Device control method and apparatus, electronic device, and storage medium

Abstract

The application discloses a device control method and device, an electronic device and a storage medium. The device control method is applied to an electronic device, and comprises the following steps: obtaining current use information of the electronic device; inputting the current use information into each preset model in a plurality of preset models to obtain an application output by each preset model, wherein the preset model is trained based on historical use information when the electronic device runs the application multiple times and application information of a next application run after each time of running the application; determining at least one application from the applications output by the plurality of preset models as a target application; and performing a preprocessing operation before the target application runs. The method can better realize prediction of the application and improve efficiency when the application runs.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more specifically, to a device control method, apparatus, electronic device, and storage medium. Background Art

[0002] With the rapid development of electronic devices, various applications can be installed on these devices to meet the diverse needs of users. The configuration of current electronic devices is also constantly improving, and electronic devices can install a large number of applications, for example, dozens or even hundreds of applications. Due to the continuous improvement of user needs, the functions implemented by the applications installed on electronic devices are increasing, and the resources required for their operation are also increasing. This may cause the application to freeze and crash during operation. Summary of the Invention

[0003] In view of the above problems, the present application proposes a device control method, apparatus, electronic device and storage medium.

[0004] In a first aspect, an embodiment of the present application provides a device control method, which is applied to an electronic device, and the method includes: obtaining current usage information of the electronic device; inputting the current usage information into each preset model of a plurality of preset models, and obtaining the application output by each preset model, wherein the preset model is trained based on historical usage information when the electronic device runs an application multiple times, and application information of the next application run after each application is run multiple times; determining at least one application from the applications output by the multiple preset models as a target application; and performing preprocessing operations before the target application runs.

[0005] In a second aspect, an embodiment of the present application provides a device control device, which is applied to an electronic device, and the device includes: an information acquisition module, an application prediction module, an application determination module and a preprocessing module, wherein the information acquisition module is used to obtain current usage information of the electronic device; the application prediction module is used to input the current usage information into each preset model of a plurality of preset models to obtain the application output by each preset model, and the preset model is trained based on historical usage information when the electronic device runs the application multiple times, and application information of the next application run after each application is run multiple times; the application determination module is used to determine at least one application from the applications output by the multiple preset models as a target application; and the preprocessing module is used to perform preprocessing operations before the target application is run.

[0006] In a third aspect, an embodiment of the present application provides an electronic device comprising: one or more processors; a memory; and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the device control method provided in the first aspect above.

[0007] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which program code is stored. The program code can be called by a processor to execute the device control method provided in the first aspect above.

[0008] The solution provided by this application obtains the current usage information of the electronic device, inputs the current usage information into each preset model in a plurality of preset models, obtains the application output by each preset model, and each preset model is trained based on the historical usage information of the electronic device when running applications multiple times, as well as the application information of the next application run after each application run. Then, at least one application is determined as the target application from the applications output by the multiple preset models, and finally preprocessing work is performed before the target application runs. In this way, it is possible to predict the application to be run based on multiple preset models for application prediction, and determine the target application to be run based on the results output by the multiple preset models, and perform preprocessing operations on the predicted target application before running, which can achieve more accurate application prediction and thus improve the efficiency of application runtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A flow chart of a device control method according to an embodiment of the present application is shown.

[0011] Figure 2 A flow chart of a device control method according to another embodiment of the present application is shown.

[0012] Figure 3 A flowchart of step S240 in the device control method provided by another embodiment of the present application is shown.

[0013] Figure 4 Another flow chart of step S240 in the device control method provided in another embodiment of the present application is shown.

[0014] Figure 5A flow chart of a device control method according to another embodiment of the present application is shown.

[0015] Figure 6 A flow chart of a device control method according to another embodiment of the present application is shown.

[0016] Figure 7 A block diagram of a device control apparatus according to an embodiment of the present application is shown.

[0017] Figure 8 4 is a block diagram of an electronic device for executing a device control method according to an embodiment of the present application.

[0018] Figure 9 It is a storage unit in an embodiment of the present application for storing or carrying program codes for implementing the device control method in accordance with an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0020] With the rapid development of technology, electronic devices such as smartphones and tablets are becoming increasingly powerful. People can install various applications on these devices to watch movies, play games, browse the web, and video chat. Furthermore, as user demands continue to increase, applications are becoming increasingly versatile and demanding higher quality. Consequently, the resources consumed by applications during operation are also increasing.

[0021] Therefore, to meet the needs of users to install and run more and more applications, the configuration of electronic devices is also getting higher and higher. However, when electronic devices run many applications, delays and freezes may still occur. For example, the system memory of the electronic device may be insufficient, causing the application to freeze. Among them, system memory is RAM (random-access memory), which is the main area of ​​the system for temporarily storing program instructions and data.

[0022] In traditional technology, when an application is found to be stuck after running, the user can manually optimize the system or application to meet the user's needs for smooth use of the application, or the system can automatically optimize when it detects a stuck phenomenon, but this has affected the user's normal use of the application.

[0023] To address the above issues, the inventors have proposed the device control method, apparatus, electronic device, and storage medium provided in the embodiments of this application. These methods can predict the application to be run based on multiple preset models for application prediction, determine the target application to be run based on the output of the multiple preset models, and perform preprocessing operations on the predicted target application before running, thereby enabling more accurate application prediction and improving the efficiency of application runtime. The specific device control method is described in detail in the subsequent embodiments.

[0024] See also Figure 1 , Figure 1 The flowchart of the device control method provided by one embodiment of the present application is shown. The device control method is used to predict the application to be run based on multiple preset models for application prediction, and to determine the target application to be run based on the results output by the multiple preset models, and to perform preprocessing operations on the predicted target application before running, which can achieve application prediction more accurately, thereby improving the efficiency of the application when it is running. In a specific embodiment, the device control method is applied to Figure 7 The device control device 400 and the electronic device 100 ( Figure 8 ). The following will take electronic devices as an example to illustrate the specific process of this embodiment. Of course, it can be understood that the electronic devices used in this embodiment can be smart phones, tablet computers, smart watches, PC computers, etc., and are not limited here. Figure 1 The process shown in FIG. 1 is described in detail. The device control method may specifically include the following steps:

[0025] Step S110: Acquire current usage information of the electronic device.

[0026] Current electronic devices can be installed with a variety of applications to meet user needs. Although the performance and configuration of electronic devices are constantly improving and advancing, when running applications on electronic devices, delays and lags may still occur due to the large number of currently running application processes or the large number of resources required by the running applications.

[0027] To avoid delays and freezes when applications are started and running, you can predict which applications are about to start so that you can perform corresponding processing and preparation before the application starts. This will prevent subsequent applications from freezing when they are started and running, thereby improving the efficiency of application operation.

[0028] In an embodiment of the present application, an electronic device can obtain current usage information so as to predict the application to be run based on the current usage information of the electronic device. Among them, the current usage information may include application information of the first application currently running on the electronic device and the status information of the electronic device. The status information of the electronic device may include the current time, date, week, charging status, peripheral connection status, network connection status, etc., wherein the charging status indicates whether the electronic device is in a charging or uncharging state, the peripheral connection status indicates whether the electronic device is connected to a peripheral device (such as headphones, speakers, etc.), and the network connection status indicates whether the electronic device is connected to the network. Of course, the specific usage information may not be used as a limitation. For example, it may also include information about the previous one or more applications run by the electronic device before running the first application. The application information may be information that identifies the identity of the application, such as the package name of the application, etc., which is not limited here. It can be understood that when the electronic device runs an application, it is associated with the usage information of the electronic device, so the usage information of the electronic device can be used to predict the application to be run.

[0029] In some embodiments, an electronic device can obtain current usage information of the electronic device to predict the application to be run when the prediction conditions for predicting the application are met. Prediction conditions may include: the currently running application is a set application, the current time is a set time, the currently running application is switched to the background, the currently running application is exited, or the duration of the currently running application reaches a set duration. Of course, the specific prediction conditions are not limited. In other words, the electronic device can obtain current usage information and predict the application only when the prediction conditions are met, without having to perform application predictions all the time, thereby saving power consumption.

[0030] Step S120: Input the current usage information into each preset model of a plurality of preset models to obtain the application output by each preset model, wherein the preset model is pre-trained to determine the application to be run by the electronic device based on the input usage information of the electronic device.

[0031] In an embodiment of the present application, after obtaining the current usage information of the electronic device, the electronic device can input the obtained current usage information into each of the multiple preset models. Each preset model is trained based on the historical usage information of the electronic device when running applications multiple times, as well as the application information of the next application run after each application run.

[0032] In some embodiments, the preset model is obtained through machine learning. Specifically, a training data set is first collected, wherein the attributes or characteristics of one type of data in the training data set are different from those of another type of data. Then, a neural network is trained and modeled using the collected training data set according to a preset algorithm, thereby summarizing patterns based on the training data set to obtain the preset. The training data set may, for example, include historical usage information of multiple applications run by the electronic device in the past, as well as the next application run after each application run. The trained preset model can be used to output an application that is run after the currently running application included in the current usage information based on the current usage information of the electronic device.

[0033] In this embodiment, the multiple different preset models may include models obtained by training with different machine learning algorithms using training samples for the electronic device. The type of machine learning algorithm is not limited, and may be, for example, a neural network, logistic regression, decision tree, random forest, etc. The multiple preset models may also include models obtained by training with the same machine learning algorithm using different training samples for the electronic device. The specific method for obtaining the multiple different preset models is not limited.

[0034] In some embodiments, multiple preset models can be stored locally on the terminal device, and the electronic device can directly read the determined target model locally, which can effectively avoid the influence of network factors on the speed of inputting the preset model into the current usage information, thereby improving the prediction speed and thus improving the user experience; multiple preset models can also be stored on the server, and when the electronic device needs to predict the application to be run, it can send a request to the server to call the target model, thereby reducing the storage space occupied by the electronic device by storing the trained preset model on the server and reducing the impact on the normal operation of the electronic device. This is not limited here.

[0035] In this embodiment, the current usage information of the electronic device can be input into multiple trained preset models, and the application that the electronic device will run output by each preset model in the multiple preset models can be obtained, wherein the application that the electronic device will run is the next application that the preset model predicts the electronic device will run after running the currently running application.

[0036] In some embodiments, the content output by the preset model may be application information of the application to be run. The application information may be information representing the identity of the application, such as the application package name, logo, etc., which is not limited here. Based on the output application information, the electronic device can determine which application to be run as output by the preset model.

[0037] Step S130: Determine at least one application from the applications output by the multiple preset models as a target application.

[0038] In an embodiment of the present application, after obtaining the application output by each preset model in multiple preset models, the electronic device can determine the target application to be run as the final prediction based on all the applications output by the multiple preset models.

[0039] Due to the differences in preset models, some of the applications output in multiple preset models may be the same or different. In some embodiments, when the applications output in multiple preset models include multiple different applications, all applications included in all output results can be determined, and then for each application in all applications, the number of preset models that output each identical application can be determined, and then from the sorting of the corresponding numbers of all applications from large to small, the applications corresponding to the top N numbers are determined, or the applications whose numbers are greater than the specified number are determined, and the determined applications are used as target applications. Wherein, N is a positive integer, and the specific value of N may not be used as a limitation. For example, if N is 1, the application corresponding to the maximum number is selected. The specified number can be determined based on the number of preset models. For example, the specified number can be 80%, 90%, etc. of the number of preset models, which is not limited here.

[0040] In other embodiments, when the applications output from multiple preset models include multiple different applications, all applications included in all output results can be determined. Then, based on the accuracy of the different preset models, the sum of the prediction accuracy of all preset models outputting that application can be calculated for each application. Finally, based on the size of the sum, at least one application can be selected as the target application. For example, based on the size of the sum corresponding to each application in descending order, the applications corresponding to the top N numbers can be determined, or the applications whose sum is greater than a specified size can be determined, and the determined applications can be selected as the target applications. Where N is a positive integer, the specific value of N is not limited. For example, if N is 1, the applications corresponding to the largest number are selected. The specific value of the specified size is not limited. For example, the specified size can be the product of 95% and 90% of the number of all preset models, so that the preset size is related to the number of all preset models and also to the accuracy of the preset models, thereby more accurately predicting the applications that will be run. This is not limited here.

[0041] It should be noted that the number of target applications to be run obtained by the above prediction may be one or more, and the specific number may not be limited.

[0042] Step S140: performing pre-processing operations before the target application is run.

[0043] In an embodiment of the present application, after predicting the target application to be run, the electronic device can perform a preprocessing operation before the target application is run. The preprocessing operation is used to increase the speed of the target application when it is run to avoid lag when starting the target application.

[0044] In some embodiments, the electronic device performs preprocessing operations before running the target application, which may be to optimize the memory resources of the electronic device, for example, to clean up residual processes after the application is run, or to clean up processes with lower priority (i.e., lower importance), or to clean up processes of applications that will not be operated for a long time.

[0045] Furthermore, the electronic device may obtain the memory resources required when the target application is running, and then optimize the memory resources according to the memory resources required when the target application is running.

[0046] As a specific implementation method, the electronic device may compare the memory resources required for the target application to run with the remaining memory resources currently remaining in the electronic device; if the remaining memory resources currently remaining in the electronic device are not less than (greater than or equal to) the memory resources required for the target application to run, then the memory resources may not be optimized; if the remaining memory resources currently remaining in the electronic device are less than the memory resources required for the target application to run, then the memory resources may be released so that the remaining memory resources are not less than the memory resources required for the target application to run.

[0047] As another specific implementation, the electronic device can calculate the difference between the remaining memory resources currently remaining in the electronic device and the memory resources required when the target application is running; if the difference is less than a set threshold, the memory resources can be released so that the difference between the remaining memory resources and the memory resources required when the target application is running is not less than the set threshold; if the difference is not less than the set threshold, the memory resources may not be optimized. In this way, it can be ensured that a certain amount of memory resources can be left after running the target application, avoiding the jamming of the electronic device. The specific set threshold may not be used as a limitation, for example, it can be 100 megabytes (Mb), 200Mb, etc.

[0048] Of course, the specific strategy for optimizing the memory resources of the electronic device may not be limited.

[0049] In other embodiments, the electronic device may also preload the target application. Specifically, the electronic device may preload the resources required by the target application. For example, if the target application is social software, the startup screen, contact list, and message log of the target application may be preloaded. For another example, if the target application is a game, the game loading screen may be preloaded. The specific preloading process and loaded data resources are not limited.

[0050] In some other embodiments, the electronic device performs preprocessing before running the target application, which may be to display a control corresponding to the target application, and the control is used to trigger the entry into the target application. The electronic device displays the control of the target application by displaying a pop-up window of the control corresponding to the target application in the currently displayed interface. The control of the target application may be an application icon corresponding to the target application, or an application link for entering the target application. The specific form of the control is not limited.

[0051] In some other embodiments, the electronic device performs preprocessing before the target application runs, which can be the electronic device allocating network resources to the electronic device. It is understandable that if the target application is an application that needs to use the network, the electronic device can pre-adjust the network resources according to the network resources required for the target application to run, and prepare the network resources allocated to the target application. For example, the electronic device can pre-reduce the network resources allocated to the idle process to free up the network resources allocated to the target application. Of course, the specific method of allocating network resources is not limited.

[0052] In the embodiments of the present application, the specific manner in which the electronic device performs the preprocessing operation is not limited. The above-mentioned methods of optimizing memory resources, preloading target applications, and allocating network resources may be performed one by one, or two or all of them may be selected. Of course, the electronic device may also perform other hardware resource allocations, which are not limited here.

[0053] The device control method provided in the embodiment of the present application obtains the current usage information of the electronic device, inputs the current usage information into each preset model in a plurality of preset models, obtains the application output by each preset model, and each preset model is trained based on the historical usage information of the electronic device when running applications multiple times, as well as the application information of the next application run after each application run. Then, at least one application is determined as the target application from the applications output by the multiple preset models, and finally preprocessing work is performed before the target application runs. In this way, it is possible to predict the application to be run based on multiple preset models for application prediction, and determine the target application to be run based on the results output by the multiple preset models, and perform preprocessing operations on the predicted target application before running, which can more accurately realize application prediction and thus improve the efficiency of application runtime.

[0054] See also Figure 2 , Figure 2 The flow chart of the device control method provided by another embodiment of the present application is shown. The device control method is applied to the above electronic device. Figure 2The process shown in FIG. 1 is described in detail. The device control method may specifically include the following steps:

[0055] Step S210: Acquire current usage information of the electronic device.

[0056] Step S220: Input the current usage information into each preset model of multiple preset models to obtain the application output by each preset model, where the preset model is trained based on the historical usage information when the electronic device runs the application multiple times, and the application information of the next application run after each application is run multiple times.

[0057] In the embodiment of the present application, step S210 and step S220 can refer to the contents of the aforementioned embodiment and will not be repeated here.

[0058] Step S230: Obtain the prediction accuracy of each preset model.

[0059] In an embodiment of the present application, after obtaining the application output by each of the multiple preset models, the electronic device may determine the target application to be run as a prediction from all the output results of the multiple preset models based on the prediction accuracy of each preset model. The prediction accuracy is the accuracy of the application output by each preset model based on the input usage information of the electronic device.

[0060] In the embodiment of the present application, the prediction accuracy of each preset model can be obtained by:

[0061] Obtain a test data set, wherein the test data set includes historical usage information of the electronic device when running an application multiple times, and application information of the next application run after each application is run multiple times; based on the test data set, test each preset model of the multiple preset models to obtain the prediction accuracy of each preset model.

[0062] In some embodiments, the test data set may be a collection of partial sample data from a plurality of sample data obtained by the electronic device before training each preset model. The partial sample data includes historical usage information from multiple runs of an application, as well as application information for the next application to be run after the currently running application.

[0063] In some embodiments, when testing a preset model, the electronic device may input usage information from the test data into the preset model, obtain application information output by the preset model, and compare the output application information with the application information of the next application in the test data. If the two are the same, it indicates that the preset model's prediction is accurate; if they are different, it indicates that the preset model's prediction is inaccurate. After testing the preset model using multiple test data, the ratio of the number of times the preset model's prediction is accurate to the total number of tests can be calculated to obtain the accuracy of the preset model.

[0064] In some embodiments, the prediction accuracy of the above preset model can be pre-calculated by the electronic device, and when the electronic device needs to use the prediction accuracy of the preset model, it can directly read it from the local computer. The prediction accuracy of the above preset model can also be calculated by a server, and when the electronic device needs to use the prediction accuracy of the preset model, it can obtain the prediction accuracy from the server.

[0065] Step S240: determining at least one application from the applications output by the plurality of preset models as a target application according to the prediction accuracy of each preset model.

[0066] In an embodiment of the present application, after obtaining the prediction accuracy of each preset model, the electronic device can determine at least one application as a target application from all applications output by multiple preset models based on the accuracy of different preset models.

[0067] As an implementation, see Figure 3 , based on the prediction accuracy of each preset model, determining at least one application from the applications output by the multiple preset models as a target application, which may include:

[0068] Step S241A: According to the applications output by the multiple preset models and the prediction accuracy of each preset model, a first sum value of the accuracy of the preset models outputting the same application is obtained.

[0069] It is understandable that due to differences in preset models, some of the applications output by multiple preset models may be the same or different. When the applications output by multiple preset models include multiple different applications, the electronic device may determine all applications included in all output results, and then calculate, for each application among all applications, a first sum of the prediction accuracy rates of all preset models outputting that application.

[0070] For example, multiple preset models include: Model 1, Model 2, Model 3, Model 4, and Model 5. The prediction accuracies of Model 1, Model 2, Model 3, Model 4, and Model 5 are 90%, 92%, 91%, 95%, and 93%, respectively. The output of Model 1 includes Application 1, the output of Model 2 includes Application 2, the output of Model 3 includes Application 1, the output of Model 4 includes Application 3, and the output of Model 5 includes Application 2. Therefore, all output applications include Application 1, Application 2, and Application 3. For Application 1, the models output for Application 1 are Model 1 and Model 3. Therefore, the prediction accuracies of Model 1 and Model 3 are added together, resulting in a first sum of 1.81. For Application 2, the models output for Application 2 are Model 2 and Model 5. Therefore, the prediction accuracies of Model 2 and Model 5 are added together, resulting in a first sum of 1.85. For Application 3, the model output for Application 3 is only Model 4, resulting in a first sum of 0.95.

[0071] Step S242A: According to a first sum of the accuracy rates corresponding to each application in the applications output by the multiple preset models, at least one application whose first sum meets a first specified condition is selected as a target application.

[0072] In some embodiments, after calculating the first sum corresponding to each application, the electronic device may select one or more applications whose first sums satisfy the first specified condition as target applications based on the first sum corresponding to each application. Because the prediction results of each preset model are referenced and the prediction results are filtered based on the prediction accuracy of each preset model, application predictions can be more accurate.

[0073] In some embodiments, the first specified condition may include: the first sum of the accuracy is greater than a specified threshold; or the sum of the accuracy is in the top N positions in the order of all first sums from large to small, where N is a positive integer. It is understandable that the electronic device may select an application whose first sum is greater than the specified threshold as the target application, or may select an application whose first sum is in the top N positions in the order of all first sums from large to small as the target application. The specific value of the specified threshold may not be used as a limitation. For example, the specified threshold may be the product of 95% and 90% of the number of all preset models, so that the specified threshold is related to the number of all preset models and also to the accuracy of the preset models, thereby more accurately predicting the application to be run, which is not limited here. N is a positive integer, and the specific value of N may not be used as a limitation. For example, N may be 1, that is, the application corresponding to the largest first sum is selected as the target application.

[0074] As another embodiment, the output of each preset model also includes a probability value corresponding to an application. It will be appreciated that each preset model outputs a predicted application to be run and a probability value for the application to be run based on the input mobile terminal usage information. For example, if the output of a preset model includes application 1 and a probability value of 90%, it means that there is a 90% probability that the mobile terminal will run application 1.

[0075] In this implementation, see Figure 4 , based on the prediction accuracy of each preset model, determining at least one application from the applications output by the multiple preset models as a target application, which may include:

[0076] Step S241B: Obtain the product of the probability value corresponding to the application output by each preset model and the prediction accuracy of each preset model.

[0077] In this embodiment, when the results output by the preset models include probability values ​​corresponding to applications, the electronic device may also calculate the product of the probability value of the application output by each preset model and its prediction accuracy.

[0078] For example, multiple preset models include: Model 1, Model 2, Model 3, Model 4 and Model 5. The prediction accuracy of Model 1, Model 2, Model 3, Model 4 and Model 5 are: 90%, 92%, 91%, 95% and 93% respectively. The results output by Model 1 include: Application 1 and a probability value of 90%, the results output by Model 2 include: Application 2 and a probability value of 95, the results output by Model 3 include: Application 1 and a probability value of 100, the results output by Model 4 include: Application 3 and a probability value of 98, and the results output by Model 5 include: Application 2 and a probability value of 85%. For model 1, the product of the output probability value of application 1 and its prediction accuracy is: 90%*90%=0.81; for model 2, the product of the output probability value of application 2 and its prediction accuracy is: 92%*95%=0.874; for model 3, the product of the output probability value of application 1 and its prediction accuracy is: 100%*91%=0.91; for model 4, the product of the output probability value of application 3 and its prediction accuracy is: 98%*95%=0.931; for model 5, the product of the output probability value of application 2 and its prediction accuracy is: 85%*93%=0.7905.

[0079] Step S242B: According to the product corresponding to the application output by each preset model, obtain a second sum value of the products corresponding to the same application output by the multiple preset models.

[0080] In some embodiments, the electronic device obtains the sum of the products of the probability value of each application output by each preset model and its prediction accuracy, which is equivalent to obtaining the product corresponding to each output application. The electronic device can determine all applications included in all output results, and then calculate and output a second sum of the products calculated above for each application in all applications.

[0081] For example, in the above example, all the output applications include application 1, application 2, and application 3. For application 1, the products corresponding to application 1 include 0.81 and 0.91, and the second sum is 0.81+0.91=; for 2, the products corresponding to application 2 include 0.874 and 0.7905, and the second sum is 0.874+0.7905=; for application 3, the products corresponding to application 3 include 0.931, and the second sum is 0.931.

[0082] Step S243B: According to the second sum value corresponding to each application in the applications output by the multiple preset models, at least one application whose second sum value meets a second specified condition is selected as a target application.

[0083] In some embodiments, after calculating the second sum corresponding to each application, the electronic device may select one or more applications whose second sums satisfy the second specified condition as target applications based on the second sum corresponding to each application. Because the prediction results of each preset model are referenced and the prediction results are screened based on the prediction accuracy of each preset model, application predictions can be more accurate.

[0084] In some embodiments, the second specified condition may include: the second sum is greater than a set threshold; or the second sum is in the top N positions in the order of all second sums from large to small, where N is a positive integer. It is understandable that the electronic device may select an application whose second sum is greater than the set threshold as the target application, or may select an application whose second sum is in the top N positions in the order of all second sums from large to small as the target application. The specific value of the set threshold may not be used as a limitation and is not limited here. N is a positive integer, and the specific value of N may not be used as a limitation. For example, N may be 1, which means that the application corresponding to the largest second sum is selected as the target application.

[0085] In the embodiments of the present application, the number of applications determined may be one or more, depending on the prediction accuracy of each preset model. For example, if there are multiple applications whose first sum is greater than a specified threshold, then multiple applications can be determined from all applications. If there are multiple applications determined, all of them can be used as target applications.

[0086] Step S250: performing pre-processing operations before running the target application.

[0087] In the embodiment of the present application, step S250 can refer to the content of the aforementioned embodiment and will not be repeated here.

[0088] The device control method provided in the embodiment of the present application obtains the current usage information of the electronic device, inputs the current usage information into each preset model of multiple preset models, obtains the application output by each preset model, and each preset model is pre-trained to determine the application that the electronic device will run based on the input usage information of the electronic device, and then determines at least one application as the target application from the applications output by the multiple preset models based on the prediction accuracy of each preset model, and finally performs preprocessing work before the target application runs. Since predictions are made through multiple preset models and the target application that is ultimately used as the prediction result is determined with reference to the accuracy of the preset models, it is possible to achieve more accurate application predictions, thereby improving the efficiency of the application during runtime.

[0089] See also Figure 5 , Figure 5 A flow chart of a device control method according to another embodiment of the present application is shown. The device control method is applied to the above-mentioned electronic device. Figure 5 The process shown in FIG. 1 is described in detail. The device control method may specifically include the following steps:

[0090] Step S310: Acquire a training data set, where the training data set includes historical usage information of the electronic device when running applications multiple times, and application information of the next application to be run after each application is run multiple times.

[0091] Regarding the preset model in the aforementioned embodiment, the embodiments of the present application also include a training method for the preset model. It is worth noting that the training of the preset model can be performed in advance based on the acquired training data set. Subsequently, each time a prediction is made for an application to be run, a prediction can be made based on the preset model, without the need to train the preset model each time a prediction is made for an application to be run.

[0092] In some embodiments, an electronic device may collect a usage record of an application of the electronic device within a period of time (e.g., 30 days). The usage record includes historical usage information when the application is running, as well as application information of the next application to be run after the application is run. Specifically, the usage record includes x and y, where x represents historical usage information when the application is running, and is defined as x=[t, d, A c ,A1,A2,c,e,n],y=A nAmong them, t represents the current time, and its value range is [0,23], with one hour as one unit; d represents the day of the week, and its value range is [0,6], with a total of 7 values, namely Monday to Sunday; A c Indicates the currently running application, A1 and A2 respectively indicate the two applications that were run before the currently running application; c indicates the current charging status, with a value of {0,1}, where 0 indicates not charging and 1 indicates charging; e indicates the current headphone status, with a value of {0,1}, where 0 indicates not connected to the headphone and 1 indicates connected to the headphone; n indicates the current network status, with a value of {0,1}, where 0 indicates not connected to the network and 1 indicates connected to the network; A n Indicates the next application to be used after running the currently running application.

[0093] Furthermore, the application usage records collected above can be used as sample data, where x can be a label and y is the label result. In other words, each usage information is labeled as the application information of the next application, and each set of sample data includes an input sample x and an output sample y.

[0094] Specifically, the training data used to train the preset model can be a portion of the sample data above, thereby constituting a training data set; and the test data can be another portion of the sample data above, thereby constituting a test data set. For example, the training data can be 70% of the sample data, and the test data can be 30% of the sample data, and the specific proportions are not limited.

[0095] Step S320: Based on the training data set, each usage information is used as input data, and the application information of the next application corresponding to each usage information is used as output data, and multiple machine learning algorithms are used for training to obtain multiple different preset models. Each preset model can determine the application that the electronic device will run based on the input usage information of the electronic device.

[0096] In an embodiment of the present application, different machine learning algorithms can be used for training the training data set of the electronic device to obtain a plurality of different preset models. Among them, the machine learning algorithms used may include: neural network, long short-term memory (LSTM) network, threshold recurrent unit, simple recurrent unit, autoencoder, decision tree, random forest, feature mean classification, classification regression tree, hidden Markov, K-nearest neighbor (k-NearestNeighbor, KNN) algorithm, logistic regression model, Bayesian model, Gaussian model and KL divergence (Kullback–Leibler divergence), etc. The specific machine learning algorithm may not be used as a limitation. For example, when using the logistic regression algorithm for training, the loss function can be Among them, m is the total number of training data, y i represents the true label of the i-th data item, and y′ represents the label calculated by the logistic regression algorithm. For another example, when using the random forest algorithm for application prediction learning, the number of decision trees should be moderate, such as N = 50. During the tree generation process, the training sample is divided by calculating the information entropy, and the formula is: Among them, n represents the number of types of labels in the training set data, p i Represents the probability that the training sample label belongs to the i-th category.

[0097] The following takes a neural network as an example to illustrate how to train an initial model based on a training data set.

[0098] The neural network uses the usage information x of electronic devices from a set of data in the training dataset as input data, and the application information y of the next application run by the electronic device after the current application is labeled in the set of data as output data. Neurons in the input layer are fully connected to neurons in the hidden layer, and neurons in the hidden layer are fully connected to neurons in the output layer, effectively extracting latent features of varying granularity. Multiple hidden layers are also possible, enabling better fitting of nonlinear relationships and increasing the accuracy of the trained pre-set model.

[0099] It is understood that the training process for the preset model can be completed by the electronic device or not. When the training process is not completed by the electronic device, the electronic device can only act as a direct user or an indirect user. That is, the electronic device can send the current usage information of the electronic device to the server storing the preset model and obtain the second application predicted to be run from the server.

[0100] In some embodiments, the trained preset model can be stored locally in the electronic device, or in a server that is communicatively connected to the electronic device. Storing the preset model in the server can reduce the storage space occupied by the electronic device and improve the operating efficiency of the electronic device.

[0101] In some implementations, the preset model may periodically or irregularly acquire new training data to train and update the preset model.

[0102] In some embodiments, the electronic device can also test multiple trained models based on the above test data set and calculate the prediction accuracy of different preset models. M Indicates the prediction accuracy of the preset model M.

[0103] Specifically, when obtaining the prediction accuracy of each preset model, the electronic device can use the usage information of multiple test data sets to input into the preset model based on the test data set, obtain the application information output by the preset model, and compare the output application information with the application information of the next application in the test data. If the two are the same, it indicates that the preset model's prediction is accurate; if the two are different, it indicates that the preset model's prediction is inaccurate. After testing the preset model using multiple test data sets, the ratio of the number of times the preset model's prediction is accurate to the total number of times it is tested can be obtained, thereby obtaining the prediction accuracy of the preset model.

[0104] Step S330: Acquire current usage information of the electronic device.

[0105] Step S340: Input the current usage information into each preset model of a plurality of preset models to obtain the application output by each preset model, wherein the preset model is pre-trained to determine the application to be run by the electronic device based on the input usage information of the electronic device.

[0106] Step S350: Determine at least one application from the applications output by the multiple preset models as a target application.

[0107] Step S360: performing pre-processing operations before running the target application.

[0108] In the embodiment of the present application, steps S330 to S360 can refer to the contents of the aforementioned embodiment and will not be repeated here.

[0109] The device control method provided in the embodiment of the present application provides a method for training a preset model, and the initial model is trained by training data marked with application information of the next application that the electronic device will run after running the currently running application, thereby obtaining a preset model. The preset model can be used to output application information of the next application that the electronic device will run based on the input usage information of the electronic device. It is possible to predict the application to be run based on multiple preset models for application prediction, and determine the target application to be run based on the results output by multiple preset models, and perform preprocessing operations on the predicted target application before running, which can achieve more accurate application prediction and thus improve the efficiency of application runtime.

[0110] See also Figure 6 , Figure 6 A flow chart of a device control method according to another embodiment of the present application is shown. The device control method is applied to the above-mentioned electronic device. Figure 6 The process shown in FIG. 1 is described in detail. The device control method may specifically include the following steps:

[0111] Step S410: Acquire current usage information of the electronic device.

[0112] Step S420: Input the current usage information into each preset model of multiple preset models to obtain the application output by each preset model, where the preset model is trained based on the historical usage information when the electronic device runs the application multiple times, and the application information of the next application run after each application is run multiple times.

[0113] In the embodiment of the present application, step S410 and step S420 can refer to the contents of the aforementioned embodiment and will not be repeated here.

[0114] Step S430: Determine the number of preset models outputted by each application in all the applications according to all the applications outputted by the multiple preset models.

[0115] It is understandable that due to differences in preset models, some of the applications output by multiple preset models may be the same or different. When the applications output by multiple preset models include multiple different applications, all applications included in all output results can be determined, and then, for each application in all applications, the number of preset models that output each identical application can be determined.

[0116] Step S440: selecting at least one application whose quantity meets the specified quantity condition from the applications output by the plurality of preset models as a target application.

[0117] In some implementations, after obtaining the number of preset models corresponding to each application, the electronic device can select at least one application from all applications that meets a specified number requirement as the target application. This allows the predicted results of the applications to be run based on multiple preset models to be jointly determined as the target application.

[0118] In some embodiments, the specified quantity condition may include: determining the top N numbers in the sorting of the numbers corresponding to all applications from large to small, or determining the number greater than the specified number. It is understandable that the electronic device can determine the applications corresponding to the top N numbers from the sorting of the numbers corresponding to all applications from large to small based on the above-determined number corresponding to each application, or determine the applications whose number is greater than the specified number, and use the determined applications as target applications. Wherein, N is a positive integer, and the specific value of N may not be used as a limitation. For example, if N is 1, the application corresponding to the maximum number is selected. The specified number can be determined based on the number of preset models. For example, the specified number can be 80%, 90%, etc. of the number of preset models, which is not limited here.

[0119] Step S450: performing pre-processing operations before the target application is run.

[0120] In the embodiment of the present application, step S450 can refer to the content of the aforementioned embodiment and will not be repeated here.

[0121] The device control method provided in the embodiment of the present application obtains the current usage information of the electronic device, inputs the current usage information into each preset model in a plurality of preset models, obtains the application output by each preset model, and each preset model is pre-trained to determine the application that the electronic device will run based on the input usage information of the electronic device, and then selects at least one application that meets the specified quantity conditions from the applications output by the plurality of preset models as the target application based on the number of the same application output by the plurality of preset models, and finally performs preprocessing work before the target application runs. Since predictions are made through a plurality of preset models and the target application that is ultimately used as the prediction result is determined with reference to the accuracy of the preset models, it is possible to achieve more accurate application predictions, thereby improving the efficiency of the application during runtime.

[0122] See also Figure 7, which shows a structural block diagram of a device control device 400 provided in an embodiment of the present application. The device control device 400 applies the above-mentioned electronic device, and the device control device 400 includes: an information acquisition module 410, an application prediction module 420, an application determination module 430 and a pre-processing module 440. Among them, the information acquisition module 410 is used to obtain the current usage information of the electronic device; the application prediction module 420 is used to input the current usage information into each preset model of a plurality of preset models, and obtain the application output by each preset model, and the preset model is trained based on the historical usage information when the electronic device runs the application multiple times, and the application information of the next application run after each application is run multiple times; the application determination module 430 is used to determine at least one application from the applications output by the multiple preset models as the target application; the pre-processing module 440 is used to perform pre-processing operations before the target application is run.

[0123] In some embodiments, the application determination module 430 may include: an accuracy acquisition unit for acquiring the prediction accuracy of each preset model; an application selection unit for determining at least one application from the applications output by the multiple preset models as a target application based on the prediction accuracy of each preset model.

[0124] As an embodiment, the application selection unit may include: a first calculation subunit, used to obtain a first sum of the accuracy of the preset models outputting the same application based on the applications output by the multiple preset models and the prediction accuracy of each preset model; a first selection subunit, used to select at least one application whose first sum meets a first specified condition as a target application based on the first sum of the accuracy corresponding to each application in the applications output by the multiple preset models.

[0125] In this embodiment, the first specified condition includes: the first sum of the accuracy rates is greater than a specified threshold; or the sum of the accuracy rates is in the top N positions of all first sums sorted from large to small, where N is a positive integer.

[0126] As another embodiment, the result output by each preset model also includes a probability value corresponding to the application. The application selection unit may include: a second calculation subunit, which is used to obtain the product of the probability value corresponding to the application output by each preset model and the prediction accuracy of each preset model before obtaining the prediction accuracy of each preset model in the multiple preset models; a third calculation subunit, which is used to obtain a second sum value of the products corresponding to the same application output by the multiple preset models based on the product corresponding to the application output by each preset model; and a second selection subunit, which is used to select at least one application whose second sum value meets the second specified condition as the target application based on the second sum value corresponding to each application in the applications output by the multiple preset models.

[0127] In some embodiments, the device control apparatus 400 may further include: a test data acquisition module and a model testing module. The test data acquisition module is configured to acquire a test data set, the test data set including historical usage information of the electronic device when multiple applications are run, and application information of the next application run after each of the multiple application runs; and the model testing module is configured to test each of the multiple preset models based on the test data set to obtain a prediction accuracy rate for each preset model.

[0128] In other embodiments, the application determination module 430 may include: a quantity determination unit for determining the number of preset models corresponding to each application among all the applications output by the multiple preset models; and a selection execution unit for selecting at least one application whose number meets a specified quantity condition from the applications output by the multiple preset models as a target application.

[0129] In some embodiments, the device control device 400 may also include a training data acquisition module and a model training module. The training data acquisition module is used to obtain a training data set before inputting the current usage information into each preset model of the multiple preset models and obtaining the application output by each preset model. The training data set includes historical usage information of the electronic device when running applications multiple times, and application information of the next application run after each application run when running applications multiple times; the model training module is used to use a plurality of machine learning algorithms to train based on the training data set, using each usage information as input data and the application information of the next application corresponding to each usage information as output data to obtain a plurality of different preset models. Each of the preset models can determine the application to be run by the electronic device based on the input usage information of the electronic device.

[0130] In some embodiments, the pre-processing module 440 performs pre-processing operations before the target application is executed, including one or more of the following methods:

[0131] Optimizing memory resources of the electronic device according to memory resources required when the target application is running;

[0132] preloading the target application;

[0133] Displaying a control corresponding to the target application, wherein the control is used to jump to the target application;

[0134] The network resources of the electronic device are allocated according to the network resources required when the target application is running.

[0135] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described devices and modules can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0136] In several embodiments provided in this application, the coupling between modules may be electrical, mechanical or other forms of coupling.

[0137] In addition, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The above-mentioned integrated modules may be implemented in the form of hardware or software functional modules.

[0138] In summary, the solution provided by this application obtains the current usage information of the electronic device, inputs the current usage information into each preset model in a plurality of preset models, obtains the application output by each preset model, and each preset model is trained based on the historical usage information of the electronic device when running applications multiple times, as well as the application information of the next application run after each application is run. Then, at least one application is determined as the target application from the applications output by the multiple preset models, and finally, preprocessing work is performed before the target application runs. In this way, it is possible to predict the application to be run based on a plurality of preset models for application prediction, and determine the target application to be run based on the results output by the multiple preset models, and perform preprocessing operations on the predicted target application before running, which can achieve more accurate application prediction and thus improve the efficiency of application runtime.

[0139] Please refer to Figure 8, which shows a structural block diagram of an electronic device provided in an embodiment of the present application. The electronic device 100 can be an electronic device capable of running applications, such as a smartphone, a tablet computer, a smart watch, a PC, etc. The electronic device 100 in the present application may include one or more of the following components: a processor 110, a memory 120, and one or more applications, wherein the one or more applications may be stored in the memory 120 and configured to be executed by one or more processors 110, and the one or more programs are configured to execute the method described in the aforementioned method embodiment.

[0140] The processor 110 may include one or more processing cores. The processor 110 utilizes various interfaces and circuits to connect various components within the electronic device 100. It executes instructions, programs, code sets, or instruction sets stored in the memory 120, as well as accesses data stored in the memory 120, to perform various functions of the electronic device 100 and process data. Optionally, the processor 110 may be implemented using at least one of the following hardware forms: a digital signal processing (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). The processor 110 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may not be integrated into the processor 110 and may be implemented separately via a communications chip.

[0141] The memory 120 may include a random access memory (RAM) or a read-only memory (ROM). The memory 120 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 120 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the following various method embodiments, etc. The data storage area may also store data created by the electronic device 100 during use (such as a phone book, audio and video data, chat history data), etc.

[0142] Please refer to Figure 9 , which shows a block diagram of a computer-readable storage medium provided in an embodiment of the present application. The computer-readable medium 800 stores program code, which can be called by a processor to execute the method described in the above method embodiment.

[0143] The computer-readable storage medium 800 can be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer-readable storage medium 800 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 800 has storage space for program code 810 for executing any of the method steps described above. These program codes can be read from or written to one or more computer program products. The program code 810 can be compressed, for example, in a suitable form.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A device control method, characterized in that: Applied to electronic equipment, the method includes: Acquiring current usage information of the electronic device; Inputting the current usage information into each of a plurality of preset models to obtain an application and a probability value corresponding to the application output by each preset model, wherein the preset model is trained based on historical usage information of the electronic device when the electronic device runs applications multiple times and application information of a next application run after each of the multiple application runs, and the plurality of preset models are trained using different machine learning algorithms; Obtaining the prediction accuracy of each preset model; Obtaining the product of the probability value corresponding to the application output by each preset model and the prediction accuracy of each preset model; Obtaining, according to the product corresponding to the application output by each preset model, a second sum of the products corresponding to the same application output by the multiple preset models; Selecting, based on the second sum value corresponding to each application in the applications output by the multiple preset models, at least one application whose second sum value satisfies a second specified condition as a target application, the second specified condition including: the second sum value being greater than a set threshold, or the second sum value being in the top N positions of all second sum values ​​sorted from largest to smallest, where N is a positive integer; Perform pre-processing operations before the target application is run.

2. The method according to claim 1, characterized in that Determining at least one application as a target application from the applications output by the plurality of preset models according to the prediction accuracy of each preset model includes: Obtaining a first sum of the accuracy rates of the preset models outputting the same application based on the applications output by the multiple preset models and the prediction accuracy rate of each preset model; According to a first sum value of the accuracy corresponding to each application in the applications output by the multiple preset models, at least one application whose first sum value meets a first specified condition is selected as a target application.

3. The method according to claim 2, characterized in that The first specified condition includes: The first sum of the accuracy is greater than the specified threshold; or The sum of the accuracy rates is in the top N positions of all first sums sorted from largest to smallest, where N is a positive integer.

4. The method according to claim 1, wherein Before obtaining the prediction accuracy of each of the multiple preset models, the method further includes: Acquire a test data set, the test data set including historical usage information of the electronic device when the electronic device runs the application multiple times, and application information of the next application run after each application run during the multiple application runs; Each of the plurality of preset models is tested according to the test data set to obtain a prediction accuracy of each preset model.

5. The method according to claim 1, wherein Determining at least one application from the applications output by the plurality of preset models as a target application includes: Determining, based on all applications output by the multiple preset models, the number of preset models outputting the application corresponding to each application in all the applications; At least one application whose quantity meets a specified quantity condition is selected from the applications output by the plurality of preset models as a target application.

6. The method according to claim 1, characterized in that Before inputting the current usage information into each of the plurality of preset models and obtaining the application output by each preset model and the probability value corresponding to the application, the method further includes: Acquire a training data set, the training data set including historical usage information of the electronic device when the electronic device runs the application multiple times, and application information of the next application run after each application run during the multiple application runs; According to the training data set, each usage information is used as input data, and the application information of the next application corresponding to each usage information is used as output data, and multiple machine learning algorithms are used for training to obtain multiple different preset models. Each preset model can determine the application that the electronic device will run based on the input usage information of the electronic device.

7. The method according to any one of claims 1 to 6, characterized in that The pre-processing operation before the target application is executed includes one or more of the following methods: Optimizing memory resources of the electronic device according to memory resources required when the target application is running; preloading the target application; Displaying a control corresponding to the target application, wherein the control is used to jump to the target application; The network resources of the electronic device are allocated according to the network resources required when the target application is running.

8. A device control device, characterized in that: Applied to electronic equipment, the device includes: an information acquisition module, an application prediction module, an application determination module and a preprocessing module, wherein: The information acquisition module is used to acquire current usage information of the electronic device; The application prediction module is configured to input the current usage information into each of a plurality of preset models, and obtain an application output by each preset model and a probability value corresponding to the application, wherein the preset model is trained based on historical usage information of the electronic device when the electronic device runs applications multiple times, and application information of the next application to be run after each of the multiple application runs, and the plurality of preset models are trained using different machine learning algorithms; The application determination module is configured to obtain a prediction accuracy rate of each preset model; obtain a product of a probability value corresponding to an application output by each preset model and the prediction accuracy rate of each preset model; obtain, based on the product corresponding to the application output by each preset model, a second sum of the products corresponding to the same application output by the multiple preset models; and select, based on the second sum corresponding to each application among the applications output by the multiple preset models, at least one application whose second sum satisfies a second specified condition as a target application, the second specified condition including: the second sum being greater than a set threshold, or the second sum being in the top N positions of all second sums sorted from largest to smallest, where N is a positive integer; The preprocessing module is used to perform preprocessing operations before the target application is run.

9. An electronic device, characterized in that: include: one or more processors; Memory; One or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, which can be called by a processor to execute the method according to any one of claims 1 to 7.

Images