Usage parameter adjusting method, model training method and electronic equipment

Abstract

The invention provides a use parameter adjusting method, a model training method and electronic equipment, and relates to the technical field of artificial intelligence. And when the electronic equipment receives a first trigger operation of the electronic equipment, the electronic equipment responds to the first trigger operation, executes a target operation, and inputs the current feature data into the parameter preference model corresponding to the first application scene to obtain a predicted value of each use parameter in at least one use parameter corresponding to the first application scene. Afterwards, for each usage parameter, the electronic device can adjust the parameter value of the usage parameter into the predicted value of the usage parameter, thereby realizing timely adjustment of the usage parameter in different application scenarios, improving accuracy of determination of the usage parameter, satisfying usage parameter requirements of a user in different application scenarios, and improving usage experience of the user.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method for adjusting usage parameters, a model training method, and an electronic device. Background Art

[0002] With the development of Internet technology, more and more smart devices are beginning to appear in people's lives. When using smart devices (such as mobile phones), users can adjust the device's usage parameters (such as display brightness, volume, color temperature, etc.) according to their own needs, so that the smart device can meet the user's personal preferences.

[0003] However, currently the usage parameters of smart devices are usually manually triggered and adjusted, resulting in low adjustment efficiency of the usage parameters and low user experience. Summary of the invention

[0004] The embodiments of the present application provide a method for adjusting usage parameters, a model training method and an electronic device, which are used to improve the adjustment efficiency of usage parameters and enhance the user experience.

[0005] To achieve the above objectives, the embodiments of the present application adopt the following technical solutions:

[0006] In a first aspect, a method for adjusting a usage parameter is provided, which can be applied to an electronic device, the electronic device including a parameter preference model corresponding to a plurality of application scenarios, the parameter preference model being used to predict a predicted value of the usage parameter in the corresponding application scenario. In the method, when the electronic device receives a first trigger operation of the electronic device, in response to the first trigger operation, performs a target operation, inputs current feature data into the parameter preference model corresponding to the first application scenario, and obtains a predicted value of each usage parameter in at least one usage parameter corresponding to the first application scenario; wherein the first trigger operation is used to trigger the electronic device to perform the target operation, the target operation including at least one of displaying a first interface, starting or entering a first mode. Afterwards, for each usage parameter, the electronic device can adjust the parameter value of the usage parameter to the predicted value of the usage parameter.

[0007] In the present application, since the target parameter preference model is obtained by training the initial model according to the training feature set, that is to say, the target parameter preference model is used to reflect the user's usage habits (or parameter adjustment habits), that is, the predicted parameter values ​​corresponding to the usage parameters obtained by predicting the target parameter preference model can match the user's usage habits and be more in line with the setting requirements of the target user, thereby realizing timely adjustment of usage parameters in different application scenarios, improving the accuracy of parameter value determination, and eliminating the need for users to manually adjust parameters, thereby improving the user's usage experience.

[0008] In a possible implementation of the first aspect, when the target operation includes displaying the first interface, the process in which the electronic device inputs the current feature data into the parameter preference model corresponding to the first application scenario may specifically include: the electronic device determines whether the application scenario has changed. Then, when the application scenario has changed, the electronic device obtains the parameter preference model corresponding to the first application scenario. Then, the electronic device inputs the current feature data into the parameter preference model corresponding to the first application scenario.

[0009] In the present application, the electronic device only performs parameter prediction when the application scenario changes. In this way, not only can the parameter values ​​in different application scenarios be adjusted in a timely manner to meet the user's parameter usage requirements in different application scenarios, but also unnecessary resource waste caused by parameter value prediction for the same application scenario can be reduced, thereby reducing unnecessary power consumption losses.

[0010] In a possible implementation of the first aspect, the above method also includes: when a new parameter setting result is obtained, the electronic device can update the parameter preference model corresponding to the usage parameter corresponding to the new feature data based on the new feature data in the new parameter setting result and the setting value of the usage parameter corresponding to the new feature data.

[0011] In the present application, after obtaining the parameter preference model, the electronic device can update the corresponding parameter preference model according to the usage parameters manually adjusted by the user during daily use of the mobile phone, that is, using the new feature data and the setting values ​​of the usage parameters corresponding to the new feature data. In this way, the updated parameter preference model can better learn the user's usage parameter setting habits, ensure that the predicted values ​​obtained by the updated parameter preference model are more in line with the user's usage habits, thereby improving the accuracy of the usage parameter prediction.

[0012] In a possible implementation manner of the first aspect, the new usage parameter value includes the predicted value and / or a usage parameter value manually adjusted by the user.

[0013] In a possible implementation of the first aspect, the process of updating the parameter preference model may specifically include: the electronic device sets a first weight for the new feature data, wherein the first weight belongs to a preset weight range. Thereafter, the electronic device updates the parameter preference model corresponding to the usage parameter corresponding to the new feature data according to the new feature data with the first weight and the setting value of the usage parameter corresponding to the new feature data.

[0014] In the present application, since each new feature data is set with a corresponding first weight, it means that the impact of each feature data on the model training is different, that is, the electronic device can update the parameter preference model in a targeted manner according to the first weight, so that the updated parameter preference model can better learn the user's usage parameter setting habits, thereby improving the accuracy of parameter prediction.

[0015] In a possible implementation of the first aspect, when the number of new feature data includes multiple ones, the above-mentioned first weight is related to the setting time of the usage parameters corresponding to the new feature data, and the earlier the setting time of the usage parameters corresponding to the new feature data, the smaller the weight of the new feature data.

[0016] In the present application, if the setting time of the usage parameters corresponding to the new feature data is earlier, it means that the setting time of the usage parameters is longer, and the probability that the user preference may change is greater. Therefore, the electronic device can set the weight of the new feature data to be smaller; if the setting time of the usage parameters corresponding to the new feature data is later, it means that the setting time of the usage parameters is shorter, and the probability that the user preference may change is smaller. Therefore, the electronic device can set the weight of the new feature data to be larger. In this way, the parameter preference model can quickly learn the user's recent preferences, so that the updated parameter preference model can better predict the usage parameter values ​​and improve the accuracy of usage parameter prediction.

[0017] In a possible implementation of the first aspect, the above-mentioned first weight is related to the determined object corresponding to the setting value of the usage parameter corresponding to the feature data. If the setting value of the usage parameter corresponding to the new feature data is determined in response to an adjustment operation input by a user, the first weight of the new feature data is a first preset weight; if the setting value of the usage parameter corresponding to the new feature data is determined based on a parameter preference model, the first weight of the new feature data is a second preset weight; wherein the second preset weight is smaller than the first preset weight.

[0018] In the present application, if the setting value of the usage parameter corresponding to the new feature data is determined in response to the adjustment operation input by the user, it means that the setting value is set by the user according to his own needs, therefore, the electronic device can set the first weight higher, that is, set it to the first preset weight; if the setting value of the usage parameter corresponding to the new feature data is determined based on the parameter preference model, it means that the setting value is predicted by the electronic device according to the user's usage habits, and has a certain error, therefore, the electronic device can set the first weight of the new feature data lower, that is, set it to the second preset weight. In this way, the updated parameter preference model can better predict the usage parameter value and improve the accuracy of the usage parameter prediction.

[0019] In a possible implementation of the first aspect, the process of the electronic device inputting the current feature data into the parameter preference model corresponding to the first application scenario may specifically include: the electronic device determines whether there is a parameter setting result in the parameter setting results in the pre-stored position in which the first feature data is the same as the current feature data; if there is no parameter setting result in the parameter setting results in the pre-stored position in which the first feature data is the same as the current feature data, the electronic device may input the current feature data into the parameter preference model corresponding to the first application scenario, wherein the parameter setting result includes a historical setting value of the usage parameter and the first feature data corresponding to the historical setting value.

[0020] In the present application, if there is no parameter setting result in the pre-stored position in which the first characteristic data is the same as the current characteristic data, it indicates that the electronic device has not set the usage parameters in the environment corresponding to the current characteristic data. Therefore, in order to accurately determine the usage parameter value, the current characteristic data can be input into the parameter preference model corresponding to the first application scenario to predict the usage parameter value, so that the predicted parameter value is more in line with the user's preference, which not only saves the user's time in manually setting the parameters, but also improves the user's usage experience.

[0021] In a possible implementation of the first aspect, the above method also includes: when there is a parameter setting result in the parameter setting result in the pre-stored position in which the first feature data is the same as the current feature data, the electronic device can determine the historical setting value of the usage parameter in the parameter setting result as the predicted value corresponding to the usage parameter.

[0022] In the present application, if there is a parameter setting result in the pre-stored position in which the first characteristic data is the same as the current characteristic data, it indicates that the electronic device has set the usage parameter under the environment corresponding to the current characteristic data. Therefore, the electronic device can directly use the setting value of the usage parameter corresponding to the current characteristic data as the predicted value corresponding to the usage parameter, thereby avoiding unnecessary prediction of the usage parameter.

[0023] In a possible implementation of the first aspect, the usage parameters corresponding to the application scenario include brightness, volume and color temperature; when the usage parameter is brightness, the above-mentioned current feature data include user age, user city, user occupation, user education level, average daily usage time, application scenario type and current ambient brightness; or, when the usage parameter is volume, the above-mentioned current feature data include user age, user city, user occupation, user education level, average daily usage time, application scenario type and current ambient noise; or, when the usage parameter is color temperature, the above-mentioned current feature data include user age, user city, user occupation, user education level, average daily usage time, application scenario type, current ambient noise and media playback mode.

[0024] In this embodiment, the current feature data corresponding to different usage parameters are different, that is, the parameter prediction value obtained by predicting based on the current feature data can better meet the user's habit of adjusting the usage parameter, further improving the accuracy of the usage parameter prediction.

[0025] In a second aspect, a model training method is provided, which can be applied to an electronic device. In the method, the electronic device obtains a first training feature set. The first training feature set includes feature data of a user and usage parameter setting values ​​corresponding to the feature data, and the feature data includes a user portrait and / or user behavior data. Afterwards, the electronic device trains the initial model according to the first training feature set to obtain a parameter preference model corresponding to the application scenario.

[0026] In the present application, the initial model is trained according to the user portrait of the current user and the usage parameters manually adjusted by the current user, so that the parameter preference model can learn the parameter usage habits of the current user in different usage scenarios. That is to say, the trained parameter preference model can predict the usage parameters that need to be adjusted for electronic devices in different usage scenarios, so that the trained parameter preference model has strong generalization performance. In addition, since the parameter preference model corresponding to the application scenario is obtained by training on the terminal side, that is, the parameter preference model is an end-side model, when the end-side model is used for parameter prediction later, there is no need to request the cloud (or cloud service end), which speeds up the prediction of parameter values.

[0027] In a possible implementation of the second aspect, the above method can also be applied to a cloud service. In the method, the cloud service obtains a first training feature set. The first training feature set includes feature data of multiple users and usage parameter setting values ​​corresponding to the feature data, and the feature data includes user portraits and / or user behavior data. Afterwards, the cloud service trains the initial model according to the first training feature set to obtain a parameter preference model corresponding to the application scenario.

[0028] In the present application, a pre-constructed parameter preference model is trained based on the user portraits of multiple users and the usage parameters manually adjusted by multiple users, so that the parameter preference model can learn the parameter usage habits of multiple users in different usage scenarios. In other words, the trained parameter preference model can predict the usage parameters that need to be adjusted for electronic devices in different usage scenarios, so that the trained parameter preference model has a stronger generalization performance.

[0029] In a possible implementation of the second aspect, the above-mentioned process of training the initial model may specifically include: the cloud service screens the first training feature set based on the user portrait to obtain a first sub-training feature set corresponding to each user portrait. Afterwards, the cloud service obtains a second training feature set corresponding to various application scenarios from the first sub-training feature set. Afterwards, for each application scenario, the cloud service may obtain a third training feature set corresponding to various usage parameters corresponding to the application scenario from the second training feature set. Afterwards, for each usage parameter, the cloud service may train the initial model based on the third training feature set corresponding to the usage parameter to obtain a parameter preference model corresponding to the usage parameter.

[0030] In the present application, the cloud service screens from the first training feature set in sequence according to the type of user portrait, the type of application scenario, and the usage parameters, so that the third training feature set includes feature data corresponding to a usage parameter in the same application scenario corresponding to the same user portrait. Therefore, the parameter preference model trained by the third training feature set is more targeted and provides a basis for the subsequent accurate prediction of parameter values.

[0031] In a third aspect, the present application provides an electronic device, comprising a display screen, a memory and one or more processors; the display screen, the memory and the processor are coupled; the display screen is used to display an image generated by the processor, and the memory is used to store computer program code, wherein the computer program code comprises computer instructions; when the processor executes the computer instructions, the electronic device executes the method as described above.

[0032] In a fourth aspect, the present application provides a computer-readable storage medium, comprising computer instructions, which, when executed on an electronic device, enable the electronic device to execute the method described above.

[0033] In a fifth aspect, the present application provides a computer program product, which, when executed on an electronic device, enables the electronic device to execute the method described above.

[0034] In a sixth aspect, a chip is provided, comprising: an input interface, an output interface, a processor and a memory, wherein the input interface, the output interface, the processor and the memory are connected via an internal connection path, the processor is used to execute the code in the memory, and when the code is executed, the processor is used to execute the method as described above.

[0035] Among them, the beneficial effects that can be achieved by the model training method described in the second aspect provided above, the electronic device described in the third aspect provided above, the computer-readable storage medium described in the fourth aspect, the computer program product described in the fifth aspect, and the chip described in the sixth aspect can refer to the beneficial effects in the first aspect and any possible design method thereof, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A schematic diagram of a brightness adjustment interface provided in an embodiment of the present application;

[0037] Figure 2 A schematic diagram of an interface for enabling the automatic brightness adjustment function provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of an interface for adjusting brightness in a video scene provided in an embodiment of the present application;

[0039] Figure 4 A schematic diagram of an interface for adjusting volume in a video scene provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application;

[0041] Figure 6 A schematic diagram of the software structure of an electronic device provided in an embodiment of the present application;

[0042] Figure 7 A flowchart of a method for adjusting device parameters provided in an embodiment of the present application;

[0043] Figure 8 A schematic diagram of the structure of a neural network provided in an embodiment of the present application;

[0044] Fig. 9 A flowchart of a model training method provided in an embodiment of the present application;

[0045] Fig.10 A flow chart of a model updating method provided in an embodiment of the present application;

[0046] Fig.11 A schematic diagram of an interface for a user to return to the main interface provided in an embodiment of the present application;

[0047] Fig.12 A schematic diagram of an interface for a user to browse videos downwards provided in an embodiment of the present application;

[0048] Fig.13 A schematic diagram of an interface for a user to enter a game provided in an embodiment of the present application;

[0049] Fig.14 A schematic diagram of an interface for a user to click on a text message logo provided in an embodiment of the present application;

[0050] Fig.15 A schematic diagram of an interface for a user to exit a video screen provided in an embodiment of the present application;

[0051] Fig.16 A schematic diagram of an interface for displaying authorization information provided in an embodiment of the present application;

[0052] Fig.17 A flowchart of a method for adjusting parameters using an embodiment of the present application is provided. DETAILED DESCRIPTION

[0053] The technical scheme in the embodiment of the present application will be described below in conjunction with the accompanying drawings in the embodiment of the present application. Wherein, in the description of the present application, unless otherwise specified, the "and / or" in the present application is only a kind of association relationship describing the associated object, indicating that there can be three kinds of relationships, for example, A and / or B, which can be represented by: A exists alone, A and B exist at the same time, and B exists alone, wherein A, B can be singular or plural. And, in the description of the present application, unless otherwise specified, "multiple" refers to two or more than two. "At least one of the following (individuals)" or its similar expressions refers to any combination of these items, including any combination of single items (individuals) or plural items (individuals). For example, at least one of a, b, or c (individuals) can be represented by: a, b, c, ab, ac, bc, or abc, wherein a, b, c can be single or multiple. In addition, in order to facilitate the clear description of the technical scheme of the embodiment of the present application, in the embodiment of the present application, the words "first", "second" and the like are used to distinguish the same items or similar items with substantially the same functions and effects. Those skilled in the art will appreciate that words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not necessarily limit the difference. At the same time, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a concrete way for easy understanding.

[0054] In the related art, the usage parameters of smart devices (or electronic devices) (such as display brightness, volume, color temperature, etc.) are usually adjusted manually. If the application scenario switches (such as entering the video playback interface from the video list interface), the user needs to spend time to readjust the parameters, resulting in a low user experience.

[0055] For example, the smart device is a mobile phone, and the parameter used is a brightness parameter. Figure 1 As shown, a progress bar 11 is displayed on the interface, and the progress bar 11 is used to adjust the brightness value of the display screen of the mobile phone, that is, to adjust the brightness parameter. Specifically, in response to the user's dragging operation on the button 11A on the progress bar 11, the brightness value of the display screen in the mobile phone is adjusted so that the brightness value of the display screen is a brightness value suitable for the user to browse. It can be understood that the closer the button 11A is to the brightness mark 1B, the higher the brightness value of the display screen in the mobile phone; the closer the button 11A is to the brightness mark 1A, the lower the brightness value of the display screen in the mobile phone.

[0056] In some embodiments, in order to reduce the time for users to adjust parameters, the electronic device can automatically adjust the use parameters of the electronic device according to the environment in which the electronic device is currently located. Exemplarily, taking the use parameter as a brightness parameter as an example, the electronic device can automatically adjust the brightness value of the display screen according to the light source value of the current environment, combined with the mapping relationship between the light source value and the brightness value. Specifically, the electronic device can determine the target brightness value corresponding to the light source value of the current environment from the mapping relationship when the light source value of the current environment is obtained. After that, the electronic device can adjust the brightness value of the display screen to the target brightness value. Among them, ambient light refers to the natural light in the environment in which the electronic device is located. The ambient light can be sunlight, light, etc., without specific limitation.

[0057] It can be understood that the light source value of the current environment is determined based on the ambient light and the distance between the ambient light and the electronic device. Among them, the ambient light can be collected by a light sensor or a camera. The distance between the ambient light and the electronic device can be collected by a distance sensor. For example, under the same ambient light, if the distance between the ambient light and the electronic device is closer, it means that the brightness required when the user browses the interface is greater, so the electronic device can set the brightness value of the display screen to be higher; and if the distance between the ambient light and the electronic device is farther, it means that the brightness required when the user browses the interface is smaller, so the electronic device can set the brightness value of the display screen to be lower to reduce the power consumption loss of the electronic device. For another example, when the distance between the ambient light and the electronic device is the same, if the ambient light is stronger, it means that the current light is more sufficient, that is, the brightness required when the user browses the interface is larger, so the electronic device can set the brightness value of the display screen to be higher; and if the ambient light is weaker, it means that the current light is dim, that is, the brightness required when the user browses the interface is smaller, so the electronic device can set the brightness value of the display screen to be lower to reduce the power consumption loss of the electronic device.

[0058] For example, see Figure 2The brightness setting interface shown in the figure includes a switch control 1C for automatically adjusting the brightness. When the user wants the electronic device to automatically adjust the brightness according to the current ambient brightness, the user can turn on the switch control 1C. The electronic device responds to the user's turn-on operation and switches the switch control 1C to the on state (such as Figure 2 As shown in 1c in the figure), the automatic brightness adjustment function is turned on. After that, the mobile phone can automatically adjust the brightness value of the display screen of the electronic device according to the light source value of the current environment and the above mapping relationship.

[0059] However, the above-mentioned method of automatically adjusting the brightness of the display screen by ambient light can achieve different display screen brightness settings in different environments, but it cannot adjust the display screen brightness to match different types of people. For example, the eyes of people of different age groups have different sensitivities to brightness, that is, people of different age groups have different requirements for display screen brightness, such as younger people prefer brighter displays, and older people prefer darker displays. For another example, people of different occupations have different requirements for display screen brightness, such as people who often work in dim environments prefer darker displays.

[0060] In some embodiments, the electronic device can adjust the usage parameters of the electronic device according to the user's trigger operation for different usage scenarios (here or referred to as different application types). Specifically, the electronic device can adjust the usage parameters of the electronic device according to the user's sliding operation on the current interface. Among them, the sliding operation may include an upward sliding operation and a downward sliding operation. Exemplarily, if the user slides up on the current interface, it means that the user believes that the usage parameter value corresponding to the current interface is low, so the electronic device can increase the usage parameter value to meet the user's needs; if the user slides down on the current interface, it means that the user believes that the usage parameter value corresponding to the current interface is high, so the electronic device can lower the usage parameter value to meet the user's needs.

[0061] In one example, taking a video scene as an example, in response to a user sliding operation on the left side of the video playback interface, the mobile phone adjusts the brightness of the video playback interface. Figure 3 As shown, the user's finger slides upward on the left side of the video playback interface. Therefore, the electronic device can increase the brightness value of the video playback interface according to the sliding distance of the user's finger to meet the user's needs. It can be understood that the greater the distance the user's finger slides, the greater the adjustment amplitude of the interface brightness value; the smaller the distance the user's finger slides, the smaller the amplitude adjustment of the interface brightness value. Here, brightness is the above-mentioned usage parameter.

[0062] In another example, taking a video scene as an example, in response to a user's sliding operation on the right side of the video playback interface, the mobile phone adjusts the volume of the video playback. Figure 4 As shown, the user's finger slides upward on the right side of the video playback interface. Therefore, the electronic device can increase the volume value of the video playback according to the sliding distance of the user's finger to meet the user's needs. It can be understood that the greater the sliding distance of the user's finger, the greater the adjustment range of the video volume value; the smaller the sliding distance of the user's finger, the smaller the adjustment range of the video volume value. Here, the volume is the above-mentioned usage parameter.

[0063] However, although the above-mentioned method of adjusting usage parameters according to the scene can adjust different usage parameters according to the user's own needs in different scenes, when the electronic device enters the same scene again (such as the same application or the same application type), it usually cannot automatically adjust the current usage parameters according to the usage parameters adjusted by the user before. The user still needs to readjust the parameters according to his or her own needs, which wastes the user's adjustment time and makes the user's usage experience lower.

[0064] Therefore, in response to the above problems, the embodiment of the present application provides a method for adjusting the use parameters of an electronic device. In the method, in response to the user's trigger operation on the current interface, the electronic device determines whether the application scenario has changed. Afterwards, when the application scenario changes, the electronic device obtains the target parameter preference model corresponding to the current application scenario, wherein the target parameter preference model is obtained by training a pre-built parameter preference model (or initial model) based on a training feature set. Afterwards, the electronic device inputs the current feature data of the user into the target parameter preference model to obtain the predicted parameter value (or second predicted parameter value, or predicted value) corresponding to the use parameter, wherein the current feature data includes at least one of the user portrait, user behavior data, and current environmental parameters. Afterwards, when there is a difference between the predicted parameter value and the current use parameter value, the electronic device adjusts the parameter value corresponding to the use parameter from the current use parameter value to the predicted parameter value, realizes the timely adjustment of the use parameters in different application scenarios, improves the accuracy of the use parameter determination, meets the user's use parameter requirements in different application scenarios, and improves the user's use experience.

[0065] In the embodiment of the present application, since the target parameter preference model is obtained by training the initial model according to the training feature set, that is to say, the target parameter preference model is used to reflect the user's usage habits (or parameter adjustment habits), that is, the predicted parameter value corresponding to the usage parameter obtained by predicting the target parameter preference model can match the user's usage habits, can better meet the setting requirements of the target user, and improve the accuracy of parameter value determination, without the need for the target user to manually adjust the parameters, thereby improving the user's usage experience. In addition, in the present embodiment, the electronic device only performs parameter prediction when the application scenario changes. In this way, not only can the parameter values ​​in different application scenarios be adjusted in a timely manner to meet the user's parameter usage requirements in different application scenarios, but also the unnecessary waste of resources caused by parameter value prediction for the same application scenario can be reduced, thereby reducing unnecessary power consumption losses.

[0066] Exemplarily, the electronic device in the embodiments of the present application can be a mobile phone, a portable computer, a tablet computer, a laptop computer, a personal computer (PC), a smart home, a smart screen, a car computer, etc. The following embodiments do not impose any special restrictions on the specific form of the electronic device.

[0067] Take the above-mentioned electronic device as an example, which is a mobile phone. Figure 5 A schematic structural diagram of an electronic device 100 is shown.

[0068] The electronic device 100 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2, a mobile communication module 150, a wireless communication module 160, an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, an earphone interface 170D, a sensor module 180, a button 190, a motor 191, an indicator 192, cameras 1 to N 193, a display screen 194, and subscriber identification module (SIM) card interfaces 1 to N 195, etc. Among them, the sensor module 180 may include a light sensor 180A, a distance sensor 180B, a temperature sensor 180C, a humidity sensor 180D, a touch sensor E, etc.

[0069] It is to be understood that the structure illustrated in the embodiment of the present invention does not constitute a specific limitation on the electronic device 100. In other embodiments of the present application, the electronic device 100 may include more or fewer components than shown in the figure, or combine some components, or separate 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.

[0070] The processor 110 may include one or more processing units, for example, the processor 110 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.

[0071] The controller may be the nerve center and command center of the electronic device 100. The controller may generate an operation control signal according to the instruction operation code and the timing signal to complete the control of fetching and executing instructions.

[0072] The processor 110 may also be provided with a memory for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. The memory may store instructions or data that the processor 110 has just used or cyclically used. If the processor 110 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 110, and thus improves the efficiency of the system.

[0073] In some embodiments, the processor 110 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.

[0074] It is understandable that the interface connection relationship between the modules illustrated in the embodiment of the present invention is only a schematic illustration and does not constitute a structural limitation on the electronic device 100. In other embodiments of the present application, the electronic device 100 may also adopt different interface connection methods in the above embodiments, or a combination of multiple interface connection methods.

[0075] The charging management module 140 is used to receive charging input from the charger. The power management module 141 is used to connect the battery 142, the charging management module 140 and the processor 110. The power management module 141 receives input from the battery 142 and / or the charging management module 140, and provides power to the processor 110, the internal memory 121, the external memory, the display screen 194, the camera 193, and the wireless communication module 160. In other embodiments, the power management module 141 and the charging management module 140 can also be set in the same device.

[0076] The wireless communication function of the electronic device 100 can be implemented through the antenna 1, the antenna 2, the mobile communication module 150, the wireless communication module 160, the modem processor and the baseband processor.

[0077] Antenna 1 and antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in electronic device 100 can be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve the utilization of 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.

[0078] The mobile communication module 150 can provide wireless communication solutions including 2G / 3G / 4G / 5G etc. applied on the electronic device 100 .

[0079] The wireless communication module 160 can provide wireless communication solutions including wireless local area networks (WLAN) (such as wireless fidelity (Wi-Fi) networks), bluetooth (BT), global navigation satellite system (GNSS), frequency modulation (FM), near field communication (NFC), infrared (IR), etc., which are applied to the electronic device 100. The electronic device 100 implements the display function through a GPU, a display screen 194, and an application processor. The GPU is a microprocessor for image processing, which connects the display screen 194 and the application processor. The GPU is used to perform mathematical and geometric calculations for graphics rendering. The processor 110 may include one or more GPUs, which execute program instructions to generate or change display information.

[0080] The display screen 194 is used to display images, videos, etc. The display screen is a touch screen. In some embodiments, the electronic device 100 may include 1 or N display screens 194 , where N is a positive integer greater than 1.

[0081] The electronic device 100 can realize the shooting function through ISP, camera 193, video codec, GPU, display screen 194 and application processor.

[0082] ISP is used to process the data fed back by camera 193. For example, when taking a photo, the shutter is opened, and the light is transmitted to the camera photosensitive element through the lens. The light signal is converted into an electrical signal, and the camera photosensitive element transmits the electrical signal to ISP for processing and converts it into an image visible to the naked eye. ISP can also perform algorithm optimization on the noise, brightness, and skin color of the image. ISP can also optimize the exposure, color temperature and other parameters of the shooting scene. In some embodiments, ISP can be set in camera 193.

[0083] The camera 193 is used to capture still images or videos. The object generates an optical image through the lens and projects it onto the photosensitive element. The photosensitive element can be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS) phototransistor. The photosensitive element converts the optical signal into an electrical signal, and then passes the electrical signal to the ISP to be converted into a digital image signal. The ISP outputs the digital image signal to the DSP for processing. The DSP converts the digital image signal into an image signal in a standard RGB, YUV or other format. In some embodiments, the electronic device 100 may include 1 or N cameras 193, where N is a positive integer greater than 1.

[0084] The digital signal processor is used to process digital signals, and can process not only digital image signals but also other digital signals. For example, when the electronic device 100 is selecting a frequency point, the digital signal processor is used to perform Fourier transform on the frequency point energy.

[0085] Video codecs are used to compress or decompress digital videos. The electronic device 100 may support one or more video codecs. Thus, the electronic device 100 may play or record videos in a variety of coding formats, such as Moving Picture Experts Group (MPEG) 1, MPEG2, MPEG3, MPEG4, etc.

[0086] NPU is a neural network (NN) computing processor. By drawing on the structure of biological neural networks, such as the transmission mode between neurons in the human brain, it can quickly process input information and can also continuously self-learn. Through NPU, applications such as intelligent cognition of electronic device 100 can be realized, such as image recognition, face recognition, voice recognition, text understanding, etc.

[0087] In some embodiments, the NPU may refer to a target recognition model to process input information and obtain the user identity.

[0088] The external memory interface 120 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the electronic device 100. The external memory card communicates with the processor 110 through the external memory interface 120 to implement a data storage function, such as storing music, video and other files in the external memory card.

[0089] The internal memory 121 can be used to store computer executable program codes, which include instructions. The processor 110 executes various functional applications and data processing of the electronic device 100 by running the instructions stored in the internal memory 121. The internal memory 121 may include a program storage area and a data storage area. Among them, 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 100 (such as audio data, a phone book, etc.), etc. In addition, the internal memory 121 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.

[0090] The electronic device 100 can implement audio functions such as music playing and recording through the audio module 170, the speaker 170A, the receiver 170B, the microphone 170C, the headphone jack 170D, and the application processor.

[0091] The audio module 170 is used to convert digital audio information into analog audio signal output, and also used to convert analog audio input into digital audio signal. The audio module 170 can also be used to encode and decode audio signals.

[0092] The speaker 170A, also called a "horn", is used to convert audio electrical signals into sound signals.

[0093] The receiver 170B, also called a "earpiece", is used to convert audio electrical signals into sound signals. When the electronic device 100 receives a call or voice message, the voice can be received by placing the receiver 170B close to the human ear.

[0094] Microphone 170C, also called "microphone" or "microphone", is used to convert sound signals into electrical signals.

[0095] The earphone jack 170D is used to connect a wired earphone.

[0096] The light sensor 180A can be used to detect the light intensity of the environment in which the electronic device 100 is located. In some embodiments, the electronic device can determine the light brightness of the environment in which the electronic device 100 is located through the light sensor 180A, and automatically adjust the brightness of the display screen of the electronic device according to the light brightness. For example, if the electronic device is in a dim environment, the brightness of the display screen of the electronic device will automatically dim, that is, reduce the brightness value corresponding to the display screen. For another example, if the electronic device is in a bright environment, the brightness of the display screen of the electronic device will automatically brighten, that is, increase the brightness value corresponding to the display screen.

[0097] The distance sensor 180B, also called a displacement sensor, can be used to measure distance. The electronic device 100 can measure distance by infrared or laser. In some embodiments, the electronic device 100 can use the distance sensor 180B to measure distance in order to automatically adjust the brightness of the display screen of the electronic device. For example, if the electronic device is far away from the light source, it means that the electronic device is in a darker scene, then the brightness of the display screen of the electronic device will automatically dim, that is, reduce the brightness value corresponding to the display screen. For another example, if the electronic device is close to the light source, it means that the electronic device is in a brighter scene, then the brightness of the display screen of the electronic device will automatically brighten, that is, increase the brightness value corresponding to the display screen.

[0098] The temperature sensor 180C is used to detect the temperature of the environment in which the electronic device 100 is located. In some embodiments, the electronic device 100 can use the temperature detected by the temperature sensor 180C to execute a temperature processing strategy. For example, if the temperature of the environment in which the electronic device (such as a smart air conditioner) is located is high, the electronic device will automatically turn on the cooling mode to cool the environment. For another example, if the temperature of the environment in which the electronic device is located is low, the electronic device will automatically turn on the heating mode to heat the environment.

[0099] The humidity sensor 180D is used to detect the relative humidity of the environment in which the electronic device 100 is located. In some embodiments, the electronic device 100 can use the relative humidity detected by the humidity sensor 180D to execute a humidity processing strategy. For example, if the relative humidity of the environment in which the electronic device (such as a smart air conditioner) is located is high, the electronic device will automatically turn on the dehumidification mode to reduce the humidity of the environment. For another example, if the relative humidity of the environment in which the electronic device is located is low, the electronic device will automatically turn on the humidification mode to increase the humidity of the environment.

[0100] The touch sensor E is also called a "touch panel". The touch sensor 180E can be set on the display screen 194, and the touch sensor 180E and the display screen 194 form a touch screen, also called a "touch screen". The touch sensor 180E is used to detect a trigger operation acting on or near it. The touch sensor can pass the detected trigger operation to the application processor to determine the type of trigger event. A visual output related to the trigger operation can be provided through the display screen 194. In other embodiments, the touch sensor 180E can also be set on the surface of the electronic device 100, which is different from the position of the display screen 194.

[0101] The buttons 190 include a power button, a volume button, etc. The motor 191 can generate a vibration prompt. The indicator 192 can be an indicator light, which can be used to indicate the charging status, power change, messages, missed calls, notifications, etc. The SIM card interface 195 is used to connect a SIM card.

[0102] Figure 6 1 is a software structure diagram of the electronic device 100 provided in the embodiment of the present application. The layered architecture can divide the software into several layers, each layer has a clear role and division of labor. The layers communicate with each other through software interfaces. In some embodiments, Android TM The system is divided into four layers, from top to bottom: application layer, application framework layer, Android runtime layer, TM runtime) and system libraries, as well as the kernel layer.

[0103] The application layer can include a series of application packages.

[0104] like Figure 6 As shown, the application package may include applications such as camera, map, navigation, video, music, game, calendar, parameter control module, etc.

[0105] The application framework layer provides an application programming interface (API) and a programming framework for the applications in the application layer. The application framework layer includes some predefined functions.

[0106] like Figure 6 As shown, the application framework layer may include a window manager, a content provider, a view system, a phone manager, a resource manager, a notification manager, and the like.

[0107] The window manager is used to manage window programs. The window manager can obtain the display screen size, determine whether there is a status bar, lock the screen, capture the screen, etc.

[0108] Content providers are used to store and retrieve data and make it accessible to applications. The data may include videos, images, audio, calls made and received, browsing history and bookmarks, phone books, etc.

[0109] The view system includes visual controls, such as controls for displaying text, controls for displaying images, etc. The view system can be used to build applications. A display interface can be composed of one or more views. For example, a display interface including a text notification icon can include a view for displaying text and a view for displaying images.

[0110] The phone manager is used to provide communication functions of the electronic device 100, such as management of call status (including connecting, hanging up, etc.).

[0111] The resource manager provides various resources for applications, such as localized strings, icons, images, layout files, video files, and so on.

[0112] The notification manager enables applications to display notification information in the status bar. It can be used to convey notification-type messages and can disappear automatically after a short stay without user interaction. For example, the notification manager is used to notify download completion, message reminders, etc. The notification manager can also be a notification that appears in the system top status bar in the form of a chart or scroll bar text, such as notifications of applications running in the background, or a notification that appears on the screen in the form of a dialog window. For example, a text message is displayed in the status bar, a prompt sound is emitted, an electronic device vibrates, an indicator light flashes, etc.

[0113] Android runtime includes core libraries and virtual machines. Android runtime is responsible for scheduling and management of the Android system.

[0114] The core library consists of two parts: one part is the function that needs to be called by the Java language, and the other part is the Android core library.

[0115] The application layer and the application framework layer run in a virtual machine. The virtual machine executes the Java files of the application layer and the application framework layer as binary files. The virtual machine is used to perform functions such as object life cycle management, stack management, thread management, security and exception management, and garbage collection.

[0116] The system library may include multiple functional modules, such as surface manager, media libraries, 3D graphics processing library (such as OpenGL ES), 2D graphics engine (such as SGL), parameter control interface, etc.

[0117] The surface manager is used to manage the display subsystem and provide the fusion of 2D and 3D layers for multiple applications.

[0118] The media library supports playback and recording of a variety of commonly used audio and video formats, as well as static image files, etc. The media library can support a variety of audio and video encoding formats, such as: MPEG4, H.264, MP3, AAC, AMR, JPG, PNG, etc.

[0119] The 3D graphics processing library is used to implement 3D graphics drawing, image rendering, compositing, and layer processing.

[0120] A 2D graphics engine is a drawing engine for 2D drawings.

[0121] The parameter control interface is used to adjust the usage parameters of the electronic device. Specifically, when there is a difference between the predicted parameters and the current parameters, the electronic device can call the parameter control interface to adjust the current usage parameters to the predicted usage parameters, thereby meeting the user's usage needs.

[0122] The kernel layer is the layer between hardware and software. The kernel layer contains at least display driver, camera driver, audio driver, and sensor driver.

[0123] In some embodiments, the electronic device can adjust the current usage parameters through the parameter control interface in the system library to meet the user's usage requirements. Figure 6 As shown, in response to the user's triggering operation for entering the video application in the electronic device, the electronic device (or the corresponding application program described in a replacement manner) can execute step a, triggering the parameter control module (such as the target parameter preference model in the parameter control module) to predict the usage parameters of the electronic device to obtain the predicted usage parameters. In addition, the electronic device can execute step b, triggering the system library (such as the parameter control interface in the system library) to adjust the usage parameters of the electronic device, so that the adjusted usage parameters can meet the user's usage parameter requirements in the video scene, thereby improving the user's usage experience.

[0124] The following will take the electronic device as a mobile phone as an example, and divide the adjustment method of the usage parameters provided in the embodiment of the present application into two embodiments for introduction, wherein the first embodiment is used to introduce the training part of the parameter preference model, and the second embodiment is used to introduce the usage part of the parameter preference model.

[0125] Embodiment 1

[0126] The embodiment of the present application provides a method for adjusting usage parameters. In this embodiment, the cloud service uses the characteristic data of all users to train the initial model to obtain a parameter preference model set. The parameter preference model set includes multiple trained parameter preference models (or parameter preference models corresponding to usage parameters), wherein a trained parameter preference model is used to predict a usage parameter that needs to be adjusted for an electronic device in a usage scenario. Specifically, Figure 7 As shown, the device parameter adjustment method provided in the embodiment of the present application may include S701-S705:

[0127] S701, the cloud service obtains a first training feature set, wherein the first training feature set includes feature data of all users and usage parameter setting values ​​corresponding to the feature data of each user in the total number of users, wherein the feature data includes user portraits and user behavior data.

[0128] Among them, all users refer to all users who have manually adjusted the usage parameters. Feature data includes user identification, and user portrait and user behavior data corresponding to the user identification. The user portrait can reflect the user characteristics, which is generated based on the user's own attributes and user behavior habit data. The user's own attributes include at least one of user gender, user age, user city, user occupation, and user education. User behavior habit data may include the average daily usage time, which may be the average daily usage time of the user in any application scenario within a preset time, for example, the average daily usage time corresponding to the user watching videos in the past six months. Application scenarios may include video scenes, game scenes, e-book reading scenes, news browsing scenes, and general scenes.

[0129] The above user behavior data can reflect the user's usage habits, which may include at least two of the application scenario type, ambient brightness, ambient temperature, ambient humidity, ambient noise (in decibels), and usage parameter types when adjusting the usage parameters. The usage parameter types may include brightness, volume, color temperature, etc. Among them, the above general scenarios refer to scenarios other than the above video scenarios, game scenarios, e-book reading scenarios, and news browsing scenarios, such as desktop scenarios, payment application scenarios, etc. For various scenarios in the passing scenario, the settings of the usage parameters are similar.

[0130] It is understood that the data types included in the above characteristic data are only examples, and the characteristic data can be set according to actual conditions. In one example, the cloud service can reduce unnecessary characteristic data. For example, if the electronic device is an air conditioner, the cloud service can delete the environmental noise in the above characteristic data. In another example, the cloud service can also add required characteristic data. For example, if the electronic device is a mobile phone, the cloud service can increase the adjustment time of the usage parameters based on the above characteristic data.

[0131] In some embodiments, the above-mentioned full users refer to all users who have manually adjusted the usage parameters and users who have turned on the automatic parameter adjustment function. It can be understood that only when the automatic parameter adjustment function is turned on, that is, only when the mobile phone receives the user's authorization operation for the automatic parameter adjustment function, can the mobile phone adjust the usage parameter value corresponding to the application scenario.

[0132] It should be noted that before obtaining the first training feature set, the cloud service may construct feature variables for the user data of each user respectively to obtain feature data of all users.

[0133] S702: The cloud service screens the first training feature set based on the user portraits to obtain a first sub-training feature set corresponding to each user portrait.

[0134] Specifically, the cloud service can divide the user portraits and use the user portraits with higher similarity as one user portrait. For each user portrait, the cloud service determines feature data of the user portrait corresponding to the user portrait from the first training feature set, and uses the determined feature data as the first sub-training feature set corresponding to the user portrait.

[0135] In some embodiments, taking the case where the user profile includes the user's age, the cloud service may determine all feature data of users whose age belongs to the same preset age group as user profiles with a high similarity, that is, as the first sub-training feature set corresponding to the same user profile. For example, the cloud service may determine all feature data of users aged between 20 and 30 as a first sub-training feature set, and determine all sample data corresponding to users aged between 30 and 40 as a first sub-training feature set.

[0136] In some embodiments, taking the case where the user profile includes the user's education level as an example, all feature data of the user's education level belonging to the same education level are taken as user profiles with a high similarity, that is, determined as the first sub-training feature set corresponding to the user profiles of the same type. For example, the cloud service may determine all feature data corresponding to users with a bachelor's degree as a first sub-training feature set, and all feature data corresponding to users with a master's degree as a first sub-training feature set, and all feature data corresponding to users with a doctorate degree as a first sub-training feature set.

[0137] In other embodiments, taking the user profile including the user city as an example, the cloud service can take all feature data of the user city belonging to the same city as the user profile with a higher similarity, that is, determine it as the first sub-training feature set corresponding to the user profile of the same type. For example, the cloud service can determine all feature data corresponding to the user whose user city is City B as a first sub-training feature set, and determine all feature data corresponding to the user whose user city is City C as a first sub-training feature set. Alternatively, the cloud service can take all feature data of the user city belonging to the same province as the user profile with a higher similarity, that is, determine it as the first sub-training feature set corresponding to the user profile of the same type. For example, taking Province G as an example, Province G includes City G1, City G2, and City G3, and the cloud service can determine the feature data corresponding to the users whose user cities are G1, G2, and G3 as a first sub-training feature set. Alternatively, the cloud service can take all feature data of the user city belonging to the same region as the user profile with a higher similarity, that is, determine it as the first sub-training feature set corresponding to the user profile of the same type. For example, taking the regions including the east, west, south and north, the northern region includes city A1, city A2 and city A3. The cloud service can determine the feature data corresponding to users whose cities are A1, A2 and A3 as a first sub-training feature set.

[0138] The above example lists the method of determining user portraits with high similarity through one type of feature data. User portraits with high similarity can also be determined through multiple types of feature data. For example, feature data of users whose ages belong to the same preset age group and whose educational backgrounds belong to the same educational background can be used as the first sub-training feature set corresponding to the same type of user portrait.

[0139] S703: The cloud service obtains a second training feature set corresponding to various application scenarios from the first sub-training feature set.

[0140] Specifically, after obtaining the first sub-training feature sets corresponding to various user portraits, the cloud service can obtain second training feature sets corresponding to various application scenarios from the first sub-training feature sets corresponding to each user portrait.

[0141] Exemplarily, the above application scenarios include video scenarios, e-book reading scenarios, game scenarios, news browsing scenarios, and general scenarios. The cloud service may determine all feature data in the first sub-training feature set whose application scenario type is video scenarios as the second training feature set corresponding to video scenarios, determine all feature data in the first sub-training feature set whose application scenario type is game scenarios as the second training feature set corresponding to game scenarios, determine all feature data in the first sub-training feature set whose application scenario type is e-book reading scenarios as the second training feature set corresponding to e-book reading scenarios, determine all feature data in the first sub-training feature set whose application scenario type is news browsing scenarios as the second training feature set corresponding to news browsing scenarios, and determine all feature data in the first sub-training feature set whose application scenario type is general scenarios as the second training feature set corresponding to general scenarios.

[0142] S704: For each application scenario, the cloud service obtains, from the second training feature set, a third training feature set corresponding to various usage parameters corresponding to the application scenario.

[0143] Specifically, after obtaining the second training feature set corresponding to various application scenarios, the cloud service can obtain the third training feature set corresponding to various usage parameters from the second training feature set corresponding to the application scenario for each application scenario. It can be understood that the second training feature set includes the third training feature set corresponding to each usage parameter corresponding to the application scenario. The third training feature set is the feature data of a usage parameter of an application scenario corresponding to the same type of user portrait.

[0144] It can be understood that in addition to some common feature data (such as user gender, user age, user city, user occupation, user education, etc.), in order to increase the personalized and intelligent parameter control effect and enhance the user experience, in this embodiment, different feature data are set for different usage parameter types, that is, the feature data corresponding to different usage parameter types may be different. For example, if the above-mentioned usage parameter type is brightness, the feature data may include ambient brightness and application scenario type. For another example, if the above-mentioned usage parameter type is volume, the feature data may include ambient noise and application scenario type. For another example, if the above-mentioned usage parameter type is color temperature, the feature data may include ambient noise, application scenario type and media playback mode. Among them, the media playback mode may include a relaxed mode, a comedy mode and a horror mode, etc.

[0145] Exemplarily, as shown in Table 1, taking the usage parameter type as brightness as an example, the feature data of the user in the third training feature set may include the user's unique number (identification, ID), user age, user city, user occupation, user education, average daily usage time in the game scenario, average daily usage time in the video scenario, average daily usage time in the novel reading scenario, average daily usage time in the news reading scenario, application scenario type, ambient brightness, ambient temperature and ambient humidity, and the brightness parameter setting value corresponding to each feature data (here the brightness parameter setting value refers to the usage parameter setting value). For example, the user age of user ID1 is 32 years old, the user's city is B, the user's occupation is engineer, the user's education is a master's degree, the user's average daily usage time in the game scenario is 0.2h, the user's average daily usage time in the video scenario is 0.4h, the user's average daily usage time in the novel reading scenario is 0.2h, the user's average daily usage time in the news reading scenario is 0.8h, the application scenario type when the user adjusts the usage parameters is video, and the ambient brightness is 80cd / m 2 , the ambient temperature is 26℃, the ambient humidity is 40% rh, and the brightness parameter setting value is 120. For another example, the user ID2 is 20 years old, the user is located in S city, the user occupation is a student, the user education is a master's degree, the user's average daily usage time in the game scene is 0.5h, the user's average daily usage time in the video scene is 0.6h, the user's average daily usage time in the novel reading scene is 0.35h, the user's average daily usage time in the news reading scene is 0.2h, the application scene type when the user adjusts the usage parameters is game, and the ambient brightness is 90cd / m 2 , the ambient temperature is 28℃, the ambient humidity is 50%rh, and the brightness parameter setting value is 150.

[0146] Table 1

[0147]

[0148] S705. For each of the above usage parameters, the cloud service trains the initial model based on the third training feature set corresponding to the usage parameter to obtain a parameter preference model corresponding to the usage parameter.

[0149] Among them, the above-mentioned parameter preference model is used to predict the parameter value corresponding to a usage parameter that needs to be adjusted in an application scenario.

[0150] In some embodiments, the parameter preference model is that each usage parameter corresponds to a parameter preference model, for example, brightness corresponds to a parameter preference model, volume corresponds to a parameter preference model, etc. In other implementations, the parameter preference model may be that all usage parameter types corresponding to an application scenario correspond to only one parameter preference model, that is, the parameter preference model can predict parameter values ​​corresponding to all usage parameters.

[0151] It can be understood that since the first training feature set includes feature data of all users (or feature data of multiple users), and the feature data can include data corresponding to multiple application scenarios and multiple usage parameters, after the cloud service trains the initial model, it can obtain a parameter preference model set, and the parameter preference model set includes multiple parameter preference models.

[0152] Specifically, the process of training the parameter preference model may specifically include: for each usage parameter, the cloud service inputs the feature data in the third training feature set corresponding to the usage parameter into the initial model to train the initial model. For example, for the brightness parameter corresponding to the video scene, the cloud service may input the third training feature set corresponding to the brightness parameter into the initial model to train the initial model and obtain the parameter preference model corresponding to the brightness parameter corresponding to the video scene. The parameter preference model can output a first predicted parameter value, which represents the predicted value of the brightness parameter in the video scene.

[0153] In some embodiments, the initial model may be a multiple linear regression model, wherein the multiple linear regression model uses two or more influencing factors as independent variables to explain the change of the dependent variable.

[0154] Exemplarily, the above multivariate linear regression model can be calculated by the following formula 1:

[0155]

[0156] in, is the target variable, that is, the first prediction parameter value, w=(w1,w2,...,w n ) is the regression coefficient, X=(X1,X2,...,X n ) is a feature variable, that is, the feature data of the above-mentioned user, for example, X1 is the user's age, X2 is the user's city, X3 is the user's occupation, etc., and b is a constant.

[0157] Specifically, the multivariate linear regression model can be fitted on the training feature set using the least squares method (MSE) to obtain the coefficients in the multivariate linear regression model, that is, w and b. The least squares method can find the best function that matches the feature data of the user by minimizing the sum of squares of the errors. In other words, the first prediction parameter value can be easily obtained using the least squares method, and the sum of squares of the errors between the first prediction parameter value and the setting parameter value can be minimized.

[0158] In other embodiments, the initial model may be a neural network model. The neural network model is composed of a fully connected input layer, a hidden layer, and an output layer. The layers are fully connected, that is, any neuron in the i-th layer must be connected to any neuron in the i+1-th layer. The number of hidden layers may be 1 or multiple layers.

[0159] Exemplarily, the structure of the above neural network model is as follows Figure 8 As shown, X=(X1,x2,...,X n ) is a feature variable, that is, the feature data of the above user, for example, X1 is the user's age, X2 is the user's city, X3 is the user's occupation, etc., w=(W ij ) is the connection weight in the neural network model, i is the number of neuron layers, j is the number of neuron columns, and Y is the predicted output value of the neural network model, which is the first prediction parameter value mentioned above.

[0160] Specifically, the above neural network model is a linear relationship plus an activation function, which is used to delinearize, and the cross entropy is used as the loss function. The stochastic gradient descent algorithm, batch gradient descent algorithm and other error back propagation methods are used for learning and training, and the connection weights W in the above neural network model are adjusted and updated. ij .

[0161] It should be noted that after obtaining the third training feature set, the cloud service may preprocess the feature data in the third training feature set. Afterwards, the cloud service may train the initial model based on the preprocessed feature data to obtain the parameter preference model corresponding to the above-mentioned usage parameters.

[0162] In some embodiments, for the enumerated variables in the feature data, the enumerated variables are one-hot encoded. Among them, the enumerated variable refers to listing the values ​​of the variable one by one, and the value of the variable is limited to the listed values. The enumerated variable is used to represent the feature variable whose value is an identifier. The enumerated variable may include the user city, user occupation, user education, and application scenario type in the above-mentioned feature data. One-hot encoding refers to using an N-bit state register to encode N states, each state has its own independent register bit, and at any time, only one state register is valid. Exemplarily, taking the enumerated variable as an example of user education, the user education may include high school, technical secondary school, undergraduate, junior college, master and doctor. Afterwards, the cloud service can encode each user's education, that is, high school is 1, technical secondary school is 2, undergraduate is 3, junior college is 4, master is 5, and doctor is 6. Afterwards, if the education of user ID1 is a master, it can be determined that the user's education is 000010.

[0163] In other embodiments, for the numerical variables in the feature data, the numerical variables are normalized. Among them, the numerical variables are used to represent the feature variables whose values ​​are numerical values. The numerical variables may include the user age, the average daily usage time in the game scene, the average daily usage time in the video scene, the average daily usage time in the e-book reading scene, the average daily usage time in the news browsing scene, the ambient brightness, the ambient temperature and the ambient humidity in the above-mentioned feature data. The normalization process refers to limiting the feature data within a certain range, thereby reducing the occurrence of situations in which the model training is affected by the presence of abnormal data in the feature data. Exemplarily, taking the temperature range corresponding to the ambient temperature as 20 to 35°C as an example, if the ambient temperature in the feature data is 40°C, it can be said that the ambient temperature is abnormal data. Therefore, the cloud service can delete the feature data of the user corresponding to the abnormal data from the above-mentioned training feature set.

[0164] It can be understood that since the third training feature set contains multiple feature data, for example, taking the brightness corresponding to the game scene as an example, the feature data may include user gender, user age, user city, user occupation, user education, ambient brightness, and application scenario type. Therefore, when the cloud service trains the initial model, each feature data in the third training feature set can be assigned a corresponding weight. In this embodiment, the weight corresponding to each feature data is set to 1. In other embodiments, it can be determined according to the number of feature data corresponding to a user portrait. For example, taking the user portrait as the user age as an example, if the number of feature data for the user age between 20 and 30 years old is large, the weight corresponding to the user age can be set to 0.3; if the number of feature data for the user age between 80 and 90 years old is small, the weight corresponding to the user age can be set to 0.9.

[0165] In some embodiments, a parameter preference model can also be used to predict parameter values ​​corresponding to multiple usage parameters that need to be adjusted in an application scenario, which is equivalent to a parameter preference model including multiple parameter preference models corresponding to the above-mentioned usage scenario. For example, the parameter preference model is used to predict the parameter preference model corresponding to the volume corresponding to the video scene and the parameter preference model corresponding to the brightness.

[0166] In some embodiments, a parameter preference model can also be used to predict a parameter value corresponding to a usage parameter that needs to be adjusted in multiple application scenarios, which is equivalent to a parameter preference model including a parameter preference model corresponding to the above-mentioned multiple scenarios. For example, the parameter preference model is used to predict the parameter preference model corresponding to the volume corresponding to the video scene and the parameter preference model corresponding to the volume corresponding to the game scene.

[0167] In other embodiments, a parameter preference model can also be used to predict parameter values ​​corresponding to multiple usage parameters that need to be adjusted in multiple application scenarios, which is equivalent to a parameter preference model including multiple parameter preference models corresponding to the above-mentioned multiple scenarios. For example, the parameter preference model is used to predict the parameter preference model corresponding to the volume corresponding to the video scene, the parameter preference model corresponding to the brightness corresponding to the video scene, the parameter preference model corresponding to the brightness corresponding to the game scene, and the parameter preference model corresponding to the volume corresponding to the game scene.

[0168] In some embodiments, after obtaining the characteristic data of all users, the cloud service can also train the above initial model for each user based on the characteristic data corresponding to the user to obtain a parameter preference model corresponding to the user. In this way, the obtained parameter preference model can accurately predict the user's parameter adjustment habits and achieve accurate prediction of usage parameters.

[0169] Embodiment 2

[0170] The embodiment of the present application provides a method for adjusting usage parameters. After the parameter preference model training is completed, the mobile phone can obtain the target parameter preference model according to the user identifier, and use the target parameter preference model to predict the second predicted parameter value in the scene where the mobile phone is currently located when the application scene switches. Afterwards, when there is a difference between the second predicted parameter value and the current parameter value, the current parameter value is adjusted to the second predicted parameter value, so that the adjusted usage parameters can meet the user's usage parameter requirements in the video scene, thereby reducing the time for the user to adjust the parameters and improving the user's usage experience.

[0171] Specifically, Fig. 9 As shown, in the embodiment of the present application, the model training method provided in the embodiment of the present application may include S901-S907:

[0172] S901, the mobile phone obtains a target parameter preference model from a cloud service, wherein the target parameter preference model is a parameter preference model corresponding to a user identifier corresponding to the mobile phone.

[0173] Exemplarily, the above-mentioned user identification may be user account information (such as user ID). The mobile phone may determine whether there is a parameter preference model matching the user account information in the mobile phone. If there is no parameter preference model matching the user account information in the mobile phone, the mobile phone may obtain the parameter preference model matching the user account information from the above-mentioned cloud service, and determine the parameter preference model matching the user account information as the target parameter preference model. If there is a parameter preference model matching the user account information in the mobile phone, the mobile phone does not need to obtain the parameter preference model matching the user account information from the cloud service, that is, the mobile phone may use the parameter preference model existing in the mobile phone as the target parameter preference model. Of course, the above-mentioned target parameter preference model may also be actively sent by the cloud service.

[0174] In some embodiments, if there is no parameter preference model matching the user account information in the mobile phone, the mobile phone can send a model sending request to the cloud service, wherein the model sending request carries the user account information. Afterwards, the cloud service responds to the model sending request, determines the user portrait feature corresponding to the user based on the user account information, and selects the parameter preference model corresponding to the user portrait feature from the above parameter preference model set. Afterwards, the cloud service sends the parameter preference model corresponding to the user portrait feature to the mobile phone. Afterwards, after receiving the parameter preference model corresponding to the user portrait feature, the mobile phone can determine the parameter preference model corresponding to the user portrait feature as the target parameter preference model.

[0175] In other embodiments, if there is no parameter preference model matching the user account information in the mobile phone, the cloud service can directly send the parameter preference model corresponding to the user account information to the mobile phone after training the parameter preference model. That is, the mobile phone can directly obtain the parameter preference model corresponding to the user account information, and use the parameter preference model corresponding to the user account information as the target parameter preference model.

[0176] It can be understood that the absence of a parameter preference model matching the user account information in the mobile phone indicates that the user has not performed any historical parameter setting behavior for each scenario, or that the mobile phone only contains a small number of historical parameter setting behaviors for each scenario. The presence of a parameter preference model matching the user account information in the mobile phone indicates that the user has used the mobile phone for a period of time, that is, the mobile phone contains a large number of historical parameter setting behaviors for each scenario.

[0177] It should be noted that the above-mentioned target parameter preference model includes at least one first parameter preference model. Each first parameter preference model is used to predict the setting value of a usage parameter corresponding to a usage scenario. For example, the target parameter preference model includes three first parameter preference models, namely model 1, model 2 and model 3, model 1 is used to predict the setting value of the brightness parameter in the video scene (or described as the brightness parameter corresponding to the video scene), model 2 is used to predict the setting value of the volume parameter in the video scene, and model 3 is used to predict the setting value of the brightness parameter in the e-book reading scene.

[0178] In a possible implementation, after obtaining the target parameter preference model, the mobile phone can update the parameter preference model corresponding to the user identifier corresponding to the mobile phone, that is, the target parameter preference model, according to the usage parameters manually adjusted by the user in the daily use of the mobile phone, that is, using the new feature data and the setting values ​​of the usage parameters corresponding to the new feature data. Fig.10 As shown, when the mobile phone receives the target parameter preference model, the mobile phone can determine whether the cache data corresponding to the usage parameter has changed, so as to train the target parameter preference model with the new parameter setting result when the change occurs. Exemplarily, the mobile phone can determine whether the cache data corresponding to the usage parameter has changed by polling or monitoring.

[0179] Exemplarily, for each usage parameter, a new record is added to the cache data corresponding to the usage parameter, indicating that the cache data corresponding to the usage parameter has changed. Therefore, the mobile phone can extract the new parameter setting result and construct the feature variable for the new parameter setting result, that is, normalize the record to obtain the new feature data corresponding to the usage parameter and the setting value of the usage parameter corresponding to the new feature data. Afterwards, the mobile phone can use the new feature data and the setting value of the usage parameter corresponding to the new feature data to continue training the first parameter preference model corresponding to the usage parameter, and obtain the updated target parameter preference model corresponding to the usage parameter, so that the target parameter preference model can better learn the user's usage parameter setting habits, thereby improving the accuracy of usage parameter prediction.

[0180] Specifically, the process of continuing to train the first parameter preference model corresponding to the usage parameter using the new feature data and the setting value of the usage parameter corresponding to the new feature data may include the mobile phone setting a first weight for the new feature data. The first weight belongs to a preset weight range, and the preset weight range may be 10 to 10000.

[0181] In some embodiments, the value of the first weight may be related to the addition time corresponding to the feature data. For example, if the addition time corresponding to feature data A is 1 minute ago, that is, the time of cache data setting corresponding to the parameters in feature data A is shorter, then the mobile phone may set the weight corresponding to feature data A to 500. For another example, if the addition time corresponding to feature data B is 2 years ago, that is, the time of cache data setting corresponding to the parameters in feature data B is longer, then the mobile phone may set the weight corresponding to feature data B to 10.

[0182] In other embodiments, the value of the first weight may be related to the determined object corresponding to the setting value of the usage parameter corresponding to the feature data. For example, if the setting value of the usage parameter corresponding to the new feature data is determined in response to an adjustment operation input by a user, it means that the setting value is set by the user according to his or her own needs. Therefore, the mobile phone can set the first weight of the new feature data to a first preset weight (such as 400). For another example, if the setting value of the usage parameter corresponding to the new feature data is determined based on a parameter preference model, it means that the setting value is predicted by the mobile phone based on the user's usage habits. Therefore, the mobile phone can set the first weight of the new feature data to a second preset weight (such as 40). It can be understood that the second preset weight is less than the first preset weight.

[0183] Afterwards, the mobile phone can update the first parameter preference model corresponding to the above-mentioned usage parameters based on the new feature data with the first weight and the setting value of the usage parameter corresponding to the new feature data, and save the updated first parameter preference model to facilitate subsequent parameter prediction based on the updated target parameter preference model.

[0184] It can be understood that if the cache data corresponding to the above usage parameters has not changed, it means that the user has not manually adjusted the usage parameters, that is, the user is relatively satisfied with the usage parameter values ​​automatically adjusted by the mobile phone. Therefore, the mobile phone does not need to update the target parameter preference model received by the above mobile phone.

[0185] It should be noted that after obtaining the updated first parameter preference model, the mobile phone can delete the cached parameter setting results to re-record new parameter setting results, so as to ensure that the cached parameter setting results are more in line with the user's usage habits.

[0186] In some embodiments, when the mobile phone receives a manual adjustment operation of a user for a usage parameter, it determines the record corresponding to the usage parameter and adds it to the cache data. Afterwards, the mobile phone can adjust the current value of the usage parameter according to the cache data to meet the user's parameter adjustment requirements.

[0187] In one example, the mobile phone may report all cached data manually adjusted by the user within the target time to the cloud service to be saved in the above training feature set.

[0188] In another example, the mobile phone can report the feature data newly added by the user to the cloud service in real time to be saved in the training feature set. When the amount of uploaded feature data meets the preset amount threshold, the mobile phone can stop uploading.

[0189] In a possible implementation, the target parameter model may not be obtained from the cloud service, but may be obtained by training the mobile phone according to the usage parameter values ​​set by the user for different scenarios when using the mobile phone, and the corresponding mobile phone may not send a model sending request to the cloud service. Specifically, the mobile phone obtains a feature data set of the current user, wherein the feature data set includes multiple feature data. Afterwards, the initial model is trained according to the feature data set to obtain the target parameter preference model corresponding to the user identifier corresponding to the mobile phone.

[0190] In some embodiments, the mobile phone may not obtain the target parameter preference model first. After determining that the application scenario has changed, the mobile phone may obtain the target parameter preference model based on the user account information. There is no specific limitation.

[0191] S902: In response to a trigger operation of the user on the mobile phone, the mobile phone determines whether the application scenario has changed.

[0192] The trigger operation (or first trigger operation) is used to trigger the mobile phone to perform a target operation, and the target operation is used to display a first interface. The first interface refers to the interface displayed after the mobile phone performs the target operation.

[0193] Exemplarily, when receiving a trigger operation from the user on the current interface (or second interface) of the mobile phone, the mobile phone determines whether the application scenario has changed. Among them, the current interface refers to the interface currently displayed on the display screen of the mobile phone, that is, the interface displayed before the mobile phone executes the above-mentioned target operation. The trigger operation refers to the operation performed by the user's finger touching the display screen. In some embodiments, the trigger operation can be a sliding operation on the current interface, or a click operation on any interface content in the current interface, etc., without specific limitation. In one example, the sliding operation can be an upward sliding operation on the current interface, or a downward sliding operation on the current interface (such as refreshing the page), or a left sliding operation on the current interface, etc.

[0194] Specifically, after receiving the user's trigger operation on the current interface, the mobile phone determines whether the application scenario to which the current interface belongs and the application scenario to which the interface displayed after the trigger belongs (or referred to as the first application scenario) are the same scenario. The mobile phone can determine whether the current interface and the interface generated after the trigger are the same scenario by active detection or callback, that is, determine whether the application scenario has changed.

[0195] Exemplarily, if the application scenario to which the current interface belongs is the same as the application scenario to which the interface displayed after being triggered belongs, the mobile phone can determine that the application scenario has not changed, therefore, the mobile phone does not need to reset the usage parameters, that is, the mobile phone does not need to perform parameter prediction, and the mobile phone can execute S903, thereby avoiding unnecessary parameter predictions and reducing unnecessary power consumption losses. If the application scenario to which the current interface belongs is different from the application scenario to which the interface displayed after being triggered belongs, the mobile phone can determine that the application scenario has changed, therefore, the mobile phone can predict the usage parameter value corresponding to the interface displayed after being triggered, that is, the mobile phone can execute S904 to determine whether the current usage parameter value meets user needs, thereby determining whether the mobile phone needs to adjust the usage parameters based on the predicted usage parameter value. Among them, application scenarios can include video scenarios, game scenarios, e-book reading scenarios, news browsing scenarios, and general scenarios.

[0196] In some embodiments, if the trigger operation is a sliding operation, the mobile phone displays a corresponding interface according to the sliding position and sliding direction of the user's finger, or it can be understood as determining a sliding function corresponding to the sliding operation. Among them, the sliding function may include a function of returning to the previous interface, a function of returning to the desktop, a function of refreshing the interface, a function of browsing up, and a function of browsing down. Afterwards, the mobile phone can determine whether the application scenario has changed according to the interface displayed before and after the sliding operation is input, that is, determine whether the application scenario has changed according to the sliding function.

[0197] In one example, see Fig.11 As shown, taking the current interface as a short video playback interface as an example, if the user's finger starts from the bottom of the mobile phone display and slides upward, the mobile phone can respond to the sliding operation and display the desktop. In other words, the sliding function corresponding to the sliding operation is the return to the main interface function. Accordingly, the application scene switches from the video scene to the general scene, so the mobile phone can determine that the application scene has changed.

[0198] In another example, see Fig.12 As shown, taking the current interface as a short video playback interface as an example, if the user's finger starts from any position except the edge position in the mobile phone display screen interface and slides upward, the mobile phone can respond to the sliding operation and display the corresponding content of the next video. In other words, the sliding function corresponding to the sliding operation is the downward browsing function. Accordingly, since the interface displayed by the mobile phone is still the video interface, that is, the application scene is still the video scene, the mobile phone can determine that the application scene has not changed.

[0199] In other embodiments, if the trigger operation is a click operation, the mobile phone can respond to the clicked content and display a corresponding interface, or it can be understood as determining a click function corresponding to the click operation. The click function includes an entry function and an exit function. Afterwards, the mobile phone can determine whether the application scenario has changed based on the interface displayed before and after the click operation is input, that is, determine whether the application scenario has changed based on the click function.

[0200] In one example, the current interface is taken as the initial interface of the game. The user clicks on Fig.13 In the game initial interface shown in (a), the mobile phone can respond to the click operation of the "start game" control and display the following Fig.13 The game interface shown in (b) in the figure. That is, the click function corresponding to the click operation is the enter function. Fig.13 The initial game interface shown in (a) belongs to the general scenario, and Fig.13The game interface shown in (b) belongs to the game scene, that is, the application scene is switched from the general scene to the game scene. Therefore, the mobile phone can determine that the application scene has changed.

[0201] In one example, the current interface is a text message list interface. The user clicks on Fig.14 In the short message list interface shown in (a), the mobile phone can respond to a click operation on the short message icon 8 and display the following information: Fig.14 The SMS details interface shown in (b) includes the details of the clicked SMS (such as Hello, your courier is being delivered. If you have any questions, please contact the courier. Thank you~). In other words, the click function corresponding to the click operation is the entry function. Since the SMS list interface shown in (a) of 14 is a general scenario, and Fig.14 The SMS details interface shown in (b) is still a universal scenario, that is, the application scenario is still a universal scenario, so the mobile phone can determine that the application scenario has not changed.

[0202] In another example, taking the current interface as a video playback interface, the user clicks on Fig.15 In the "return" control shown in (a), the mobile phone can respond to the click operation of the control and display the following Fig.15 The video list interface shown in (b) in the figure. In other words, the click function corresponding to the click operation is the exit function. Fig.15 The video playback interface shown in (a) is a video scene, and Fig.15 The video list interface shown in (b) is a general scene, that is, the application scene is switched from the video scene to the general scene, so the mobile phone can determine that the application scene has changed.

[0203] In some possible implementations, after receiving a trigger operation from the user on the current interface, the mobile phone does not need to determine whether the application scenario has changed. That is, the mobile phone can directly execute the following step S904, that is, directly predict the usage parameter value based on the current application scenario to improve the accuracy of determining the usage parameter value.

[0204] In this embodiment, the electronic device is taken as an example of a mobile phone, but in fact, the electronic device can also be a smart home, such as an air conditioner, a washing machine, etc. For example, if the electronic device is an air conditioner, in response to the user's first trigger operation on the air conditioner, the air conditioner performs a target operation. The target operation is used to start or enter a first mode, and the first mode can be a sleep mode, a cooling mode, etc.

[0205] It should be noted that in order to ensure the user's privacy, it is necessary to determine whether the user has enabled the automatic parameter adjustment function before the mobile phone determines whether the application scenario has changed. When the automatic parameter adjustment function is turned on, that is, when the mobile phone receives the user's authorization operation for the automatic parameter adjustment function, the mobile phone can determine whether the application scenario has changed to adjust the usage parameter value.

[0206] For example, see Fig.16 As shown, the interface 50 displays a switch control F of personalized control, which is used to indicate whether the mobile phone turns on the automatic parameter adjustment function. When the switch control is in the off state, it indicates that the user has not enabled the automatic parameter adjustment function. When the switch control is in the on state, it indicates that the user has turned on the automatic parameter adjustment function. It can be understood that in this embodiment, the initial state of the default automatic parameter adjustment function is the off state, that is, the switch control F is a switch control in the off state.

[0207] When the user wants to turn on the automatic parameter adjustment function, the user can click the switch control F on the interface 50. In response to the user's click operation on the switch control F, the mobile phone can display the following Fig.16 The interface 51 shown in the figure shows authorization information for data collection, a rejection control 501 and an approval control 502. The authorization information is used to help users understand the authorization content and related authorization knowledge. For example, the authorization information can be embodied in the form of agreement terms.

[0208] In one example, when the user does not want to enable the automatic parameter adjustment function, the user can click the rejection control 501 on the interface 51. In response to the user's click operation on the rejection control 501, the mobile phone can return to the above display interface 50. In other words, the mobile phone will not automatically adjust the usage parameters, that is, it will not execute the device parameter adjustment method, such as predicting the usage parameter value according to the current usage scenario for subsequent adjustment of the usage parameter value.

[0209] In another example, when the user still wants to turn on the automatic parameter adjustment function, the user can click the consent control 502 on the interface 51. In response to the user's click operation on the consent control 502, the mobile phone can display the interface 52, which displays the switch control F'. In other words, the mobile phone can automatically adjust the usage parameters, that is, execute the device parameter adjustment method.

[0210] In some embodiments, the above Fig.16The interface asking the user whether to automatically adjust the parameter function is only an example, and the user can also determine whether to turn on the automatic adjustment parameter function in other ways. For example, the mobile phone can display the usage parameter options so that the user can choose to turn on or off the automatic adjustment function of which usage parameter, such as the user can choose whether to turn on the automatic adjustment function of brightness.

[0211] In other embodiments, the mobile phone may ask the user whether to automatically adjust the parameter function when installing an application (APP).

[0212] S903: The mobile phone does not predict the usage parameter value.

[0213] Specifically, when it is determined that the application scenario has not changed, it means that the mobile phone does not need to adjust the parameter value of the currently used parameter, that is, the mobile phone does not need to predict the used parameter value. In this way, the number of predictions of the used parameter value can be reduced, thereby reducing unnecessary power consumption loss.

[0214] S904, for each usage parameter corresponding to the first application scenario, the mobile phone inputs the current feature data into the first parameter preference model corresponding to the usage parameter in the above-mentioned target parameter preference model to obtain a second predicted parameter value corresponding to the usage parameter.

[0215] Specifically, when it is determined that the application scenario has changed, it means that the mobile phone may need to adjust the parameter values ​​of one or more usage parameters under the current application scenario (or the first application scenario), that is, the mobile phone needs to predict the parameter values ​​corresponding to various usage parameters. Therefore, the mobile phone can determine the first parameter preference model corresponding to the usage parameter from the above-mentioned target parameter preference model for each usage parameter corresponding to the first application scenario, and input the current feature data into the first parameter preference model to predict the parameter value, thereby obtaining the second predicted parameter value corresponding to the usage parameter, which provides a basis for subsequent judgment on whether to adjust the usage parameter value. Among them, the first application scenario refers to the scenario to which the interface displayed by the mobile phone in response to the user's trigger operation belongs. The second predicted parameter value refers to the usage parameter value predicted by the mobile phone to meet the user's usage needs.

[0216] Among them, the above-mentioned current feature data includes at least one of a user portrait, user behavior data and current environmental parameters. The user portrait includes at least one of user gender, user age, user city, user occupation, user education and average daily usage time. The user behavior data includes at least one of application scenario type and media playback mode. The current environmental parameters include at least one of current ambient brightness, current ambient temperature, current ambient humidity and current ambient noise. It should be noted that the content of the current feature data is related to the above-mentioned usage parameters.

[0217] In some embodiments, when the usage parameter is brightness, the current feature data may include user age, user city, user occupation, user education, average daily usage time, application scenario type, current environment brightness, etc. That is, the second predicted parameter value obtained by performing parameter value prediction based on the current feature data is a predicted brightness value.

[0218] In some embodiments, when the usage parameter is volume, the current feature data may include user age, user city, user occupation, user education, average daily usage time, application scenario type, current environmental noise, etc. That is, the second predicted parameter value obtained by performing parameter value prediction based on the current feature data is the predicted volume value.

[0219] In other embodiments, when the usage parameter is color temperature, the current feature data may include user age, user city, user occupation, user education, average daily usage time, application scenario type, current environmental noise, and media playback mode, etc. That is, the second predicted parameter value obtained by performing parameter value prediction based on the current feature data is a predicted color temperature value.

[0220] In this implementation, the first parameter preference model corresponding to the usage parameter corresponding to the first application scenario can reflect the user's usage habits, that is, the user's setting habits of the usage parameter in the application scenario. By judging the application scenario, the predicted value of the usage parameter in the current application scenario is determined, so that the predicted usage parameter value can be consistent with the user's usage habits, and the accurate prediction of the usage parameter value can be achieved, which can meet the different needs of users in different application scenarios and improve the user's usage experience. In addition, the mobile phone can predict the parameter value of the usage parameter corresponding to the first application scenario based on the user's previous setting preferences, so that the predicted parameter value is more consistent with the user's preference, which not only saves the user's time in manually setting parameters, but also improves the user's usage experience.

[0221] In a possible implementation, the mobile phone can cache the historical setting values ​​of the usage parameters and their corresponding characteristic data (or first characteristic data) to obtain the parameter setting results. Caching means storing the setting values ​​of the usage parameters and their corresponding characteristic data in a preset location. Afterwards, when the application scenario changes, the mobile phone can directly search for the parameter setting results in the preset location.

[0222] The setting value of the usage parameter may be a parameter value manually set by the user and / or the second predicted parameter value. For example, the mobile phone may cache a preset number of parameter setting results, which may be 1,000 or 800. After obtaining a new parameter setting result, if the number of parameter setting results is equal to the preset number, the mobile phone may delete the parameter setting result with the earliest cache time from the cache, and then add the new parameter setting result to the pre-stored position, thereby implementing a cache update, so that the parameter setting results in the pre-stored position can reflect the user's latest usage habits.

[0223] Among them, the setting value of the usage parameter in the parameter setting result and its corresponding first characteristic data indicate the setting status of the usage parameter when the characteristic data is the first characteristic data. In simple terms, it indicates the setting status of the usage parameter when the mobile phone environment is the environment corresponding to the first characteristic data.

[0224] In some embodiments, when the application scenario (or usage scenario) changes, for each usage parameter corresponding to the current application scenario, the mobile phone can first search the cached parameter setting results to see whether there is a parameter setting result with the same first feature data as the current feature data. If so, it indicates that the mobile phone has set the usage parameter in the environment corresponding to the current feature data. Therefore, the mobile phone can directly use the setting value of the usage parameter corresponding to the current feature data as the second predicted parameter value corresponding to the usage parameter. If not, it indicates that the mobile phone has not set the usage parameter in the environment corresponding to the current feature data. The mobile phone can execute the above S904 to determine the second predicted parameter value, thereby avoiding unnecessary prediction of the usage parameter.

[0225] S905: The mobile phone determines whether there is a difference between a second predicted parameter value corresponding to the usage parameter and a current usage parameter value corresponding to the usage parameter.

[0226] The current usage parameter value refers to the usage parameter value corresponding to the case where the mobile phone does not receive the above trigger operation.

[0227] In one embodiment, after obtaining the second predicted parameter value corresponding to the usage parameter, the mobile phone can compare the second predicted parameter value with the current usage parameter value. If the second predicted parameter value is the same as the current usage parameter value, it means that there is no difference between the second predicted parameter value and the current usage parameter value. Therefore, the mobile phone does not need to adjust the current usage parameter value, that is, the mobile phone can continue to maintain the current usage parameter value. If the second predicted parameter value is different from the current usage parameter value, it means that there is a difference between the second predicted parameter value and the current usage parameter value. Therefore, the mobile phone needs to adjust the current usage parameter value so that the usage parameter value reaches the second predicted parameter value, thereby improving the user experience.

[0228] S906: The mobile phone adjusts the parameter value corresponding to the usage parameter from the current usage parameter value to the second predicted parameter value.

[0229] Exemplarily, taking the usage parameter as volume as an example, if the current usage parameter value is 50 and the second predicted parameter value is 90, the mobile phone can adjust the volume parameter value from 50 to 90 to meet the user's usage needs.

[0230] S907: The mobile phone does not adjust the currently used parameter value.

[0231] Specifically, after determining that there is no difference between the second predicted parameter value and the currently used parameter value, the mobile phone does not need to adjust the currently used parameter value, that is, the mobile phone can continue to maintain the currently used parameter value, thus reducing unnecessary power consumption loss.

[0232] In some embodiments, after determining that the application scenario has changed, the parameter control module in the mobile phone can predict the use parameter value based on the current feature data, so as to adjust the use parameter value in time when there is a difference between the predicted parameter value and the current use parameter value. Figure 5 The structure shown and Fig.17 The adjustment process of the usage parameters shown in the figure introduces in detail the process of predicting the usage parameter values ​​by the parameter control module in the mobile phone.

[0233] S1601: The model training module in the cloud service obtains a first training feature set, wherein the first training feature set includes feature data of all users.

[0234] In some embodiments, the feature data of the user in the training feature set may be obtained from the mobile phone. Specifically, the mobile phone may send the feature data of the user corresponding to the mobile phone within the target time to the cloud service, so that the cloud service can receive the feature data of multiple users (or the feature data of all users).

[0235] S1602: The model training module in the cloud service screens the first training feature set based on the user portrait to obtain a first sub-training feature set corresponding to each user portrait.

[0236] S1603. The model training module in the cloud service obtains a second training feature set corresponding to various application scenarios from the above-mentioned first sub-training feature set.

[0237] S1604. For each application scenario, the model training module in the cloud service obtains a third training feature set corresponding to various usage parameters corresponding to the application scenario from the second training feature set.

[0238] S1605. For each of the above usage parameters, the model training module in the cloud service trains the initial model according to the above third training feature set to obtain a parameter preference model corresponding to the usage parameter.

[0239] Specifically, after obtaining the third training feature set, the model training module can train the initial model for each usage parameter according to the third training feature set corresponding to the usage parameter to obtain a trained parameter preference model, that is, a parameter preference model corresponding to the usage parameter.

[0240] It can be understood that the model training module in the cloud service can train multiple different parameter preference models, each parameter preference model corresponds to a usage parameter of an application scenario corresponding to the same type of user portrait. That is to say, the model training module can obtain a parameter preference model set after inputting different feature data into the initial model for training, and the parameter preference model set includes multiple parameter preference models.

[0241] S1606: The model delivery module in the cloud service receives the parameter preference model set sent by the model training module, wherein the parameter preference model set includes a plurality of parameter preference models corresponding to the usage parameters.

[0242] S1607. The model sending module in the cloud service receives the user identifier sent by the model updating submodule in the parameter control module in the mobile phone.

[0243] The parameter control module includes a model update submodule and a parameter control submodule. The model update submodule is used to update the target parameter preference model to reduce the error of the predicted parameter value. The parameter control submodule is used to control the use parameter value and, when there is a difference between the second predicted parameter value and the current use parameter value, call the parameter control interface to adjust the use parameter value.

[0244] In some embodiments, the user identification may be user account information.

[0245] S1608. The model issuing module determines a target parameter preference model from the above parameter preference model set according to the user identifier.

[0246] The target parameter preference model is a parameter preference model corresponding to a user identifier corresponding to the mobile phone. The target parameter preference model includes at least one first parameter preference model.

[0247] S1609: The model updating submodule receives the target parameter preference model sent by the model sending module.

[0248] S1610: When the application scenario changes, the parameter control submodule sends a parameter prediction request to the model update submodule.

[0249] S1611. When receiving the above-mentioned parameter prediction request, the model updating submodule inputs the current feature data into the first parameter preference model corresponding to each usage parameter corresponding to the first application scenario to obtain a second prediction parameter value corresponding to the usage parameter.

[0250] Among them, the first application scenario represents the current application scenario.

[0251] S1612. The parameter control submodule receives the second prediction parameter value sent by the model updating submodule.

[0252] S1613. The parameter control submodule determines whether there is a difference between the second predicted parameter value and the currently used parameter value.

[0253] In some embodiments, when there is a difference between the second predicted parameter value corresponding to the usage parameter and the current usage parameter value corresponding to the usage parameter, the parameter control submodule can call the parameter control interface to adjust the usage parameter value; when there is no difference between the second predicted parameter value corresponding to the usage parameter and the current usage parameter value corresponding to the usage parameter, the parameter control submodule does not need to adjust the current usage parameter value, that is, there is no need to call the parameter control interface.

[0254] S1614. The parameter control submodule calls the parameter control interface to adjust the parameter value corresponding to the usage parameter from the current usage parameter value to the second predicted parameter value.

[0255] The parameter control interface predefines a parameter adjustment function. Specifically, after determining that the second predicted parameter value is different from the current parameter value, the parameter control interface can adjust the parameter value through the parameter adjustment function to meet the user's usage requirements.

[0256] In some embodiments, after adjusting the usage parameter value to the second predicted parameter value, the parameter control interface can send the second predicted parameter value and the user's set parameter value to the current user's feature data set, so that the mobile phone can subsequently use the above-mentioned model update submodule to update the target parameter preference model.

[0257] In some embodiments, the above application introduces the setting scheme of the usage parameters (such as brightness, volume, color temperature and other parameters) by taking a mobile phone as an example. In addition, the setting scheme can also be applied to other electronic devices, such as smart homes and smart screens. For example, the setting scheme introduced in this application is applied to smart air conditioners, which can predict the usage parameters (such as temperature, humidity, air volume and other parameters) in different usage scenarios (such as different environments and different time periods) based on a parameter preference model that reflects the user's usage habits. This application does not limit the applied electronic devices and usage parameters.

[0258] In some embodiments, 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 for adjusting usage parameters as described above.

[0259] In some embodiments, the present application provides a computer program product, which, when executed on an electronic device, enables the electronic device to execute the method for adjusting usage parameters as described above.

[0260] Through the description of the above implementation methods, technical personnel in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0261] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0262] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0263] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0264] 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 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, which is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor (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: U disk, mobile hard disk, read only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0265] The above contents are only specific implementation methods of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application shall be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for adjusting a usage parameter, characterized in that: Applied to an electronic device, the electronic device includes parameter preference models corresponding to multiple application scenarios, the parameter preference models are used to predict predicted values ​​of usage parameters in corresponding application scenarios, the method includes: The electronic device receives a first trigger operation on the electronic device; the first trigger operation is used to trigger the electronic device to perform a target operation, and the target operation includes at least one of displaying a first interface, starting or entering a first mode; The electronic device executes the target operation in response to the first trigger operation, inputs the current feature data into the parameter preference model corresponding to the first application scenario, and obtains a predicted value of each usage parameter in at least one usage parameter corresponding to the first application scenario; wherein the first application scenario is the scenario in which the electronic device is located after executing the target operation; the current feature data includes at least one of a user portrait, user behavior data, and a current environmental parameter; and the at least one usage parameter corresponding to the first application scenario refers to a parameter that can be adjusted by the electronic device in the first preset scenario; For each usage parameter, the electronic device adjusts a parameter value of the usage parameter to a predicted value of the usage parameter.

2. The method according to claim 1, characterized in that: The target operation includes displaying a first interface, and inputting the current feature data into a parameter preference model corresponding to the first application scenario includes: When the application scenarios corresponding to the first application scenario and the second interface are different, the electronic device inputs the current feature data into the parameter preference model corresponding to the first application scenario; the second interface is the interface displayed before the electronic device performs the target operation.

3. The method according to claim 1 or 2, characterized in that: The method further comprises: When a new parameter setting result is obtained, the electronic device updates the parameter preference model corresponding to the usage parameter corresponding to the new feature data based on the new feature data in the new parameter setting result and the setting value of the usage parameter corresponding to the new feature data.

4. The method according to claim 3, characterized in that: The electronic device updates the parameter preference model corresponding to the usage parameter corresponding to the new feature data based on the new feature data in the new parameter setting result and the setting value of the usage parameter corresponding to the new feature data, including: The electronic device sets a first weight for the new feature data; wherein the first weight belongs to a preset weight range; The electronic device updates a parameter preference model corresponding to the usage parameter corresponding to the new feature data according to the new feature data with the first weight and the setting value of the usage parameter corresponding to the new feature data.

5. The method according to claim 4, characterized in that The number of the new feature data is multiple, and the earlier the setting time of the usage parameters corresponding to the new feature data is, the smaller the weight of the new feature data is; and / or, if the setting value of the usage parameter corresponding to the new feature data is determined in response to an adjustment operation input by a user, the first weight of the new feature data is a first preset weight, and if the setting value of the usage parameter corresponding to the new feature data is determined based on the parameter preference model, the first weight of the new feature data is a second preset weight, and the second preset weight is less than the first preset weight.

6. The method according to any one of claims 1 to 5, characterized in that The step of inputting the current feature data into a parameter preference model corresponding to the first application scenario includes: When there is no parameter setting result in the pre-stored location in which the first characteristic data is identical to the current characteristic data, the electronic device inputs the current characteristic data into a parameter preference model corresponding to the first application scenario; wherein the parameter setting result includes the historical setting value of the usage parameter and the first characteristic data corresponding to the historical setting value.

7. The method according to any one of claims 1 to 5, characterized in that The method further comprises: In the case that there is a parameter setting result in which the first characteristic data is the same as the current characteristic data in the parameter setting result in the pre-stored position, the predicted value corresponding to the use parameter is a historical setting value of the use parameter in the parameter setting result.

8. The method according to any one of claims 1 to 7, characterized in that The usage parameters corresponding to the application scenario include brightness, volume and color temperature; When the usage parameter is brightness, the current feature data includes user age, user city, user occupation, user education, average daily usage time, application scenario type, and current ambient brightness; or, When the usage parameter is volume, the current feature data includes user age, user city, user occupation, user education, average daily usage time, application scenario type, and current environmental noise; or, When the usage parameter is color temperature, the current feature data includes user age, user city, user occupation, user education level, average daily usage time, application scenario type, current environmental noise, and media playback mode.

9. A model training method, characterized in that: include: Acquire a first training feature set; wherein the first training feature set includes feature data of a user and setting values ​​of usage parameters in an application scenario corresponding to the feature data, wherein the feature data includes a user portrait and / or user behavior data; Based on the first training feature set, the initial model is trained to obtain a parameter preference model corresponding to the application scenario.

10. The method according to claim 9, characterized in that The number of the users is multiple, and the initial model is trained based on the first training feature set to obtain a parameter preference model corresponding to the application scenario, including: Based on the user portraits, the first training feature set is screened to obtain first sub-training feature sets corresponding to each user portrait; Acquire a second training feature set corresponding to various application scenarios from the first sub-training feature set; For each application scenario, obtaining, from the second training feature set, third training feature sets corresponding to various usage parameters corresponding to the application scenario; For each usage parameter, the initial model is trained based on a third training feature set corresponding to the usage parameter to obtain a parameter preference model corresponding to the usage parameter.

11. An electronic device, characterized in that: The electronic device includes a display screen, a memory and one or more processors; the display screen, the memory and the processor are coupled; the display screen is used to display an image generated by the processor, and 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 executes the method described in any one of claims 1 to 10.

12. A computer-readable storage medium, characterized in that: The method comprises computer instructions, which, when executed on an electronic device, cause the electronic device to execute the method according to any one of claims 1 to 10.

Images