CN118051287A - Application recommendation method and electronic device - Google Patents

Abstract

The application provides an application program recommending method and electronic equipment, and relates to the technical field of computers. The scheme improves the application recommendation hit rate of the electronic equipment. The specific scheme is as follows: the electronic equipment displays a first interface, wherein the first interface comprises a first control, and the first control is used for displaying application icons of application programs to be recommended to a user; when the electronic equipment is located in a first position area and the system time belongs to a first period, the electronic equipment displays an application icon of a first application in the first control; the first weight of the first application is higher than the first weight of the second application, the application icon of the second application is not displayed on the first control, the first weight is used for indicating the enabling probability of the application program in the scene which is located in the first position area, and the system time belongs to the first period.

Description

Application recommendation method and electronic device

Technical Field

The present application relates to the field of computer technology, and in particular to an application recommendation method and electronic device.

Background technique

At present, the number of applications installed in electronic devices is increasing. In order to facilitate users to quickly start a target application from a large number of applications, the electronic device can help users select some applications, such as recommended applications, and centrally display application icons of the recommended applications, thereby facilitating users to find and start the target application.

At present, electronic devices have a high accuracy in recommending high-frequency applications, but a low accuracy in recommending long-tail applications. Long-tail applications are applications with low usage popularity. Although long-tail applications are generally less popular, in certain specific scenarios, they are applications that users are likely to use.

Summary of the invention

The embodiments of the present application provide an application recommendation method and an electronic device for improving the hit rate of application recommendations.

To achieve the above objectives, the embodiments of the present application adopt the following technical solutions:

In a first aspect, an application recommendation method provided by an embodiment of the present application is applied to an electronic device, the method comprising: the electronic device displays a first interface, the first interface includes a first control, and the first control is used to display an application icon of an application to be recommended to a user. When the electronic device is located in a first location area and the system time belongs to a first time period, the electronic device displays the application icon of the first application in the first control; wherein the first weight of the first application is higher than the first weight of the second application, the application icon of the second application is not displayed in the first control, and the first weight is used to indicate the probability of enabling the application in the scenario where the electronic device is located in the first location area and the system time belongs to the first time period.

In the above embodiment, the electronic device can use the time scene characteristics and the space scene characteristics to determine the first application with regular usage in the time and space scene. The above first application can include applications with lower usage popularity. In this way, after the electronic device displays the application icon of the first application in the first control, it can increase the exposure of applications with lower usage popularity but regular usage, thereby effectively improving the hit rate of application recommendations.

In some embodiments, the method also includes: when the electronic device is located in a first location area and the system time belongs to a first time period, the electronic device displays an application icon of a third application in the first control; wherein the second weight of the third application is higher than the second weight of the fourth application, and the application icon of the fourth application is not displayed in the first control, and the second weight is used to indicate the probability of the application being used by the user in the current scenario, and the second weight can be determined by the frequency of interaction between the user and the application in various scenarios (scenarios indicated by multiple scenario characteristics) in historical operation data.

In the above embodiment, the electronic device can not only recommend to the user the first application determined by combining the spatiotemporal characteristics, but also recommend to the user the third application determined by combining multiple scene features that can characterize the current scene, thereby avoiding missing out on recommending long-tail applications that are regularly used in scenes with fixed spatiotemporal characteristics, and improving the accuracy of recommendations.

In some embodiments, the electronic device includes a first recommendation model and a second recommendation model, the first recommendation model is used to determine the second weight corresponding to each of the applications in different scenarios, and the second recommendation model is used to determine the first weight corresponding to each of the applications in different scenarios. Before the electronic device displays the first application and the third application in the first control, the method further includes: the electronic device obtains the first time and the first location information; when the first time belongs to the first time period and the first location information indicates the first location area, the electronic device uses the second recommendation model to determine the first application, and when the first weights of the applications are arranged in order from large to small, the first application is the application whose first weight is arranged before the first place. In addition, the electronic device can also use the first recommendation model to determine the third application, and when the second weights of the applications are arranged in order from large to small, the third application is the application whose second weight is arranged before the second place.

In some embodiments, the first location information includes one or a combination of a service set identifier provided by a first network, a cell identifier provided by a first base station, and first longitude and latitude information, wherein a signal coverage area of the first network and the first base station belongs to the first location area.

In some embodiments, before the electronic device displays the first application in the first control, the method further includes: the electronic device obtains a first day number and a second day number, the first day number is the number of days when the first application has been enabled in a scenario where the electronic device is located in the first location area and the system time belongs to a first time period, and the second day number is the number of days when the electronic device detects that it is located in the first location area and the system time belongs to the first time period; the electronic device determines, based on the first day number and the second day number, a first weight corresponding to the first application when the electronic device is located in the first location area and the system time belongs to the first time period; the electronic device updates the first recommendation model based on the first application and the corresponding first weight.

In the above embodiment, the electronic device can update the first weight corresponding to the first application in the first recommendation model in various spatiotemporal scenarios according to the change in the number of days the user uses the first application in various spatiotemporal scenarios. Of course, the electronic device can also update the first weights of other applications in different spatiotemporal scenarios in the same way, so that the recommended applications determined by the electronic device in various spatiotemporal scenarios can change with the user's habits.

In some embodiments, the method also includes: the electronic device obtains interaction data, the interaction data includes the number of interactions between the user and each type of application when different location information, different time information, different user status information, and different device status information are detected; the electronic device performs clustering based on the location information, the time information, the user status information, and the device status information in the interaction data to obtain the second recommendation model.

In the above embodiment, the second recommendation model can improve the recognition of real-time scenes indicated by multiple scene features by learning interaction data with multiple scene features, thereby improving the matching degree of application recommendations for various real-time scenes.

In some embodiments, the electronic device obtains interaction data, including: when the electronic device detects a first operation for any of the applications, the electronic device collects the location information, time information, user status information, and device status information; wherein, when the first service set identifier and the first cell identifier are collected, the first service set identifier is determined to be the detected location information, and the first cell identifier is associated with the first service set identifier; when the first cell identifier is collected and the first cell identifier has been associated with the first service set identifier, the first service set identifier is determined to be the detected location information.

In some embodiments, when the service set identifier and the cell identity identifier are not collected when the first operation is detected, the electronic device determines that the location information is a first identifier, and the first identifier indicates an unidentified geographic space.

In some embodiments, when the first application includes a fifth application and a sixth application, and the third application includes a sixth application and a seventh application, the method further includes: the electronic device determines the recommendation weight of the sixth application based on the first weight, the second weight, the first recommendation factor and the second recommendation factor corresponding to the sixth application, the first recommendation factor being used to indicate the recommendation hit rate of the first recommendation model, and the second recommendation factor being used to indicate the recommendation hit rate of the second recommendation model; the electronic device determines the recommendation weight of the sixth application based on the first weight and the first recommendation factor corresponding to the fifth application; the electronic device determines the recommendation weight of the seventh application based on the second weight and the second recommendation factor corresponding to the seventh application.

In some embodiments, the application icons of the fifth application, the sixth application, and the seventh application are arranged on the first control in descending order according to the corresponding recommendation weights. In addition, as the scene in which the electronic device is located changes, the recommended application determined by the electronic device will also change, so that the application icons in the first control will also change. After the application icons in the first control change, the application icons in the first control can be arranged irregularly.

In a second aspect, an electronic device is provided by an embodiment of the present application, and the electronic device includes one or more processors and a memory; the memory is coupled to the processor, and the memory is used to store computer program code, and the computer program code includes computer instructions. When the one or more processors execute the computer instructions, the one or more processors are used to: display a first interface, wherein the first interface includes a first control, and the first control is used to display an application icon of an application to be recommended to a user; when the electronic device is located in a first location area and the system time belongs to a first time period, the application icon of the first application is displayed in the first control; wherein the first weight of the first application is higher than the first weight of the second application, the application icon of the second application is not displayed in the first control, and the first weight is used to indicate the probability of enabling the application in a scenario where the electronic device is located in the first location area and the system time belongs to the first time period.

In some embodiments, the one or more processors are used to: display an application icon of a third application in the first control when the electronic device is located in the first location area and the system time belongs to the first time period; wherein the second weight of the third application is higher than the second weight of the fourth application, the application icon of the fourth application is not displayed in the first control, and the second weight is used to indicate the frequency of interaction between the user and the application in the current scenario.

In some embodiments, the electronic device includes a first recommendation model and a second recommendation model, the first recommendation model is used to determine the second weight corresponding to each of the applications in different scenarios, and the second recommendation model is used to determine the first weight corresponding to each of the applications in different scenarios. Before the electronic device displays the first application and the third application in the first control, the one or more processors are used to: obtain first time and first location information; when the first time belongs to the first time period and the first location information indicates the first location area, use the second recommendation model to determine the first application, and when the first weights of the applications are arranged in order from large to small, the first application is the application whose first weight is arranged before the first place; use the first recommendation model to determine the third application, and when the second weights of the applications are arranged in order from large to small, the third application is the application whose second weight is arranged before the second place.

In some embodiments, the first location information includes one or a combination of a service set identifier provided by a first network, a cell identifier provided by a first base station, and first longitude and latitude information, wherein a signal coverage area of the first network and the first base station belongs to the first location area.

In some embodiments, before displaying the first application in the first control, the one or more processors are used to: obtain a first day and a second day, the first day being the number of days on which the first application has been enabled when the electronic device is located in the first location area and the system time belongs to a first time period, and the second day being the number of days on which the electronic device detects that it is located in the first location area and the system time belongs to the first time period; determine, based on the first day and the second day, a first weight corresponding to the first application when the electronic device is located in the first location area and the system time belongs to the first time period; and update the first recommendation model based on the first application and the corresponding first weight.

In some embodiments, the one or more processors are used to: obtain interaction data, the interaction data including the number of interactions between the user and each type of application when different location information, different time information, different user status information, and different device status information are detected; and perform clustering based on the location information, the time information, the user status information, and the device status information in the interaction data to obtain the second recommendation model.

In some embodiments, the one or more processors are used to: collect the location information, time information, user status information, and device status information when a first operation for any of the applications is detected; wherein, when a first service set identifier and a first cell identifier are collected, the first service set identifier is determined to be the detected location information, and the first cell identifier is associated with the first service set identifier; when the first cell identifier is collected and the first cell identifier has been associated with the first service set identifier, the first service set identifier is determined to be the detected location information.

In some embodiments, the one or more processors are used to: when detecting that a service set identifier and a cell identity identifier are not collected when the first operation is performed, determine that the location information is a first identifier, and the first identifier indicates an unidentified geographic space.

In some embodiments, the one or more processors are used to: when the first application includes the fifth application and the sixth application, and the third application includes the sixth application and the seventh application, the one or more processors are used to: determine the recommendation weight of the sixth application based on the first weight, the second weight, the first recommendation factor and the second recommendation factor corresponding to the sixth application, the first recommendation factor is used to indicate the recommendation hit rate of the first recommendation model, and the second recommendation factor is used to indicate the recommendation hit rate of the second recommendation model; determine the recommendation weight of the sixth application based on the first weight and the first recommendation factor corresponding to the fifth application; determine the recommendation weight of the seventh application based on the second weight and the second recommendation factor corresponding to the seventh application.

In a third aspect, an embodiment of the present application provides a computer storage medium, including computer instructions. When the computer instructions are executed on an electronic device, the electronic device executes the method in the above-mentioned first aspect and possible embodiments thereof.

In a fourth aspect, the present application provides a computer program product. When the computer program product is run on the above-mentioned electronic device, the electronic device executes the method in the above-mentioned first aspect and possible embodiments thereof.

It can be understood that the electronic devices, computer storage media and computer program products provided in the above-mentioned aspects are all applied to the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above and will not be repeated here.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 is an example diagram of a display interface of an electronic device provided in the related art;

FIG2 is one of the exemplary diagrams of the display interface of the electronic device provided in the embodiment of the present application;

FIG3 is a diagram showing an example of a hardware structure of an electronic device provided in an embodiment of the present application;

FIG4 is a second example diagram of a display interface of an electronic device provided in an embodiment of the present application;

FIG5 is an example diagram of determining a location tag in an embodiment of the present application;

FIG6 is an example diagram of a geo-fence provided in an embodiment of the present application;

FIG. 7 is a third example diagram of a display interface of an electronic device provided in an embodiment of the present application;

FIG8 is a fourth example diagram of a display interface of an electronic device provided in an embodiment of the present application;

FIG9 is a flow chart of an application recommendation method provided in an embodiment of the present application;

FIG10 is an example diagram of a decision tree model provided in an embodiment of the present application;

FIG11 is an example diagram of determining recommendation factors of model 1 provided in an embodiment of the present application;

FIG. 12 is an example diagram of a chip system provided in an embodiment of the present application.

Detailed ways

In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this embodiment, unless otherwise specified, "plurality" means two or more.

The implementation of this embodiment will be described in detail below with reference to the accompanying drawings.

The embodiment of the present application provides an application recommendation method, which is applied to an electronic device, wherein a plurality of applications (including system applications and third-party applications) are installed in the electronic device, and different applications can provide different application services to users.

In order to facilitate users to use the application services provided by the application, application icons of various applications are displayed on the desktop of the electronic device. The above application icons can be used as the application startup entry for the user to start (or open) the corresponding application. That is, after the electronic device receives the user's operation on the application icon, it can run the corresponding application, such as displaying the application interface of the application.

As more and more applications are installed in electronic devices, more and more application icons need to be displayed. In this scenario, the application icons of the application can be displayed on multiple sub-screens on the desktop. Of course, the application icons displayed on different sub-screens can be different. In the related art, when a user wants to use the application service provided by a specified application, he needs to flip through multiple sub-screens to find the application icon corresponding to the specified application.

For example, as shown in FIG1 , the unlocked electronic device displays the first sub-screen of the desktop (such as interface 101), and the main interface displays application icons corresponding to multiple applications. While the electronic device displays interface 101, the user wants to use WeChat ^TM . If the application icon of WeChat ^TM is not displayed in interface 101, the user needs to operate the electronic device to trigger the electronic device to switch to display other sub-screens, so that the user can find the application icon of WeChat ^TM on other sub-screens. That is, the user can perform a sliding operation on the display screen of the electronic device to trigger the electronic device to switch to display different sub-screens until the user finds the application icon of WeChat ^TM .

For example, during the display of interface 101, the electronic device may display interface 102 in response to a user's sliding operation on the display screen. In the case where the interface 102 includes an application icon 103 of WeChat ^™ , the user may click on the application icon 103 to trigger the electronic device to start WeChat ^™ . If there is no application icon 103 on the interface 102, the user needs to continue sliding on the display screen to trigger the electronic device to switch the display interface to other sub-screens. Obviously, the human-computer interaction efficiency of the whole process is low.

In order to improve the above problems, an embodiment of the present application provides an application recommendation method. In this solution, the electronic device can display a recommendation control during the screen-on period. Exemplarily, the above recommendation control can be displayed on any display interface of the electronic device, such as a lock screen interface, a sub-screen, a negative first screen, and an application interface, etc. The embodiment of the present application does not limit this. The display interface for displaying the recommendation control can also be referred to as a first interface, and the above recommendation control can also be referred to as a first control.

The recommendation control includes one or more application icons, which may be referred to as recommendation icons. The recommendation icons are application icons of recommended applications determined by the electronic device. The recommended applications may include applications that the electronic device evaluates from installed applications that the user wishes to use in the current scenario.

Exemplarily, the electronic device can identify the current scenario of the electronic device from one or more dimensions such as time, space, and state (e.g., may include user state or device state), and then determine the applications that the user may use in the current scenario based on the applications that are ranked high in popularity among users in various scenarios. The above-mentioned usage popularity is used to indicate the frequency of interaction between the user and the application.

As another example, the electronic device can identify the scene in which the electronic device is currently located from one or more dimensions such as time and space, and then determine the applications that the user may use in the current scene based on the applications that the user uses regularly in various scenes (such as the regularity of the time periods that are enabled and the regularity of the location of the device when using the application). For example, the electronic device will start the attendance application once every morning at 9:00, so it can be determined that in the scene of 9:00 in the morning, the attendance application is a regularly used application. For another example, the electronic device will enable the public transportation application at the bus stop, so it can be determined that the public transportation application is a regularly used application in the scene of the bus stop.

In this way, in different scenarios, the electronic device can recommend recommended icons that meet the user's current needs to the user through the recommendation control.

For example, as shown in FIG. 2 , taking the electronic device as a mobile phone as an example, the main interface (e.g., interface 201) can be displayed after the mobile phone is unlocked. The interface 201 includes recommendation controls, such as a recommendation card 202. The recommendation card 202 includes application icons of recommended applications such as WeChat ^™ , public transportation, gallery, and video. The above-mentioned applications such as WeChat ^™ , public transportation, gallery, and video are all applications that the electronic device evaluates to be likely to use in the user's current scenario. If the user wants to use WeChat ^™ at this time, then the application icon of WeChat ^™ can be selected in the recommendation card 202 to trigger the electronic device to enable WeChat ^™ . In this way, the user saves the operation of triggering the electronic device to switch to display different sub-screens, thereby improving the efficiency of human-computer interaction and the intelligence of the electronic device.

For example, the electronic device can be not only a mobile phone, but also a television, a tablet computer, a desktop, a laptop, a handheld computer, a notebook computer, a personal computer (PC), a netbook, a personal digital assistant (PDA), and other electronic devices with a display screen. The embodiments of the present application do not impose any special restrictions on the specific form of the electronic device.

As shown in Figure 3, the embodiment of the present application takes a mobile phone as an example to illustrate the structure of the electronic device provided in the embodiment of the present application. The electronic device (such as a mobile phone) may include: a processor 310, an external memory interface 320, an internal memory 321, a universal serial bus (USB) interface 330, a charging management module 340, a power management module 341, a battery 342, an antenna 1, an antenna 2, a mobile communication module 350, a wireless communication module 360, an audio module 370, a speaker 370A, a receiver 370B, a microphone 370C, an earphone interface 370D, a sensor module 380, a button 390, a motor 391, an indicator 392, a camera 393, a display screen 394, and a subscriber identification module (SIM) card interface 395, etc.

Among them, the above-mentioned sensor module 380 may include sensors such as pressure sensor, gyroscope sensor, air pressure sensor, magnetic sensor, acceleration sensor, distance sensor, proximity light sensor, fingerprint sensor, temperature sensor, touch sensor, ambient light sensor and bone conduction sensor.

It is to be understood that the structure illustrated in this embodiment does not constitute a specific limitation on the electronic device. In other embodiments, the electronic device may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

The processor 310 may include one or more processing units, for example, the processor 310 may include an application processor (AP), a modem processor, a graphics processor (GPU), an image signal processor (ISP), a controller, a memory, a video codec, a digital signal processor (DSP), a baseband processor, and/or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.

The controller can be the nerve center and command center of the electronic device. The controller can generate operation control signals according to the instruction operation code and timing signal to complete the control of fetching and executing instructions.

The processor 310 may also be provided with a memory for storing instructions and data. In some embodiments, the memory in the processor 310 is a cache memory. The memory may store instructions or data that the processor 310 has just used or cyclically used. If the processor 310 needs to use the instruction or data again, it may be directly called from the memory. This avoids repeated access, reduces the waiting time of the processor 310, and thus improves the efficiency of the system.

In some embodiments, the processor 310 may include one or more interfaces. The interface may include an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver/transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input/output (GPIO) interface, a subscriber identity module (SIM) interface, and/or a universal serial bus (USB) interface, etc.

It is understandable that the interface connection relationship between the modules illustrated in this embodiment is only a schematic illustration and does not constitute a structural limitation of the electronic device. In other embodiments, the electronic device may also adopt different interface connection methods in the above embodiments, or a combination of multiple interface connection methods.

The charging management module 340 is used to receive charging input from a charger. The charger can be a wireless charger or a wired charger. While the charging management module 340 is charging the battery 342, it can also power the electronic device through the power management module 341.

The power management module 341 is used to connect the battery 342, the charging management module 340 and the processor 310. The power management module 341 receives input from the battery 342 and/or the charging management module 340, and supplies power to the processor 310, the internal memory 321, the external memory, the display screen 394, the camera 393, and the wireless communication module 360. In some embodiments, the power management module 341 and the charging management module 340 can also be set in the same device.

The wireless communication function of the electronic device can be implemented through antenna 1, antenna 2, mobile communication module 350, wireless communication module 360, modem processor and baseband processor. In some embodiments, antenna 1 and mobile communication module 350 of the electronic device are coupled, and antenna 2 and wireless communication module 360 are coupled, so that the electronic device can communicate with the network and other devices through wireless communication technology, such as wearable devices.

Antenna 1 and antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in the electronic device can be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve the utilization of the antennas. For example, antenna 1 can be reused as a diversity antenna for a wireless local area network. In some other embodiments, the antenna can be used in combination with a tuning switch.

The mobile communication module 350 can provide solutions for wireless communications including 2G/3G/4G/5G, etc., applied to electronic devices. The mobile communication module 350 may include at least one filter, a switch, a power amplifier, a low noise amplifier (LNA), etc. The mobile communication module 350 can receive electromagnetic waves from the antenna 1, and filter, amplify, etc. the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation.

The mobile communication module 350 can also amplify the signal modulated by the modem processor and convert it into electromagnetic waves for radiation through the antenna 1. In some embodiments, at least some functional modules of the mobile communication module 350 can be set in the processor 310. In some embodiments, at least some functional modules of the mobile communication module 350 can be set in the same device as at least some modules of the processor 310.

The wireless communication module 360 can provide wireless communication solutions for electronic devices including WLAN (such as (wirelessfidelity, Wi-Fi) network), Bluetooth (bluetooth, BT), global navigation satellite system (global navigation satellite system, GNSS), frequency modulation (frequency modulation, FM), nearfield communication technology (nearfield communication, NFC), infrared technology (infrared, IR), etc.

Among them, GNSS may include Beidou navigation satellite system (BDS), global positioning system (GPS), global navigation satellite system (GLONASS), quasi-zenith satellite system (QZSS) and/or satellite based augmentation systems (SBAS).

The wireless communication module 360 may be one or more devices integrating at least one communication processing module. The wireless communication module 360 receives electromagnetic waves via the antenna 2, modulates the electromagnetic wave signal and performs filtering, and sends the processed signal to the processor 310. The wireless communication module 360 may also receive a signal to be sent from the processor 310, modulate the signal, amplify the signal, and convert it into an electromagnetic wave for radiation via the antenna 2.

The electronic device implements the display function through a GPU, a display screen 394, and an application processor. The GPU is a microprocessor for image processing, connecting the display screen 394 and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. The processor 310 may include one or more GPUs that execute program instructions to generate or change display information.

The display screen 394 is used to display images, videos, etc. The display screen 394 includes a display panel.

The electronic device can realize the shooting function through ISP, camera 393, video codec, GPU, display screen 394 and application processor. ISP is used to process the data fed back by camera 393. Camera 393 is used to capture static images or videos. In some embodiments, the electronic device may include 1 or N cameras 393, where N is a positive integer greater than 1.

The external memory interface 320 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the electronic device. The external memory card communicates with the processor 310 through the external memory interface 320 to implement a data storage function. For example, files such as music and videos can be stored in the external memory card.

The internal memory 321 may be used to store computer executable program codes, which include instructions. The processor 310 executes various functional applications and data processing of the electronic device by running the instructions stored in the internal memory 321. For example, in an embodiment of the present application, the processor 310 may execute the instructions stored in the internal memory 321, and the internal memory 321 may include a program storage area and a data storage area.

The program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc. The data storage area may store data created during the use of the electronic device (such as audio data, a phone book, etc.), etc. In addition, the internal memory 321 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc.

It is to be understood that the structures illustrated in the embodiments of the present application do not constitute specific limitations on the electronic device. In other embodiments of the present application, the electronic device may include more or fewer components than shown in the figure, or combine certain components, or split certain components, or arrange the components differently. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

The methods in the following embodiments can all be implemented in a device having the above hardware structure.

In some embodiments, during the operation of the electronic device, the interactive operations between the user and each application can be detected in real time. For example, during the display of the desktop, the user clicks the application icon corresponding to the application. Another example is during the display of the application interface provided by the application, the user operates any function control in the application interface. Another example is during the display of the multitasking interface, the user operates the thumbnail window corresponding to the application. Another example is the user triggers the operation of sending a control command to the application, etc.

In some embodiments, the electronic device may assign corresponding scene features and application identifiers to each detected interaction operation, that is, associate the interaction operation with the corresponding scene features and application identifiers. The scene features may be features in at least one dimension such as time, space, device status, and user status.

As an example, after an electronic device detects an interactive operation, the electronic device determines the application corresponding to the interactive operation. Then, the interactive operation is associated with the application identifier of the application, for example, the application identifier can be the package name of the application. Afterwards, the electronic device can obtain the time information of detecting the interactive operation, such as time information a, and use the time information a as the feature of the interactive operation in the time dimension.

Exemplarily, the time information a may include time information. For example, at 9 am, if a user triggers the operation of starting a public transportation application, the electronic device may associate the interaction operation with the public transportation application identifier, and may also associate the interaction operation with the system time information of 09:00:00.

As another example, the time information a may also include date information. For another example, at 9:00 a.m. on October 11, 2022, it is detected that the user has indicated an interaction operation with the public transportation application. The electronic device may associate the interaction operation with the public transportation application identifier, and may also associate the interaction operation with system time information such as October 11, 2022, Tuesday, 09:00:00, etc.

In this way, the electronic device can determine the top usage popularity (e.g., the usage popularity is ranked in the top k positions, k is a positive integer) corresponding to different time periods in the time dimension according to the time characteristics and application identification corresponding to the interactive operations, as the recommended applications corresponding to the time period. For example, the electronic device counts the number of interactive operations with the same application identification and time information belonging to the same time period based on the interactive operations detected within T days (a positive integer greater than 1, and the value of T can be set in advance). It can be understood that the same application label indicates the same application. In this way, through the above statistics, the electronic device can determine the number of interactive operations corresponding to each application in each number segment, that is, the electronic device can determine the top usage popularity corresponding to each time period according to the number of interactive operations corresponding to each application in each time period, that is, the application with a relatively high number of interactive operations.

In addition, the electronic device can determine the usage of each application according to the interactive operations corresponding to each application, and then evaluate the application with regular usage in the time dimension. It can be understood that in any time period (such as time period a), if the electronic device detects at least one interactive operation for any application (such as application b), then the electronic device can determine that the user used application b in time period a. As an exemplary example, the electronic device obtains the number of days that the user has used application b in time period a within T days. If the corresponding number of days of use is greater than m days, it can be determined that application b is a regularly used application in time period a. The above m is a positive integer greater than 0 and not greater than T, and the value of the above m can be related to the value of T. If application b is a regularly used application in time period a, then even if application b is not a top-ranked application in time period a, it can be determined as a recommended application in time period a.

That is, the electronic device can determine the recommended applications corresponding to each time feature according to the interactive operations corresponding to each time feature. In this way, during the actual operation of the electronic device, the electronic device can identify the time feature corresponding to the current scene according to the system time. If the time feature of the current scene corresponds to a recommended application, the electronic device can display the application icon of the recommended application corresponding to the current scene through the recommendation control. That is, when the electronic device detects that the system time reaches different time periods, different recommended applications can be displayed on the recommendation card. As shown in FIG4, the electronic device determines that the recommended applications corresponding to the time period to which the time 8:00 belongs include applications such as WeChat ^TM , public transportation, gallery, and video. When the system time is detected to reach 8:00, the electronic device can display the application icons of applications such as WeChat ^TM , public transportation, gallery, and video on the recommendation control (such as recommendation card 401). The electronic device determines that the recommended applications corresponding to the time period to which the time 10:00 belongs include applications such as takeout, memo, email, and conference. When the system time is detected to reach 10:00, the electronic device can display the application icons of applications such as takeout, memo, email, and conference on the recommendation control (such as recommendation card 402).

In other embodiments, after the electronic device detects an interactive operation, the electronic device determines the application corresponding to the interactive operation, then associates the interactive operation with the application identifier of the application, and then the electronic device can obtain a location tag corresponding to the current geographic space, and use the location tag as a feature of the interactive operation in the spatial dimension.

Exemplarily, the above-mentioned location tag can be a base station identifier carried in the base station signal collected by the electronic device, such as a cellular cell identity (cellID). It can be understood that after the electronic device enters the signal coverage range of the base station, it can access the base station and receive the cellID sent by the base station. Since the position of the base station is fixed, the spatial position of the base station signal that can be received (that is, the signal coverage range) is also fixed. In this way, the electronic device can indicate the current geographical location of the electronic device based on the cellID in the base station signal received in real time, that is, use the obtained cellID as the location tag. That is, when the electronic devices are connected to the same base station, the detected interactive operations correspond to the same location tag.

As another example, the location tag may also be a service set identifier (SSID) of a wireless-fidelity (Wi-Fi) network to which the electronic device accesses. It is understandable that the coverage of a Wi-Fi network is fixed, and electronic devices also need to enter the coverage of the Wi-Fi network to receive corresponding Wi-Fi information. The Wi-Fi information may carry an SSID, and different SSIDs may indicate different Wi-Fi networks. In this way, the electronic device may indicate the current geographical location of the electronic device based on the SSID obtained in real time, that is, use the obtained SSID as a location tag. That is, when the electronic device accesses the same WiFi network, the detected interactive operations correspond to the same location tag.

As another example, the location tag may also be the longitude and latitude information collected by a positioning system in the electronic device.

In some possible examples, the location tag may also be one of or a combination of cellID, SSID, and longitude and latitude information. For example, a combination of longitude and latitude information and cellID is used as a location tag. For another example, a combination of cellID and SSID is used as a location tag. In this application, the location tag only needs to be recognizable by the electronic device, and the content of the location tag is not specifically limited.

Taking the example that the location tag is a combination of longitude and latitude information and cellID, each time the electronic device detects an interactive operation, it can use the positioning system (such as GPS positioning technology) to collect the corresponding longitude and latitude information and obtain the cellID corresponding to the currently connected base station. In this way, each interactive operation corresponds to a longitude and latitude information and a cellID. Then, the longitude and latitude information is used to cluster the interactive operations to obtain multiple clusters. Each cluster includes multiple interactive operations, and the longitude and latitude information corresponding to each interactive operation is close to the location indicated in the geographic space. Then, the cellID corresponding to the interactive operation in the same cluster is used as the geographic fence of the spatial position indicated by the cluster. If the cellIDs corresponding to different interactive operations belong to the same geographic fence, the interactive operation has the same location tag. If the cellIDs corresponding to different interactive operations belong to different geographic fences, the interactive operation has different location tags.

Taking the location tag as a combination of longitude and latitude information and SSID as an example, each time the electronic device detects an interactive operation, it can use the positioning system (such as GPS positioning technology) to collect the corresponding longitude and latitude information and obtain the SSID corresponding to the currently connected Wi-Fi network. In this way, each interactive operation corresponds to a longitude and latitude information and an SSID. Then, the longitude and latitude information is used to cluster the interactive operations to obtain multiple clusters. Each cluster includes multiple interactive operations, and the longitude and latitude information corresponding to each interactive operation is close to the location indicated in the geographic space. Then, the SSID corresponding to the interactive operation in the same cluster is used as the geographic fence of the spatial position indicated by the cluster. If the SSIDs corresponding to different interactive operations belong to the same geographic fence, the interactive operation has the same location tag. If the SSIDs corresponding to different interactive operations belong to different geographic fences, the interactive operation has different location tags.

Taking the location tag being a combination of cellID and SSID as an example, each time the electronic device detects an interactive operation, the cellID corresponding to the base station currently accessed by the electronic device and/or the SSID corresponding to the accessed Wi-Fi network can be obtained.

For example, if the electronic device is connected to a base station but not to any Wi-Fi network, only the corresponding cellID can be obtained. If the electronic device is connected to a Wi-Fi network but not to any base station, only the corresponding SSID can be obtained. If the electronic device is connected to a base station and a Wi-Fi network at the same time, the corresponding cellID and SSID can be obtained at the same time. If the electronic device is not connected to a base station and a Wi-Fi network, no information may be obtained.

In some embodiments, the electronic device may determine the location tag of the interaction operation detected this time according to the acquisition of the SSID and/or cellID.

As an implementation, as shown in Fig. 5, the electronic device determines whether the cellID is obtained. If the electronic device does not obtain the cellID, that is, the electronic device is not connected to any base station, the identifier "-1" is used as the location tag of the interactive operation.

When the electronic device obtains the cellID, that is, when the electronic device has accessed any base station, the SSID matching the cellID is searched. Exemplarily, the electronic device queries the interaction operations collected within T days and corresponding to the same cellID, such as interaction operation a. The SSID corresponding to each interaction operation a is obtained as the matching SSID. If all interaction operations a do not correspond to an SSID (that is, there is no matching SSID), and the number of interaction operations a is greater than or equal to a specified value (such as 20), the obtained cellID is determined as the location tag of the interaction operation detected this time. If all interaction operations a do not correspond to an SSID, and the number of interaction operations a is less than a specified value (such as 20), the label "other" is determined as the location tag of the interaction operation detected this time.

In addition, when the interaction operation a includes at least one interaction operation b, where the interaction operation b is an operation corresponding to both the cellID and the SSID in the interaction operation a. If the SSIDs corresponding to the interaction operations b are the same (that is, there is only one matching SSID), the SSID corresponding to the interaction operation b is used as the location tag of the interaction operation detected this time.

For example, when an electronic device detects a first operation (i.e., an interactive operation) for any application, if a first service set identifier (e.g., an SSID provided by a WiFi network) and a first cell identifier (a cell ID provided by a base station) are collected, the first service set identifier can be determined as the location information (or location tag) corresponding to this detection, and the first cell identifier is associated with the first service set identifier. When the first cell identifier is collected and the first cell identifier is associated with the first service set identifier, the first service set identifier is determined to be the detected location information.

When interaction operation a includes at least one interaction operation b, if the SSIDs corresponding to interaction operations b are not all the same, then the number b of interaction operations b corresponding to different SSIDs is counted. In this way, the electronic device can use the SSID with the largest number b as the location tag of the interaction operation detected this time.

For example, as shown in Table 1:

Table 1

As shown in Table 1, interaction operation a is an interaction operation corresponding to cellID 1234. The above interaction operation a includes 430 interaction operations b. Among them, the number of interaction operations corresponding to SSID "223" in interaction operation b is 60, the number of interaction operations corresponding to SSID "224" in interaction operation b is 200, the number of interaction operations corresponding to SSID "225" in interaction operation b is 70, and the number of interaction operations corresponding to SSID "226" in interaction operation b is 100. In this way, SSID "224" can be used as the location tag of the interaction operation detected this time.

In addition, the coverage of the same Wi-Fi network may usually overlap with the coverage of multiple base stations. That is, when an electronic device accesses the same Wi-Fi network, it may access different base stations at different time periods. That is, when the SSID of the detected interactive operation is the same, the corresponding cellID may be different.

For example, as shown in Figure 6, in the home scenario, when the electronic device is connected to the home WiFi access node, the electronic device can access base station 1. Of course, the electronic device may also access base station 2, base station 3 or base station 4. When the electronic device detects an interactive operation, whether it is accessed to base station 1, base station 2, base station 3 or base station 4, the SSID corresponding to the home WiFi access node can be used as a location tag. In other words, whether it is accessed to base station 1, base station 2, base station 3 or base station 4, as long as they are all connected to the home WiFi access node, it can be determined that the electronic device has entered the same geographic space (that is, entered the home scenario). The cellIDs corresponding to the above base stations 1, 2, 3 and 4 can be used as geographic fences for home scenarios.

After the electronic device determines the location tag corresponding to each interactive operation, the electronic device can cluster the interactive operations detected within T days according to the location tag to obtain the number of interactive operations corresponding to each type of cellID under different location tags. For example, the results obtained after clustering are shown in Table 2 below:

Table 2

Among them, the above-mentioned 123123, 147147, 159159, 126126, 101010 and 148148 are cellIDs corresponding to different base stations, and the above-mentioned -1 represents no cellID. As shown in Table 2, the number of interaction operations with location tag 1 and corresponding to cellID "126126" is 116, the number of interaction operations with location tag 1 and corresponding to cellID "101010" is 112, the number of interaction operations with location tag 2 and corresponding to cellID "123123" is 390, the number of interaction operations with location tag 2 and no cellID is 23, the number of interaction operations with location tag 2 and corresponding to cellID "147147" is 1, the number of interaction operations with location tag 2 and corresponding to cellID "159159" is 1, the number of interaction operations with identifier "-1" is 190, and the number of interaction operations with identifier "other" and corresponding to cellID "148148" is 77. The above-mentioned identifier "-1" may refer to determining that the first service set identifier is a detected first identifier when the first service set identifier and the first cell identity identifier are collected.

Through the above clustering, it can be determined that the geo-fences corresponding to location tag 1 include cellID "126126" and cellID "101010", the geo-fences corresponding to location tag 2 include cellID "123123", cellID "147147" and cellID "159159", and the geo-fences corresponding to the identifier "other" include cellID "148148". The identifier "-1" indicates a scenario where the electronic device cannot collect the cellID, for example, it is located in a place without base station coverage, or the electronic device does not have a SIM card installed.

In some embodiments, when an interactive operation corresponds to a location tag, the electronic device can also determine the most popular applications for different location tags in the spatial dimension based on the location tag and application identifier corresponding to the interactive operation, and use them as recommended applications for the location tag. For example, under different location tags, the applications ranked in the top k in terms of usage frequency, where k is a positive integer.

For example, the electronic device counts the number of interactive operations with the same application identifier and the same location tag based on the interactive operations detected within T (a positive integer greater than 1) days. It can be understood that the same application tag indicates the same application, and the same location tag indicates the same geographic space, that is, the above number of interactive operations can indicate the number of times the user interacts with the same application in the same geographic space. In this way, through the above statistics, the electronic device can determine the number of interactions between the user and each application in different geographic spaces (which can be represented by location tags), and can also further determine the top applications (applications with a higher number of interactions) used by users in each geographic space, and use them as recommended applications corresponding to the geographic space, which can also be called recommended applications corresponding to the location tags.

In addition, the electronic device can determine the usage of each application according to the interactive operations corresponding to each application, that is, the usage of each application by the user in each geographic space (indicated by the location tag). Then, based on the spatial dimension, the application with the regular usage in each geographic space is evaluated and used as the recommended application of the geographic space, which can also be called the recommended application corresponding to the location tag.

For example, the electronic device determines the number of days that the electronic device appeared in geographic space a and used application b based on the interactive operations detected within T days. If the above-mentioned number of days is greater than m, it is determined that application b is an application with regular usage. If geographic space a is indicated by location tag a, then application b can be used as a recommended application corresponding to location tag a.

After determining the recommended applications corresponding to different location tags, when the electronic device detects that the current scene matches the geo-fence corresponding to the location tag, the recommended application corresponding to the location tag can be displayed on the recommendation card.

For example, after clustering the interactive operations with location tags, it is determined that the geo-fences corresponding to location tag 2 indicating the home scene include cellID "123123", cellID "147147" and cellID "159159", and it is also determined that the recommended applications corresponding to location tag 2 include applications such as WeChat ^TM , news, gallery, and video. It is determined that the geo-fences corresponding to location tag 1 indicating the company scene include cellID "126126" and cellID "101010", and it is also determined that the recommended applications corresponding to location tag 1 include applications such as document editing, office, email, and conference.

In this way, as shown in Figure 7, the electronic device detects that the current scene matches the geographic fence corresponding to the location tag 2. For example, by parsing the cellID "123123", cellID "147147" or cellID "159159" from the received base station signal, it can be determined that the electronic device is in the home scene. At this time, the electronic device can display application icons of applications such as ^WeChatTM , news, gallery, video, etc. on the recommendation control (such as recommendation card 701).

As shown in Figure 7, the electronic device detects that the current scene matches the geographic fence corresponding to the location tag 1. For example, by parsing the cellID "126126" or cellID "101010" from the received base station signal, it can be determined that the electronic device is located in the company scene. At this time, the electronic device can display application icons for document editing, office, email, conference, etc. on the recommendation control (such as recommendation card 702).

In other possible embodiments, after the electronic device detects an interactive operation, it can also obtain the device status information (such as power information) when the interactive operation is detected, and use it as a feature of the device status dimension. Then, based on the device status information corresponding to each interactive operation, the top-ranked applications in different device states and the applications with regular usage are determined as the corresponding recommended applications in the device state. In this way, it is possible to recommend adapted applications to users based on the real-time device state of the electronic device. For example, when the battery is too low, a power bank rental application is pushed.

In other possible embodiments, after the electronic device detects an interactive operation, the user state information when the interactive operation is detected (e.g., the user is in motion) can be obtained and used as a feature of the user state dimension. Then, based on the user state information corresponding to each interactive operation, the top-ranked applications and the applications with regular usage in different user states are determined as the corresponding recommended applications in the user state, thereby recommending the required applications to the user based on the user's real-time state. For example, as shown in Figure 8, when the user is in motion, the electronic device can display application icons for sports health, pedometer, camera, video, and other applications on the recommendation control (i.e., recommendation card 801).

Of course, during the actual operation of the electronic device, the electronic device can give multiple types of scene features to the detected interactive operations, such as having time information and location tags, etc. In this way, the electronic device can determine the recommended applications corresponding to the scenes indicated by the multiple types of scene features.

That is, in some embodiments, as shown in FIG9 , the electronic device detects an interactive operation (e.g., a user clicks on an application icon) and determines the scene features corresponding to the interactive operation, such as one or more of time information, location tag, device status information, user status information, etc. Based on the interactive operations detected within T days and the corresponding various scene feature data, a preset algorithm is used to train model 1 (the first recommended model) and model 2 (the second recommended model) respectively. Among them, the above-mentioned interactive operations and the corresponding various scene feature data can be collectively referred to as interactive data.

Among them, Model 1 is a model trained by a machine learning algorithm. In the process of training and prediction, Model 1 will fully consider a large number of scene features. That is, Model 1 can be trained based on the interactive operations detected within T days and the corresponding multiple scene features combined with the machine learning algorithm. In this way, Model 1 can realize the recognition and application recommendation for complex scenes.

Model 2 uses a simple statistical algorithm to learn and predict the recommended applications in a specific spatiotemporal scenario based on simple scenario features (e.g., time features and spatial features). That is, Model 2 can be trained based on the interactive operations detected within T days and the corresponding specific scenario features.

In this way, the electronic device can determine the recommended application 1 and the corresponding frequency of use weight (second weight) based on the real-time detected scene features and model 1. The electronic device can determine the recommended application 2 and the corresponding regular weight (first weight) based on the real-time detected scene features and model 2. Then, the weights corresponding to the recommended application 1 and the recommended application 2 are fused, and a list of applications actually recommended is output, including the recommended application 1, the recommended application 2 and the fused weight values. In short, as a supplement to model 1, after model 2 is configured in the electronic device, in specific scenarios, the exposure of long-tail applications on the recommendation controls can be increased, thereby improving the accuracy of long-tail application recommendations.

For example, when the electronic device is located in the first location area (spatial feature) and the system time belongs to the first time period (time feature), that is, when the electronic device detects the first location information, it can be determined that the electronic device is located in the first location area. It can be understood that the first location information includes a service set identifier (SSID) provided by the first network (the WiFi network currently accessed by the electronic device), a cell identifier (cellID) provided by the first base station (the base station currently accessed by the electronic device), and first longitude and latitude information (longitude and latitude information detected at the current location) or a combination thereof, wherein the signal coverage area of the first network and the first base station belongs to the first location area.

When the electronic device detects that the system time point is the first time (a time point within the first time period), it can be determined that the system time of the electronic device belongs to the first time period. At this time, using model 2, the determined recommended application 2 includes a first application, and the first application can be an application in the recommended application 2 that can be displayed in the recommendation control. In the spatiotemporal scenario indicated by the first location area and the first time period, if the first weights of the installed applications are arranged in order from large to small, the first weight of the first application is arranged before the first place (the top k), and the first weight of the second application is arranged after the first place, that is, the first weight of the first application is higher than the first weight of the second application. In addition, in the current spatiotemporal scenario, the application icon of the second application is not displayed in the recommendation control.

When the electronic device is located in the first location area and the system time belongs to the first time period, the electronic device can also determine the recommended application 1 including the third application based on the multiple types of scene features detectable in the current scene. It can be understood that in the process of determining the recommended application 1, the scene features considered are not limited to the first location area (spatial features) and the first time period (time features), and may include more or fewer types of scene features. In addition, if the second weights of the installed applications are arranged in order from large to small, the second weight of the third application is arranged before the second place (top k), and the second weight of the fourth application is arranged after the second place, that is, the first weight of the third application is higher than the second weight of the fourth application. In addition, in the current scene, the application icon of the fourth application is not displayed in the recommendation control.

Of course, the first application and the third application may be the same application or different applications. For example, the first application may include the fifth application and the sixth application, and the third application may include the sixth application and the seventh application.

As described in the above embodiments, the machine learning algorithm used in the training model 1 may be: decision tree, LightGBM, logistic regression model, XGBoost and other algorithms.

Taking Model 1 as an example, which is a model trained by a decision tree algorithm, the above-mentioned Model 1 can be a decision tree model created based on the interactive operations collected within T days and the corresponding scene features. The principle of establishing a decision tree model can refer to relevant technologies and will not be repeated here.

For example, the obtained decision tree model can be shown as FIG10. In the decision tree model, the first node corresponds to the number of interactions corresponding to all applications. FIG10 uses a digital sequence to display the number of interactions of each application. Each node in FIG10 corresponds to a digital sequence. Each sequence bit in the digital sequence corresponds to an application. The value of each sequence bit is the number of interactions of the corresponding application. For example, the first sequence bit refers to the calendar application. As shown in FIG10, the value of the first sequence bit of the first node is 1, indicating that the number of interactions of the calendar application is 1.

In addition, the first node can be split into multiple child nodes based on a certain scene feature. For example, as shown in FIG10 , the second node and the third node are split based on whether there is a location tag indicating a home scene. The second node corresponds to the number of interactions corresponding to each application in the home scene. The third node corresponds to the number of interactions corresponding to each application in the non-home scene.

Similarly, the second node and the third node can also be split according to other scene features. For example, as shown in Figure 10, the second node is split according to whether the time information is Monday to Thursday to obtain multiple child nodes, namely, the fourth node and the fifth node. Among them, the fourth node corresponds to the home scene, and the time information is the number of interactions corresponding to each application from Monday to Thursday. The fifth node corresponds to the home scene, and the time information is the number of interactions corresponding to each application from Friday to Sunday.

In addition, the fourth node can also be split into the sixth node and the seventh node. The sixth node corresponds to the home scenario, and the time information is the number of interactions corresponding to each application from Monday to Wednesday. The seventh node corresponds to the home scenario, and the time information is Thursday, and the number of interactions corresponding to each application.

The fifth node can also be split into the eighth and ninth nodes. The eighth node corresponds to the home scene, the time information is from 0:00 to 11:00 from Friday to Sunday, and the number of interactions corresponding to each application. The ninth node corresponds to the home scene, the time information is from 12:00 to 23:00 from Friday to Sunday, and the number of interactions corresponding to each application.

In addition, the third node can be split into the tenth node and the eleventh node. The tenth node corresponds to the number of interactions of each application in the company scenario. The eleventh node corresponds to the number of interactions of each application in the scenarios of not being at home and not being in the company.

The tenth node can also be split into the twelfth node and the thirteenth node. The twelfth node is in the company scenario, and the time information is the number of interactions corresponding to each application from Monday to Friday. The thirteenth node is in the company scenario, and the time information is the number of interactions corresponding to each application from Saturday to Sunday.

The eleventh node can also be split into the fourteenth node and the fifteenth node. The fourteenth node corresponds to the number of interactions of each application in the cafeteria scenario. The fifteenth node corresponds to the number of interactions of each application in the scenarios of not at home, not at work, and not in the cafeteria.

In this way, the electronic device can detect the scene features corresponding to the current scene in real time, and search for matching nodes from model 1 according to the detected scene features. Then, according to the number of interactions of each application corresponding to the matching node, the application with the highest usage popularity is evaluated as the application adapted to the current scene, that is, recommended application 1.

For example, the electronic device detects that the scene features corresponding to the current scene include the location tag of the home scene and the time information is Thursday. Then, according to the decision tree model shown in Figure 10, it can be determined that the current scene matches the seventh node. Then, according to the number of interactions of each application in the seventh node, the applications with the highest usage popularity are evaluated.

In some examples, the process of evaluating the top applications in terms of usage popularity may be: normalizing the number of interactions of each application in the matching node to obtain the usage frequency weight of each application, which may indicate the frequency of interaction between the user and the application in the scenario represented by the multiple scenario features. Among them, the higher the usage frequency weight, the higher the usage popularity of the corresponding application. In this case, the electronic device may determine the application whose usage frequency weight is greater than the preset weight threshold 1 as the top application in the current scenario. Alternatively, in this case, the electronic device may sort the applications according to the usage frequency weight, and determine the applications ranked in the top k as the top applications. Among them, k is a positive integer.

Taking model 1 as an example, the model is trained by algorithms such as LightGBM, logistic regression or XGBoost. The above model 1 can be a model obtained after training based on data such as interactive operations and corresponding scene features collected within T days. In this way, model 1 can output multiple applications and usage probabilities based on the scene features collected by the electronic device in real time. The usage probability is the probability predicted by model 1 that the application is suitable for the current scene. Then, the applications ranked before the top k in usage probability can be regarded as the most popular applications.

In some embodiments, the above model 2 may be a regular weight indicating each application program in each scenario. For example, as shown in the following Table 3:

table 3

Among them, the scene can be represented by a combination of spatial features and time features. For example, in Table 3, the combination of location tag 1 and time period 9 indicates scene 1, and the combination of location tag 2 and time period 10 indicates scene 2.

In addition, a day can be quantified into multiple time periods in advance, for example, quantified into 24 time periods, such as time period 0, time period 1...time period 24. Among them, time period 0 indicates 0 o'clock to 1 o'clock, time period 1 indicates 1 o'clock to 2 o'clock, and so on, time period 9 indicates 9 o'clock to 10 o'clock, time period 10 indicates 10 o'clock to 11 o'clock, and so on, time period 23 indicates 23 o'clock to 0 o'clock. It can be understood that if the electronic device appears in the geographic space indicated by location tag 1 within time period 9 (that is, 9 o'clock to 10 o'clock), then the electronic device can detect scene 1. If the electronic device appears in the geographic space indicated by location tag 1 within time period 10 (that is, 10 o'clock to 11 o'clock), then the electronic device can detect scene 2.

As shown in Table 3, the electronic device detected scene 1 in 22 days, so the number of days scene 1 appeared is 22 days. In these 22 days, the electronic device detected that the user used application a in scene 1 for 17 days, and the regular weight corresponding to application a was 0.77. Among them, the regular weight can be the ratio between the number of days application a is used and the number of days scene 1 appears, which can indicate the probability of enabling the application in a specific spatiotemporal scenario represented by the spatiotemporal features. In addition, in these 22 days, the electronic device detected that the user used application b in scene 1 for 11 days, and the regular weight corresponding to application b was 0.5. The same applies to the others and will not be repeated again.

In some embodiments, the usage days in different scenarios in Table 3 are obtained by counting the interactive operations detected within T days.

An example is given to determine the number of days application a is used in scenario 1 (location tag 1 and time period 9) and scenario 2 (location tag 1 and time period 10).

The first method: First, obtain training data 1 and training data 2. Among them, training data 1 and training data 2 are both interactive operations for application a, the difference is that the spatiotemporal scenarios in which training data 1 and training data 2 are collected are different. Training data 1 includes interactive operation c with location tag 1 and moment information 1. Training data 2 includes interactive operation e with location tag 1 and moment information 2. The above-mentioned moment information 1 refers to any time point belonging to time period 9, and the above-mentioned moment information 2 refers to any time point belonging to time period 10. As shown in the above embodiment, a day can be quantified into 24 time periods in advance, the above-mentioned time period 9 is a time period indicating 9 o'clock to 10 o'clock, and the above-mentioned time period 10 is a time period indicating 10 o'clock to 11 o'clock. In this way, there is no overlapping part between training data 1 and training data 2. Then, according to the detection date of interactive operation c in training data 1, the number of different dates corresponding to interactive operation c is counted as the number of days of use of application a under scenario 1 (location tag 1 and time period 9). According to the detection date of the interaction operation e in the training data 2, the number of different dates corresponding to the interaction operation e is counted as the number of days the application a is used in the scenario 2 (location tag 1 and time period 10).

The second method: first obtain training data 3 and training data 4. Among them, training data 3 and training data 4 are both interactive operations for application a, and the difference is that the spatiotemporal scenes of training data 3 and training data 4 are different. Training data 3 includes interactive operation w with location tag 1 and time information 3. Training data 4 includes interactive operation q with location tag 1 and time information 4. The above-mentioned time information 3 is any time point belonging to the sliding window corresponding to time period 9, and the above-mentioned time information 4 refers to any time point belonging to the sliding window corresponding to time period 10. As shown in the above embodiment, a day can be quantified into 24 time periods in advance. The above-mentioned time period 9 is a time period indicating 9 o'clock to 10 o'clock. When the sliding window length is 15 minutes, the sliding window corresponding to the above-mentioned time period 9 is 8:45 to 10:15. The above-mentioned time period 10 is a time period indicating 10 o'clock to 11 o'clock. When the sliding window length is 15 minutes, the sliding window corresponding to the above-mentioned time period 10 is 9:45 to 11:15. That is, there is an overlapping part between training data 3 and training data 4. It can be understood that the above sliding window duration of 15 minutes is only an example, and the embodiment of the present application does not specifically limit the sliding window duration.

Then, according to the detection date of interaction operation w in training data 3, the number of different dates corresponding to interaction operation w is counted as the number of days application a is used in scenario 1 (position tag 1 and time period 9). According to the detection date of interaction operation q in training data 4, the number of different dates corresponding to interaction operation q is counted as the number of days application a is used in scenario 2 (position tag 1 and time period 10).

In summary, both the first method and the second method can obtain the number of days application a is used in scenario 1 (the scenario indicated by location tag 1 and time period 9), as well as the number of days application a is used in scenario 2 (the scenario indicated by location tag 1 and time period 10). The embodiments of the present application can also adopt other methods to determine the number of days application a is used in different scenarios, and there is no specific limitation on this.

It is understandable that the usage days of other applications in various scenarios can also be obtained in the above manner, which will not be repeated here. For example, the first day number (application usage days) and the second day number (scenario appearance days) corresponding to the first application.

In short, after creating model 2, the electronic device can detect the spatial features and time features corresponding to the current scene in real time. According to the detected spatial features (such as location tags) and time features (such as time information), query the matching scene in model 2. When the matching scene is queried, obtain the applications used in the scene and the corresponding regular weights. Then, the applications whose regular weights are greater than the preset weight threshold 2 can be used as the applications with regular usage in the current scene, that is, recommended application 2.

It can be understood that whether model 1 is used to obtain recommended application 1 with high usage popularity, or model 2 is used to obtain recommended application 2 with regular usage, in essence, the electronic device uses different recommendation algorithms (model 1 and model 2) to evaluate the application programs that need to be recommended to the user from different angles. In the embodiment of the present application, other recommendation algorithms can also be used to evaluate the application programs that need to be recommended to the user from other angles, such as recommended application 3.

For example, other recommendation algorithms may be application jump prediction models, which can predict applications that have a jump relationship with the foreground application and use them as corresponding recommended applications. Usually, when an application is running in the foreground of an electronic device, the application jump prediction model can be enabled to predict the corresponding recommended application and display it on the application interface of the foreground application. The specific implementation process can refer to relevant technologies and will not be repeated here.

It is understandable that the same application may exist in the recommended applications determined by different recommendation algorithms. In addition, in the process of screening recommended applications, various recommendation algorithms will assign corresponding weights to the recommended applications, and the weights can represent the degree to which the electronic device recommends the application. Of course, different recommendation algorithms assign different types of weights. For example, recommended application 1 corresponds to a frequency of use weight, recommended application 2 corresponds to a regularity weight, and recommended application 3 corresponds to other weights. If the same application has multiple types of weight values, that is, in a scenario where multiple types of recommendation algorithms determine that the application is a recommended application, the multiple types of weight values of the recommended application can be fused. This process can also be called the fusion of multiple types of recommended applications.

As an implementation method, the way to fuse multiple types of weight values is to obtain the recommendation factor of each type of recommendation algorithm, such as the first recommendation model corresponds to the first recommendation factor, and the second recommendation model corresponds to the second recommendation factor. Among them, the recommendation factor can indicate the recommendation accuracy of the corresponding recommendation algorithm for the application. For example, the higher the proportion of recommended applications predicted by the recommendation algorithm that are selected by users, the larger the recommendation factor corresponding to the recommendation algorithm; the lower the proportion of recommended applications predicted by the recommendation algorithm that are selected by users, the smaller the recommendation factor corresponding to the recommendation algorithm. In some embodiments, historical detection data can be used to evaluate the recommendation factors corresponding to various recommendation algorithms. The specific process can refer to relevant technologies and will not be repeated here. The above-mentioned historical detection data may include interaction data detected by the electronic device within a specified number of days, wherein the interaction data includes detected interaction operations and corresponding scene labels.

Take determining the recommendation factor of model 1 as an example. For the convenience of description, t is used to refer to the current day, t-1 refers to the day before t, t-2 refers to the day before t-1, and so on, t-T represents the day before t-T+1. As shown in Figure 11, the electronic device obtains the interaction data detected within T-1 days before t-1, that is, the interaction data detected by the electronic device within t-2, t-3, ... t-T. Then, using the interaction data detected within the above T-1 days, according to the method provided in the above embodiment, model 1 is created. Afterwards, the electronic device uses model 1 to predict the recommended applications to the user based on the various scenarios detected by the electronic device in t-1. Then, the application programs actually enabled by the user detected by the electronic device in various scenarios of t-1 are obtained. Then, based on the predicted recommended applications and the actually enabled applications in each scenario, the recommendation accuracy of model 1 is calculated as the recommendation factor of model 1. The recommendation factor calculation method of other recommendation algorithms is similar, and will not be repeated here. Of course, based on the interaction data detected within T-1 days, the created model 1 is only used to test the recommendation factor corresponding to model 1. The model 1 actually used in the electronic device can be a model trained based on the interaction data detected within T days (that is, within t-1, t-2, ..., t-T).

After determining the recommendation factor of each type of recommendation algorithm, the electronic device can calculate the product of the weight assigned to each recommended application by each type of recommendation algorithm and the recommendation factor as the recommendation weight of the recommended application. When a recommended application is recommended by multiple types of recommendation algorithms, the recommended application corresponds to multiple recommendation weights. In this scenario, the multiple recommendation weights are superimposed as the final recommendation weight of the recommended application.

For example, the recommended applications predicted by the electronic device using model 1 and model 2 include WeChat ^TM , WeLink ^TM and Zhihu ^TM . Model 1 assigns a usage frequency weight of 0.2 to WeChat ^TM , a usage frequency weight of 0.3 to WeLink ^TM , and a usage frequency weight of 0.1 to Zhihu ^TM . Model 2 assigns a regularity weight of 0.6 to WeChat ^TM , a regularity weight of 0.2 to WeLink ^TM , and a regularity weight of 0.3 to Zhihu ^TM . In addition, the electronic device determines that the recommendation factor corresponding to model 1 is 0.98, and the recommendation factor corresponding to model 2 is 0.95. Then the final recommendation weight corresponding to WeChat ^TM is: 0.2*0.98+0.6*0.95=0.76, the final recommendation weight corresponding to WeLink ^TM is: 0.3*0.98+0.2*0.95=0.48, and the final recommendation weight corresponding to Zhihu ^TM is: 0.1*0.98+0.3*0.95=0.38. In this way, after the fusion of weights, the electronic device determines that the recommendation level for WeChat ^TM is greater than that for WeLink ^TM , and the recommendation level for WeLink ^TM is greater than that for Zhihu ^TM .

For another example, the recommended applications determined by model 1 also include office applications, and in a scenario where the recommended applications determined by other recommendation algorithms do not include the office applications, if the frequency of use weight assigned to the office applications by model 1 is 0.3, then the final recommendation weight of the office applications is: 0.3*0.98=0.29. In this way, when evaluating the recommendation degree for the application program according to the recommendation weight, it can be determined that the recommendation degree for WeChat ^TM is greater than that for WeLink ^TM , and the recommendation degree for WeLink ^TM is greater than that for office applications, and the recommendation degree for office applications is greater than that for Zhihu ^TM .

In some embodiments, the electronic device may arrange the application icons of the recommended applications from left to right on the recommendation control in the order of the final recommendation weight from large to small. That is, the recommended application with a larger final recommendation weight will be arranged on the left, and the recommended application with a smaller final recommendation weight will be arranged on the right.

In other embodiments, the application icons on the recommendation control can also be arranged in disorder. For example, the first time it is determined that the recommended applications include WeChat ^TM and Zhihu ^TM , and the electronic device arranges the application icons of WeChat ^TM and Zhihu ^TM on the recommendation control according to the recommendation weight. As the scene in which the electronic device is located changes, the recommended applications determined for the second time include WeChat ^TM and WeLink ^TM , that is, compared with the recommended applications determined last time (first time), there are the same applications and different applications, such as Zhihu ^TM is missing, but WeLink ^TM is newly added, of course, both include WeChat ^TM . In this scenario, the application icon of Zhihu ^TM can be canceled from the recommendation control, and then the application icon of WeLink ^TM is displayed at the display position of the original application icon of Zhihu ^TM . Afterwards, as the scene in which the electronic device is located changes, the recommended applications determined for the third time include Zhihu ^TM and office applications, that is, completely different from the recommended applications determined last time (second time), in this scenario, the electronic device can cancel the display of the application icons of WeChat ^TM and WeLink ^TM on the recommendation control, and at the same time, arrange the application icons of Zhihu ^TM and office applications on the recommendation control according to the recommendation weight.

In summary, the method provided by the embodiment of the present application can effectively improve the recommendation hit rate of long-tail applications. Among them, long-tail applications may include applications with low usage popularity but regular usage, that is, applications that are often enabled in specific time and space scenarios but have low usage popularity, such as attendance applications, access control applications, smart home applications, public transportation applications, etc.

In other possible embodiments, the electronic device can also recommend application function shortcut entries to users through recommendation controls. The process of determining the function shortcut entries to be recommended is similar to the process of determining the recommended application in the aforementioned embodiment and will not be repeated here.

The embodiment of the present application also provides an electronic device, which may include: a memory and one or more processors. The memory is coupled to the processor. The memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device can execute the various steps executed by the mobile phone in the above embodiment. Of course, the electronic device includes but is not limited to the above memory and one or more processors.

The embodiment of the present application also provides a chip system, which can be applied to the terminal device in the aforementioned embodiment. As shown in Figure 12, the chip system includes at least one processor 2201 and at least one interface circuit 2202. The processor 2201 can be the processor in the above-mentioned electronic device. The processor 2201 and the interface circuit 2202 can be interconnected through a line. The processor 2201 can receive and execute computer instructions from the memory of the above-mentioned electronic device through the interface circuit 2202. When the computer instructions are executed by the processor 2201, the electronic device can execute the various steps executed by the mobile phone in the above-mentioned embodiment. Of course, the chip system can also include other discrete devices, which are not specifically limited in the embodiment of the present application.

In some embodiments, through the description of the above implementation methods, technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

Each functional unit in each embodiment of the present application can be integrated into a processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as flash memory, mobile hard disk, read-only memory, random access memory, disk or optical disk.

The above is only a specific implementation of the embodiment of the present application, but the protection scope of the embodiment of the present application is not limited thereto, and any changes or substitutions within the technical scope disclosed in the embodiment of the present application should be included in the protection scope of the embodiment of the present application. Therefore, the protection scope of the embodiment of the present application should be based on the protection scope of the claims.

Claims (12)

1. An application recommendation method, applied to an electronic device, comprising:

The electronic equipment displays a first interface, wherein the first interface comprises a first control, and the first control is used for displaying application icons of application programs to be recommended to a user;

When the electronic equipment is located in a first position area and the system time belongs to a first period, the electronic equipment displays an application icon of a first application in the first control;

The first weight of the first application is higher than the first weight of the second application, an application icon of the second application is not displayed on the first control, the first weight is used for indicating the enabling probability of the application program in the scene of being located in the first position area, and the system time belongs to the first period.

2. The method according to claim 1, wherein the method further comprises:

When the electronic equipment is located in the first position area and the system time belongs to a first period, the electronic equipment displays an application icon of a third application in the first control; the second weight of the third application is higher than the second weight of the fourth application, the application icon of the fourth application is not displayed on the first control, and the second weight is used for indicating interaction frequency between the user and the application program in the current scene.

3. The method of claim 2, wherein the electronic device includes a first recommendation model and a second recommendation model, the first recommendation model is used to determine the second weights corresponding to the applications in different scenarios, the second recommendation model is used to determine the first weights corresponding to the applications in different scenarios, and the electronic device is further configured to, before displaying the first application and the third application in the first control:

The electronic equipment acquires first time and first position information;

Determining, by the electronic device, the first application using the second recommendation model, in a case where the first time belongs to the first period and the first location information indicates the first location area, the first application being an application in which the first weights are rearranged before a first ranking when the first weights of the applications are arranged in order from large to small;

The electronic device determines the third application by using the first recommendation model, wherein the third application is an application program with the second weights rearranged before the second ranking when the second weights of the application programs are arranged in the order from the big to the small.

4. A method according to claim 3, wherein the first location information comprises one or a combination of a service set identity provided by a first network, a cell identity provided by a first base station and first latitude and longitude information, wherein the signal coverage areas of the first network and the first base station belong to the first location area.

5. The method of claim 3, wherein prior to the electronic device displaying a first application in the first control, the method further comprises:

The electronic equipment acquires a first number of days and a second number of days, wherein the first number of days is the number of days when the first application is started in a scene that the electronic equipment is located in the first position area and the system time belongs to a first period, and the second number of days is the number of days when the electronic equipment detects that the electronic equipment is located in the first position area and the system time belongs to the first period;

the electronic device determines a first weight corresponding to the first application when the electronic device is located in the first location area and the system time belongs to the first period according to the first number of days and the second number of days;

and the electronic equipment updates the first recommendation model according to the first application and the corresponding first weight.

6. A method according to claim 3, characterized in that the method further comprises:

The electronic equipment acquires interaction data, wherein the interaction data comprises the interaction times of users and various application programs under the condition that different position information, different time information, different user state information and different equipment state information are detected;

And the electronic equipment clusters according to the position information, the time information, the user state information and the equipment state information in the interaction data to obtain the second recommendation model.

7. The method of claim 6, wherein the electronic device obtaining interaction data comprises:

The electronic equipment collects the position information, the time information, the user state information and the equipment state information when detecting a first operation aiming at any application program;

Under the condition that a first service set identifier and a first cell identifier are acquired, determining the first service set identifier as the detected position information, and associating the first cell identifier with the first service set identifier;

And under the condition that the first cell identity is acquired and the first cell identity is already associated with the first service set identity, determining the first service set identity as the detected position information.

8. The method of claim 7, wherein the electronic device determines the location information as a first identification indicating unidentified geospatial if a service set identification and a cell identity are not acquired upon detection of the first operation.

9. A method according to claim 3, wherein the first application comprises a fifth application and a sixth application, and wherein in the case where the third application comprises a sixth application and a seventh application, the method further comprises:

The electronic equipment determines the recommendation weight of the sixth application according to the first weight, the second weight, the first recommendation factor and the second recommendation factor corresponding to the sixth application, wherein the first recommendation factor is used for indicating the recommendation hit rate of the first recommendation model, and the second recommendation factor is used for indicating the recommendation hit rate of the second recommendation model;

The electronic equipment determines the recommendation weight of the sixth application according to the first weight and the first recommendation factor corresponding to the fifth application;

and the electronic equipment determines the recommendation weight of the seventh application according to the second weight and the second recommendation factor corresponding to the seventh application.

10. The method of claim 9, wherein application icons of the fifth application, the sixth application, and the seventh application are arranged on the first control in an order of the corresponding recommendation weights from high to low.

11. An electronic device comprising one or more processors and memory; the memory being coupled to a processor, the memory being for storing computer program code comprising computer instructions which, when executed by one or more processors, are for performing the method of any of claims 1-10.

12. A computer storage medium comprising computer instructions which, when run on an electronic device, cause the electronic device to perform the method of any of claims 1-10.