[0376] Sector C: "Find related documents".

[0377] Additional voice parameters:

[0378] While hovering, the user says: "Next Friday at 2 PM."

[0379] The system uses real-time voice recognition to update parameters, improving the confidence of the "create schedule" intent and expanding its sector angle.

[0380] - Interaction: User releases their hand to confirm.

[0381] Result: The system automatically generates a schedule record that includes the time, title, and default meeting duration.

[0382] Scenario 5: Commercialization and Service Recommendation Model

[0383] User behavior: Users long-press and drag the text "Universal Studios" to the system-level service area and hover it in the travelogue article.

[0384] System processing (semantic recognition and memory injection):

[0385] Extracted content: Identified as a tourist attraction entity.

[0386] Injecting long-term memory: The user has a history of ticket purchase behavior.

[0387] Injecting short-term memory: The user recently searched for "ticket prices".

[0388] Intention wheel pops up:

[0389] Sector A (Idea): "View Ticket";

[0390] Sector B: "Compare prices across the entire network";

[0391] Sector C: “Map Navigation”;

[0392] Sector D: "View Guide".

[0393] The confidence score for each intent is calculated based on the semantic matching score and historical behavior statistics.

[0394] Interaction: The user slides to the "View Tickets" sector and releases their finger.

[0395] Result: The system redirected to the service aggregation page, displaying relevant ticket purchase links and price information.

[0396] The task processing method provided in this embodiment has application potential not only limited to the main scenario of "cross-application hovering", but can also be extended to the following broader interaction forms:

[0397] I. Trigger Mechanism Expansion: From drag-and-drop triggering to multiple input triggering methods

[0398] The intent generation mechanism in this implementation does not rely on the drag action itself, but on a three-layer structure of "hover + context modeling + intent prediction".

[0399] Therefore, the triggering condition can be expanded to:

[0400] 1. Long press to trigger while stationary;

[0401] 2. Cursor hover trigger (applicable to desktop systems or tablet mode);

[0402] 3. Triggered by gesture pause;

[0403] 4. Stylus hover trigger;

[0404] 5. Delayed trigger after selecting text.

[0405] Technical implementation: When the system detects that the input behavior meets the conditions of "displacement less than threshold + dwell time greater than threshold", it can enter the intent prediction process.

[0406] This extension explains that the present invention is not limited to the dragging action itself, but rather constructs a general hovering intent generation mechanism.

[0407] II. Input Object Expansion: From Single Data Objects to Composite Objects

[0408] Existing implementations predict intent based on a single data object.

[0409] In the extended implementation:

[0410] 1. Supports dragging and dropping multiple data objects together;

[0411] 2. Supports mixed input of text, images, and files;

[0412] 3. Supports input of structured objects (contact cards, meeting cards).

[0413] The system can fuse multiple object vectors: V_content = Σ V_object_i

[0414] The intent confidence is recalculated based on the combined semantics.

[0415] This extension enhances the ability to handle complex tasks.

[0416] III. Intent Generation Layer Extension: From Application-Level Intents to System-Level Task Orchestration

[0417] In existing implementations, intent is primarily mapped to a single action.

[0418] In the extended approach: the system can generate a composite intent sequence: Intent_Chain = {I1 → I2 → I3}

[0419] For example:

[0420] Drag and drop the meeting text → Create schedule → Automatically generate meeting minutes template → Notify relevant members.

[0421] This mechanism automatically derives high-frequency task chains based on historical behavior statistical models.

[0422] In terms of technical implementation, a composite intent mode can be triggered when the confidence level exceeds a set threshold.

[0423] IV. Memory Mechanism Extension: From Single-User Model to Group Collaboration Model

[0424] In enterprise scenarios: the system can build group-level behavioral statistics models.

[0425] P(Intent_i | Context, Group_Profile)

[0426] For example, when dragging and dropping meeting texts within a project team, the system prioritizes recommending "synchronize to the project management system".

[0427] This extension is based on group behavior aggregation statistics and does not change the core structure of this invention.

[0428] V. Predictive Model Extension: From Responsive Prediction to Predictive Preloading

[0429] In an extended implementation: the system can pre-calculate candidate intents at the beginning of the drag-and-drop process.

[0430] In other words, when dragging is detected to begin, the background pre-calculates the candidate intent list and confidence ranking.

[0431] If the hovering condition is met subsequently, the wheel can be rendered directly, reducing latency.

[0432] This mechanism improves the smoothness of real-time interaction.

[0433] VI. Interface Form Expansion: From a wheel-like form to a replaceable presentation layer

[0434] The core innovation of this implementation lies in the intent prediction and dynamic confidence calculation mechanism, rather than limiting the visual form of the roulette wheel.

[0435] Therefore, in the extension method:

[0436] 1. Linear menus can be used;

[0437] 2. Card-type floating shelves can be used;

[0438] 3. A circular progress bar structure can be adopted;

[0439] 4. A three-dimensional spatial distribution structure can be adopted (compatible with AR devices).

[0440] The intent ranking and confidence assignment mechanisms remain unchanged.

[0441] VII. Device Form Factor Expansion: Adaptable to foldable screens, tablets, and space computing devices

[0442] Because the intention roulette is based on an equidistant distribution structure at the center point.

[0443] In large-screen devices or foldable screen unfolded mode:

[0444] 1. The sector angle can be dynamically adjusted according to the screen size;

[0445] 2. Partial rendering is possible in multi-window mode;

[0446] 3. It can be mapped to a spatial coordinate system to achieve three-dimensional interaction.

[0447] This implementation method, by introducing a "target application hovering triggering mechanism" and "memory context injection technology," produces the following significant advantages compared to existing technologies:

[0448] 1. Solved the problem of accurately matching "operational intent" with "application capability".

[0449] Existing intent-based recommendations often rely solely on source content (e.g., dragging and dropping text recommends all text-processing apps), resulting in lengthy and inaccurate recommendation lists. This invention utilizes a **"target app" as a strong filter**, generating intents strongly related to that app (e.g., creating a new schedule, searching for dates) only when the user drags content to and hovers over it. This significantly narrows the scope of intents, raising the accuracy of intent prediction from the industry average to a "usable" or even "easy-to-use" level.

[0450] 2. Seamless compatibility between "default operations" and "advanced operations" has been achieved.

[0451] This implementation creatively utilizes "hover duration" as a switch to toggle the interaction mode.

[0452] Fast Drop: Users simply drag and drop and release, and the system executes the default "paste / send" logic, preserving the user's original high-frequency muscle memory with zero learning cost.

[0453] Hover & Think: When a user hovers briefly, the system senses that the user has a more complex need within the application and then brings up the intent wheel. This design avoids the visual disturbance caused by forcibly popping up the wheel in all drag-and-drop scenarios, achieving "on-demand intelligence."

[0454] 3. Memory-based "cognitive unloading" and "expectation alignment"

[0455] With the introduction of "memory context injection", the system is no longer a cold, impersonal data transporter.

[0456] Short-term memory solves the problem of data silos (e.g., knowing that you just looked at price comparisons in your browser, when you drag and drop a product into WeChat, the wheel will automatically prompt "send price comparison link" instead of just a product image).

[0457] Long-term memory solves the problem of repetitive configuration (e.g., knowing that you always send files to "General Manager Wang" in PDF format, the "Send" option on the wheel will automatically include the "Convert to PDF" preprocessing action). This significantly reduces the user's cognitive load, making the interaction results highly consistent with the user's psychological expectations.

[0458] 4. Improved screen space utilization and operational efficiency.

[0459] It utilizes an "intention wheel" instead of a traditional list menu, making full use of Fitts's Law.

[0460] The user's finger is located at the center of the circle, and the distance to any sector is equal and extremely short.

[0461] The sector area is dynamically adjusted according to the confidence level, making it easier to hit high-frequency intentions blindly.

[0462] This design eliminates the need for a Dock or Drop Zone at the edge of the screen, making it particularly suitable for flexible operation of large-screen devices such as foldable screens and tablets in multitasking scenarios.

[0463] It should be noted that each of the above method embodiments, or various possible implementations of each method embodiment, can be executed individually or in combination of any two or more. The specific implementation can be determined according to actual usage requirements, and this application embodiment does not impose any restrictions on this.

[0464] or,

[0465] It should be noted that the above-described method embodiments, or the various possible implementations of the method embodiments, can be executed individually, or, provided there are no contradictions, they can be combined with each other. The specific implementation can be determined according to actual usage requirements, and this application embodiment does not impose any restrictions on this.

[0466] The task execution method provided in this application can be executed by a task execution device. This application uses the example of a task execution device executing the task execution method to illustrate the task execution device provided in this application.

[0467] Figure 6 This is a schematic diagram of a task execution device provided in an embodiment of this application. Figure 6 As shown, the task execution device includes a receiving module 601 and a display module 602.

[0468] The receiving module 601 is used to receive the user's first input on the first information in the first interface;

[0469] Display module 602 is used to respond to the first input and display the task execution result corresponding to the first task intent. The first task intent is determined based on the first information and the second information. The second information includes at least one of the following: interface context information of the first interface and user historical behavior data.

[0470] In some embodiments of this application, the display module 602 is further configured to:

[0471] Before displaying the task execution result corresponding to the first task intent, at least one task control corresponding to the first task intent is displayed, and each task control in the at least one task control indicates a task;

[0472] The receiving module is further configured to receive a second input from the user to the first task control in the at least one task control;

[0473] The display module 602 is specifically used for:

[0474] In response to the second input, the task execution result corresponding to the first task control is displayed;

[0475] Wherein, the first task control is any one of the at least one task controls.

[0476] In some embodiments of this application, the at least one task control forms a control wheel; each task control is a sector control in the control wheel, and the angle of the sector control is determined based on the confidence level of the task intent corresponding to the sector control.

[0477] In some embodiments of this application, the first interface is the application interface of the first application, and the interface context information includes at least one of the following: the interface content of the first interface; the application attribute information of the first application; the geographical location information of the electronic device; and the second reference information associated with the first information.

[0478] The application attribute information includes at least one of the following: application name; application type; application function; the second reference information is the interface content of the second interface, the second interface is the second application interface of the second application accessed by the user in history, and the second application interface includes the first information.

[0479] In some embodiments of this application, combined with Figure 6 ,like Figure 7 As shown, the user's historical behavior data includes at least one item;

[0480] The device 600 further includes: a processing module 603, used for:

[0481] Obtain the first characteristic information of each user's historical behavior data;

[0482] Obtain the first fusion feature information corresponding to the first information and the interface context information;

[0483] The first task intent is determined based on the user's historical behavior data, the first feature information, and the first fused feature information.

[0484] In some embodiments of this application, the processing module 603 is specifically used for:

[0485] Obtain the first time decay value for each historical behavior data point, which is used to indicate the importance of the user's historical behavior data in the time dimension;

[0486] Calculate the first similarity between each first feature information and the first fused feature information;

[0487] Based on the historical behavior data of each user, at least one historical behavior intent feature is determined, and each user's historical behavior data corresponds to historical behavior intent feature information.

[0488] The first task intent is determined based on each of the first time decay values, each of the first similarities, each of the historical behavioral intent feature information, and the first fusion feature information.

[0489] In some embodiments of this application, the processing module 603 is specifically used for:

[0490] Based on each of the first similarities, N first feature information that satisfy the similarity condition with the first fused feature information are determined among each of the first feature information.

[0491] Based on the first similarity corresponding to each of the N first feature information and the first time decay value corresponding to each of the N first feature information, a weighted operation is performed on the N first feature information to obtain the second feature information;

[0492] The first fused feature information, the second feature information, and the third fused feature information are weighted and fused to obtain the third fused feature information, wherein the second fused feature information is the fused feature information of the at least one historical behavioral intention feature information.

[0493] Based on the third fusion feature information, the first task intent is determined; where N is a positive integer.

[0494] In some embodiments of this application, the processing module 603 is specifically used for:

[0495] The third fused feature information is input into the trained task intent prediction model, which outputs at least one task intent.

[0496] The task intent with the highest confidence level among the at least one task intent is determined as the first task intent.

[0497] In some embodiments of this application, the first input is an input dragged from a first position coordinate to a second position coordinate, and the processing module 603 is further configured to:

[0498] Before determining the task intent with the highest confidence among the at least one task intent as the first task intent, calculate the second similarity between the second task intent feature information and the third fused feature information;

[0499] Obtain the matching degree between the second task intent and the third application, wherein the matching degree is used to indicate whether the third application can execute the task corresponding to the second task intent;

[0500] Based on the second similarity and the second time decay value, a first confidence score corresponding to the second task intention is determined. The first confidence score is used to indicate the confidence of the second task intention in the time dimension.

[0501] Obtain a second confidence score corresponding to the second task intent, the second confidence score being used to indicate the likelihood of executing the task corresponding to the second task intent in the third application;

[0502] Based on the second similarity, the matching degree, the first confidence score, and the second confidence score, a third confidence score for the second task intent is determined;

[0503] Wherein, the second task intent is any one of the at least one task intents, the second task intent feature information is the task intent feature information corresponding to the second task intent, the second time decay value is the time decay value of the first historical behavior data in the at least one historical behavior data, the first historical behavior data is the historical behavior data corresponding to the historical behavior intent feature information that matches the second task intent feature information in the at least one historical behavior data, the first position coordinate is located within the display area of ​​the first information, the second position coordinate is located within the third interface, and the third interface is the application interface of the third application.

[0504] In some embodiments of this application, the processing module 603 is specifically used for:

[0505] The second similarity, the matching degree, the first confidence score, and the second confidence score are weighted to obtain the confidence weighted value corresponding to the second task intent;

[0506] Calculate the exponential mapping value of the confidence weighted value corresponding to the second task intent;

[0507] Based on the index mapping value corresponding to the second task intent, and the sum of all index mapping values ​​corresponding to the at least one task intent, a third confidence score for the second task intent is determined.

[0508] In some embodiments of this application, the processing module 603 is specifically used for:

[0509] From the at least one set of historical behavior data, determine M sets of historical behavior data that match the third application;

[0510] Based on the M historical behavioral data, determine the M historical behavioral intentions;

[0511] Determine the first number of historical behavioral intention categories corresponding to M historical behavioral intentions;

[0512] Determine a second number of the second task intentions from the M historical behavioral intentions;

[0513] The second confidence score is determined based on the second quantity, the first quantity, and M;

[0514] Where M is a positive integer.

[0515] In the task execution device provided in this application embodiment, the task execution device receives a first input from a user on first information in a first interface; responds to the first input and displays the task execution result corresponding to a first task intent. The first task intent is determined based on first information and second information, the second information including at least one of the following: interface context information of the first interface and user historical behavior data. Thus, when the task execution device receives input from a user on the first information, it can quickly and accurately identify the user's task intent based on at least one of the first information, the interface context information of the first interface, and the user's historical behavior data, and then display the task execution result corresponding to the user's task intent, thereby improving the task execution efficiency of the task execution device.

[0516] The task execution device in this application embodiment can be an electronic device or a component within an electronic device, such as an integrated circuit or a chip. The electronic device can be a terminal or other devices besides a terminal. For example, the electronic device can be a mobile phone, tablet computer, laptop computer, PDA, in-vehicle electronic device, mobile internet device (MID), augmented reality (AR) / virtual reality (VR) device, robot, wearable device, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc. It can also be a server, network attached storage (NAS), personal computer (PC), television (TV), ATM, or self-service machine, etc. This application embodiment does not specifically limit the device.

[0517] The task execution device in this application embodiment can be a device with an operating system. This operating system can be Android, iOS, or other possible operating systems; this application embodiment does not specifically limit the specific operating system.

[0518] The task execution device provided in this application embodiment can achieve... Figures 1 to 3 The various processes implemented in the method implementation examples will not be described again here to avoid repetition.

[0519] Optionally, such as Figure 8As shown, this application embodiment also provides an electronic device 800, including a processor 801 and a memory 802. The memory 802 stores a program or instructions that can run on the processor 801. When the program or instructions are executed by the processor 801, they implement the various steps of the above-described task execution method embodiment and can achieve the same technical effect. To avoid repetition, they will not be described again here.

[0520] It should be noted that the electronic devices in the embodiments of this application include the mobile electronic devices and non-mobile electronic devices described above.

[0521] Figure 9 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of this application.

[0522] The electronic device 900 includes, but is not limited to, components such as: radio frequency unit 901, network module 902, audio output unit 903, input unit 904, sensor 905, display unit 906, user input unit 907, interface unit 908, memory 909, and processor 910.

[0523] Those skilled in the art will understand that the electronic device 900 may also include a power supply (such as a battery) for supplying power to various components. The power supply may be logically connected to the processor 910 through a power management system, thereby enabling functions such as managing charging, discharging, and power consumption through the power management system. Figure 9 The electronic device structure shown does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown, or combine certain components, or have different component arrangements, which will not be elaborated here.

[0524] The user input unit 907 is used to receive the user's first input on the first information in the first interface;

[0525] Display unit 906 is used to respond to the first input and display the task execution result corresponding to the first task intent. The first task intent is determined based on the first information and the second information. The second information includes at least one of the following: the interface context information of the first interface and the user's historical behavior data within the first time period.

[0526] In some embodiments of this application, the display unit 906 is further configured to:

[0527] In some embodiments of this application, the display unit 906 is further configured to:

[0528] Before displaying the task execution result corresponding to the first task intent, at least one task control corresponding to the first task intent is displayed, and each task control in the at least one task control indicates a task;

[0529] The receiving module is further configured to receive a second input from the user to the first task control in the at least one task control;

[0530] The display unit 906 is specifically used for:

[0531] In response to the second input, the task execution result corresponding to the first task control is displayed;

[0532] Wherein, the first task control is any one of the at least one task controls.

[0533] In some embodiments of this application, the at least one task control forms a control wheel; each task control is a sector control in the control wheel, and the angle of the sector control is determined based on the confidence level of the task intent corresponding to the sector control.

[0534] In some embodiments of this application, the first interface is the application interface of the first application, and the interface context information includes at least one of the following: the interface content of the first interface; the application attribute information of the first application; the geographical location information of the electronic device; and the second reference information associated with the first information.

[0535] The application attribute information includes at least one of the following: application name; application type; application function; the second reference information is the interface content of the second interface, the second interface is the second application interface of the second application accessed by the user in history, and the second application interface includes the first information.

[0536] In some embodiments of this application, the processor 910 is used for:

[0537] Obtain the first characteristic information of each user's historical behavior data;

[0538] Obtain the first fusion feature information corresponding to the first information and the interface context information;

[0539] The first task intent is determined based on the user's historical behavior data, the first feature information, and the first fused feature information.

[0540] In some embodiments of this application, the user historical behavior data is at least one;

[0541] The processor 910 is specifically used for:

[0542] Obtain the first time decay value for each historical behavior data point, which is used to indicate the importance of the user's historical behavior data in the time dimension;

[0543] Calculate the first similarity between each first feature information and the first fused feature information;

[0544] Based on the historical behavior data of each user, at least one historical behavior intent feature is determined, and each user's historical behavior data corresponds to historical behavior intent feature information.

[0545] The first task intent is determined based on each of the first time decay values, each of the first similarities, each of the historical behavioral intent feature information, and the first fusion feature information.

[0546] In some embodiments of this application, the processor 910 is specifically used for:

[0547] Based on each of the first similarities, N first feature information that satisfy the similarity condition with the first fused feature information are determined among each of the first feature information.

[0548] Based on the first similarity corresponding to each of the N first feature information and the first time decay value corresponding to each of the N first feature information, a weighted operation is performed on the N first feature information to obtain the second feature information;

[0549] The first fused feature information, the second feature information, and the third fused feature information are weighted and fused to obtain the third fused feature information, wherein the second fused feature information is the fused feature information of the at least one historical behavioral intention feature information.

[0550] Based on the third fusion feature information, the first task intent is determined; where N is a positive integer.

[0551] In some embodiments of this application, the processor 910 is specifically used for:

[0552] The third fused feature information is input into the trained task intent prediction model, which outputs at least one task intent.

[0553] The task intent with the highest confidence level among the at least one task intent is determined as the first task intent.

[0554] In some embodiments of this application, the first input is an input that is dragged from a first position coordinate to a second position coordinate, and the processor 910 is further configured to:

[0555] Before determining the task intent with the highest confidence among the at least one task intent as the first task intent, calculate the second similarity between the second task intent feature information and the third fused feature information;

[0556] Obtain the matching degree between the second task intent and the third application, wherein the matching degree is used to indicate whether the third application can execute the task corresponding to the second task intent;

[0557] Based on the second similarity and the second time decay value, a first confidence score corresponding to the second task intention is determined. The first confidence score is used to indicate the confidence of the second task intention in the time dimension.

[0558] Obtain a second confidence score corresponding to the second task intent, the second confidence score being used to indicate the likelihood of executing the task corresponding to the second task intent in the third application;

[0559] Based on the second similarity, the matching degree, the first confidence score, and the second confidence score, a third confidence score for the second task intent is determined;

[0560] Wherein, the second task intent is any one of the at least one task intents, the second task intent feature information is the task intent feature information corresponding to the second task intent, the second time decay value is the time decay value of the first historical behavior data in the at least one historical behavior data, the first historical behavior data is the historical behavior data corresponding to the historical behavior intent feature information that matches the second task intent feature information in the at least one historical behavior data, the first position coordinate is located within the display area of ​​the first information, the second position coordinate is located within the third interface, and the third interface is the application interface of the third application.

[0561] In some embodiments of this application, the processor 910 is specifically used for:

[0562] The second similarity, the matching degree, the first confidence score, and the second confidence score are weighted to obtain the confidence weighted value corresponding to the second task intent;

[0563] Calculate the exponential mapping value of the confidence weighted value corresponding to the second task intent;

[0564] Based on the index mapping value corresponding to the second task intent, and the sum of all index mapping values ​​corresponding to the at least one task intent, a third confidence score for the second task intent is determined.

[0565] In some embodiments of this application, the processor 910 is specifically used for:

[0566] From the at least one set of historical behavior data, determine M sets of historical behavior data that match the third application;

[0567] Based on the M historical behavioral data, determine the M historical behavioral intentions;

[0568] Determine the first number of historical behavioral intention categories corresponding to M historical behavioral intentions;

[0569] Determine a second number of the second task intentions from the M historical behavioral intentions;

[0570] The second confidence score is determined based on the second quantity, the first quantity, and M;

[0571] Where M is a positive integer.

[0572] In the electronic device provided in this application embodiment, the electronic device receives a first input from a user on first information in a first interface; responds to the first input and displays the task execution result corresponding to a first task intent. The first task intent is determined based on first information and second information, the second information including at least one of the following: interface context information of the first interface and user historical behavior data. Thus, when the electronic device receives input from a user on the first information, it can quickly and accurately identify the user's task intent based on at least one of the first information, the interface context information of the first interface, and the user's historical behavior data, and then display the task execution result corresponding to the user's task intent, thereby improving the task execution efficiency of the electronic device.

[0573] It should be understood that, in this embodiment, the input unit 904 may include a graphics processing unit (GPU) 9041 and a microphone 9042. The GPU 9041 processes image data of still images or videos obtained by an image capture device (such as a camera) in video capture mode or image capture mode. The display unit 906 may include a display panel 9061, which may be configured in the form of a liquid crystal display, an organic light-emitting diode, or the like. The user input unit 907 includes at least one of a touch panel 9071 and other input devices 9072. The touch panel 9071 is also called a touch screen. The touch panel 9071 may include a touch detection device and a touch controller. Other input devices 9072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, power buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.

[0574] The memory 909 can be used to store software programs and various data. The memory 909 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, applications or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 909 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 909 in the embodiments of this application includes, but is not limited to, these and any other suitable types of memory.

[0575] Processor 910 may include one or more processing units; optionally, processor 910 integrates an application processor and a modem processor, wherein the application processor mainly handles operations involving the operating system, user interface, and applications, and the modem processor mainly handles wireless communication signals, such as a baseband processor. It is understood that the aforementioned modem processor may also not be integrated into processor 910.

[0576] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described task execution method embodiments and achieve the same technical effects. To avoid repetition, they will not be described again here.

[0577] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0578] This application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the various processes of the above-described task execution method embodiments and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0579] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0580] This application provides a computer program product that is stored in a storage medium and executed by at least one processor to implement the various processes of the task execution method embodiments described above, and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0581] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element. Furthermore, it should be noted that the scope of the methods and apparatuses in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may also include performing functions substantially simultaneously or in the reverse order, depending on the functions involved. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

[0582] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0583] The embodiments of this application have been described above with reference to the accompanying drawings. However, this application is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of this application without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of this application.

Claims

1. A task execution method, characterized in that, include: Receive the user's first input on the first information displayed on the first interface; In response to the first input, the task execution result corresponding to the first task intent is displayed; The first task intent is determined based on the first information and the second information, the second information including at least one of the following: the interface context information of the first interface and the user's historical behavior data.

2. The method according to claim 1, characterized in that, Before displaying the task execution result corresponding to the first task intent, the method further includes: Display at least one task control corresponding to the first task intent, wherein each task control indicates a task; Receive a second input from the user to a first task control in the at least one task control; The display of the task execution result corresponding to the first task intent includes: In response to the second input, the task execution result of the task indicated by the first task control is displayed.

3. The method according to claim 2, characterized in that, The at least one task control constitutes a control wheel; each task control is a sector control in the control wheel, and the angle of the sector control is determined based on the confidence level of the task intent corresponding to the sector control.

4. The method according to claim 1, characterized in that, The first interface is the application interface of the first application, and the interface context information includes at least one of the following: the interface content of the first interface, the application attribute information of the first application, the geographical location information of the electronic device, and the second reference information associated with the first information; The application attribute information includes at least one of the following: application name, application type, and application function; the second reference information includes the interface content of the second interface, which is the second application interface of the second application that the user has accessed in the past, and the second application interface includes the first information.

5. The method according to claim 1, characterized in that, The user's historical behavior data includes at least one; The method further includes: Obtain the first characteristic information of each user's historical behavior data; Obtain the first fusion feature information corresponding to the first information and the interface context information; The first task intent is determined based on the user's historical behavior data, the first feature information, and the first fused feature information.

6. The method according to claim 5, characterized in that, Determining the first task intent based on each of the user's historical behavior data, each of the first feature information, and the first fused feature information includes: Obtain the first time decay value for each historical behavior data point, which is used to indicate the importance of the user's historical behavior data in the time dimension; Calculate the first similarity between each first feature information and the first fused feature information; Based on the historical behavior data of each user, at least one historical behavior intent feature is determined, and each user's historical behavior data corresponds to historical behavior intent feature information. The first task intent is determined based on each of the first time decay values, each of the first similarities, each of the historical behavioral intent feature information, and the first fusion feature information.

7. The method according to claim 6, characterized in that, The step of determining the first task intent based on each of the first time decay values, each of the first similarities, each of the historical behavioral intent feature information, and the first fused feature information includes: Based on each of the first similarities, N first feature information that satisfy the similarity condition with the first fused feature information are determined among each of the first feature information. Based on the first similarity corresponding to each of the N first feature information and the first time decay value corresponding to each of the N first feature information, a weighted operation is performed on the N first feature information to obtain the second feature information; The first fused feature information, the second feature information, and the third fused feature information are weighted and fused to obtain the third fused feature information, wherein the second fused feature information is the fused feature information of the at least one historical behavioral intention feature information. Based on the third fusion feature information, the first task intent is determined; where N is a positive integer.

8. The method according to claim 7, characterized in that, Determining the first task intent based on the third fused feature information includes: The third fused feature information is input into the trained task intent prediction model, which outputs at least one task intent. The task intent with the highest confidence level among the at least one task intent is determined as the first task intent.

9. The method according to claim 8, characterized in that, The first input is the input that is dragged from the first position to the second position; Before determining the task intent with the highest confidence among the at least one task intents as the first task intent, the method further includes: Calculate the second similarity between the second task intent feature information and the third fused feature information; Obtain the matching degree between the second task intent and the third application, wherein the matching degree is used to indicate whether the third application can execute the task corresponding to the second task intent; Based on the second similarity and the second time decay value, a first confidence score corresponding to the second task intention is determined. The first confidence score is used to indicate the confidence of the second task intention in the time dimension. Obtain a second confidence score corresponding to the second task intent, the second confidence score being used to indicate the likelihood of executing the task corresponding to the second task intent in the third application; Based on the second similarity, the matching degree, the first confidence score, and the second confidence score, a third confidence score for the second task intent is determined; Wherein, the second task intent is any one of the at least one task intents, the second task intent feature information is the task intent feature information corresponding to the second task intent, the second time decay value is the time decay value of the first historical behavior data in the at least one historical behavior data, the first historical behavior data is the historical behavior data corresponding to the historical behavior intent feature information that matches the second task intent feature information in the at least one historical behavior data, the first position is located in the display area of ​​the first information, the second position is located in the third interface, and the third interface is the application interface of the third application.

10. The method according to claim 9, characterized in that, The step of determining the third confidence score of the second task intent based on the second similarity, the matching degree, the first confidence score, and the second confidence score includes: The second similarity, the matching degree, the first confidence score, and the second confidence score are weighted to obtain the confidence weighted value corresponding to the second task intent; Calculate the exponential mapping value of the confidence weighted value corresponding to the second task intent; Based on the index mapping value corresponding to the second task intent, and the sum of all index mapping values ​​corresponding to the at least one task intent, a third confidence score for the second task intent is determined.

11. The method according to claim 9, characterized in that, The step of obtaining the second confidence score corresponding to the second task intent includes: From the at least one set of historical behavior data, determine M sets of historical behavior data that match the third application; Based on the M historical behavioral data, determine the M historical behavioral intentions; Determine the first number of historical behavioral intent categories corresponding to the M historical behavioral intents; Determine a second number of the second task intentions from the M historical behavioral intentions; The second confidence score is determined based on the second quantity, the first quantity, and M; Where M is a positive integer.

12. A task execution device, characterized in that, include: The receiving module is used to receive the user's first input on the first information in the first interface; The display module is used to respond to the first input and display the task execution result corresponding to the first task intent. The first task intent is determined based on the first information and the second information. The second information includes at least one of the following: the interface context information of the first interface and the user's historical behavior data.

13. The apparatus according to claim 12, characterized in that, The display module is also used for: Before displaying the task execution result corresponding to the first task intent, at least one task control corresponding to the first task intent is displayed, and each task control in the at least one task control indicates a task; The receiving module is further configured to receive a second input from the user to the first task control in the at least one task control; The display module is specifically used for: In response to the second input, the task execution result of the task indicated by the first task control is displayed; Wherein, the first task control is any one of the at least one task controls.

14. An electronic device, characterized in that, It includes a processor and a memory, the memory storing a program or instructions that can run on the processor, the program or instructions being executed by the processor to implement the steps of the task execution method as described in any one of claims 1-11.

15. A computer-readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the task execution method as described in any one of claims 1-11.

Images