Application control method and related device

Abstract

The embodiment of the invention provides an application control method and a related device, and relates to the technical field of terminals. The method is applied to the electronic equipment, and comprises the following steps: in a first time period of the tth day, M applications run in the background of the electronic equipment, and searching and killing the M applications by adopting a first application list; in a second time period of the tth day, N applications run in the background of the electronic equipment, a second application list is adopted to search and kill the N applications, the number of the applications in the second application list is different from the number of the applications in the first application list, the first time period is different from the second time period, and the number of the applications in the first application list is different from the number of the applications in the second application list. M and N are both integers. Therefore, the probability that the applications which are not used by the user are searched and killed can be improved, and the power consumption of the electronic equipment can be reduced.

Description

Technical Field

[0001] This application relates to the technical field of terminals, and in particular, to an application (APP) control method and related devices. Background Art

[0002] Multiple applications can be installed on some electronic devices to provide users with various application services. Users can open multiple applications. One of the opened multiple applications can run in the foreground, and the remaining applications among the opened multiple applications can run in the background.

[0003] To save power consumption, an electronic device can kill some applications running in the background. The killed applications will terminate running, and the applications that are not killed are kept-alive applications, and the kept-alive applications keep running.

[0004] However, there is a problem that the battery life of the electronic device is relatively short in the above implementation. Summary of the Invention

[0005] Embodiments of this application provide an application control method and related devices, which are applied to the technical field of terminals. In response to a recommendation operation, an electronic device uses the application list corresponding to the time period to which the recommendation operation time belongs to kill the applications running in the background. Among them, the number of applications in the application lists corresponding to different time periods is different. In this way, the probability that applications not used by the user are killed can be increased, and thus the power consumption of the electronic device can be reduced.

[0006] In a first aspect, an embodiment of this application proposes an application control method, which is applied to an electronic device. The method includes: in a first time period of the t-th day, M applications are running in the background of the electronic device, and the M applications are killed using a first application list. In a second time period of the t-th day, N applications are running in the background of the electronic device, and the N applications are killed using a second application list. The number of applications in the second application list is different from the number of applications in the first application list. The first time period is different from the second time period, and both M and N are integers.

[0007] In this way, it is possible to obtain application lists corresponding to different time periods, and the number of applications in the application lists corresponding to different time periods is different. Using the corresponding application lists to kill the applications running in the background in different time periods can not only increase the probability that applications not used by the user are killed, thereby reducing the power consumption of the electronic device, but also reduce the probability of occurrence of applications used by the user being killed, improving the user experience.

[0008] In a possible implementation, the first application list is obtained in the following manner: when reaching the first time period of the t-th day, a first predicted application list for the first time period of the t-th day is predicted using a preset model. A target number of applications are selected from the first predicted application list to obtain the first application list. The preset model is used to output the first predicted application list when inputting the application usage data in the first time period in the L days before the t-th day and the application usage data of the R times of using applications before the first recommendation operation time. The first recommendation operation time is the time when reaching the first time period of the t-th day. The application usage data includes the start time when the application runs in the foreground and the application identifier. The target number is related to one or more of the following: the number of applications used in the first time period in the L days before the t-th day, the number of applications used in the first time period in the K days before the t-th day, the number of applications used in Q time periods before the first time period of the t-th day, and the Q time periods do not belong to the first time period.

[0009] In this way, the preset model takes into account the order of magnitude of the possible application usage probabilities of the user in the first time period. The first application list formed by selecting a target number of applications from the first predicted application list output by the preset model also takes into account the possible number of applications used by the user in the first time period, improving the accuracy of the first application list, and can increase the probability that applications not used by the user in the first time period are killed to reduce the power consumption of the electronic device.

[0010] In a possible implementation, the target number is positively correlated with one or more of the following: the number of applications used in the first time period in the L days before the t-th day, the number of applications used in the first time period in the K days before the t-th day, the number of applications used in Q time periods before the first time period of the t-th day.

[0011] In this way, predicting the target number of the first time period based on the number of applications used in the first time period in the L days before the t-th day, the number of applications used in the first time period in the K days before the t-th day, and the number of applications used in Q time periods before the first time period of the t-th day can improve the prediction accuracy of the target number of the first time period.

[0012] In a possible implementation, when the t-th day is a working day, the target quantity is related to one or more of the following: the application usage quantity in the first time period of working days within the L days before the t-th day, the application usage quantity in the first time period during the K working days before the t-th day, and the application usage quantity in the Q time periods belonging to working days before the first time period of the t-th day. When the t-th day is a holiday, the target quantity is related to one or more of the following: the application usage quantity in the first time period of holidays within the L days before the t-th day, the application usage quantity in the first time period during the K holidays before the t-th day, and the application usage quantity in the Q time periods belonging to holidays before the first time period of the t-th day.

[0013] In this way, since the habits of users using applications on working days may be different from those on holidays, differentiating the date type of the t-th day can further improve the prediction accuracy of the target quantity in the first time period.

[0014] In a possible implementation, the first time period of the t-th day is the s-th time period of the t-th day. The target quantity N ts satisfies the following formula:

[0015]

[0016] where, N ti,s is the application usage quantity in the s-th time period of the i-th target day before the t-th day. There are i - 1 target days between the i-th target day before the t-th day and the t-th day. N t,sj is the application usage quantity in the j-th time period before the s-th time period of the t-th day. There are j - 1 time periods belonging to target days between the j-th time period before the s-th time period of the t-th day and the s-th time period of the t-th day. γ1 is a weight coefficient, γ2 is a weight coefficient, γ3 is a weight coefficient, σ tsL 2 is the variance of the application usage quantity in the s-th time period of target days within the L days before the t-th day. σ tsL 2 *γ3 represents the weighting of the variance σ tsL 2 . The target days satisfy: being working days when the t-th day is a working day and being holidays when the t-th day is a holiday. * is the multiplication sign.

[0017] In this way, based on the application usage situation in the s-th time period during the K target days before the t-th day (such as ), the application usage situation in the Q time periods belonging to target days before the s-th time period of the t-th day (such as ) and the habit stability of users using applications in the s-th time period of target days (such as σtsL 2 ) to predict the target quantity N for the first time period on the t-th day ts , which can further improve the accuracy of the predicted target quantity for the first time period on the t-th day.

[0018] In a possible implementation, the multiple applications in the first prediction list are arranged in descending order of application usage probability. Select the applications with the target quantity from the first prediction application list to obtain the first application list, including: intercepting the first N ts applications from the first prediction application list to form the first application list. N ts is the target quantity.

[0019] In this way, it can be achieved that the applications in the first application list are also arranged in descending order of application usage probability. When the electronic device uses the first application list to kill the background-running applications, if it is necessary to kill the background-running applications with application identifiers in the first application list in order to make the electronic device have sufficient available memory, the electronic device can preferentially kill the applications located at the end of the first application list. It should be understood that the usage probability of the applications located at the end of the first application list is less than the usage probability of the applications located at the top of the first application list, so as to reduce the occurrence probability of the applications that the user will use being killed.

[0020] In a possible implementation, before using the first application list to kill M applications, the method further includes: determining whether the first similarity corresponding to the first time period on the t-th day is greater than the similarity threshold, and / or whether the first dispersion corresponding to the first time period on the t-th day is greater than the dispersion threshold. The first similarity represents the similarity between the application usage quantity in the first time period on the t-th day and the application usage quantity in the time period before the first time period on the t-th day, and the first dispersion represents the difference between the application usage quantity in the first time period on the t-th day and the application usage quantity in the first time periods of multiple days before the t-th day. Using the first application list to kill M applications includes: when the first similarity is less than or equal to the similarity threshold, and / or the first dispersion is less than or equal to the dispersion threshold, using the first application list to kill M applications.

[0021] In this way, if the first similarity is less than or equal to the similarity threshold, and / or the first dispersion is less than or equal to the dispersion threshold, it can indicate that the user's habit of using applications in the first time period is stable. Using the first application list for killing can increase the probability that the applications that the user will not use in the first time period on the t-th day are killed, and reduce the power consumption of the electronic device.

[0022] In a possible implementation, the second time period of the t-th day belongs to the time period of the t-th day. The second application list includes a second predicted application list, which is predicted by using a preset model when the second time period of the t-th day is reached. The preset model is also used to output the second predicted application list when the application usage data in the second time period in the L days before the t-th day and the application usage data of the application used R times before the second recommended operation time are input. The second recommended operation time is the time when the second time period of the t-th day is reached. Before using the second application list to check and kill N applications, the method further includes: determining whether the second similarity corresponding to the second time period of the t-th day is greater than the similarity threshold, and / or whether the second dispersion corresponding to the second time period of the t-th day is greater than the dispersion threshold. The second similarity represents the similarity between the application usage quantity in the second time period of the t-th day and the application usage quantity in the time period before the second time period of the t-th day. The second dispersion represents the difference between the application usage quantity in the second time period of the t-th day and the application usage quantity in the second time periods of multiple days before the t-th day. Using the second application list to check and kill N applications includes: when the second similarity is greater than the similarity threshold, and / or the second dispersion is greater than the dispersion threshold, using the second predicted application list to check and kill N applications.

[0023] In this way, if the second similarity is greater than the similarity threshold, and / or the second dispersion is greater than the dispersion threshold, it can indicate that the user's habit of using applications in the second time period is unstable. Using the second predicted application list for checking and killing can reduce the probability that the applications that the user will use in the second time period of the t-th day are checked and killed, thus improving the user experience.

[0024] In a possible implementation, the first dispersion includes the first variance or the first standard deviation. The first variance is the variance of the application usage quantity in the first time period in the L days before the t-th day. The first standard deviation is the standard deviation of the application usage quantity in the first time period in the L days before the t-th day. The first similarity is related to the application usage quantity in the first time period in the K days before the t-th day and the application usage quantity in the first Q time periods of the first time period of the t-th day.

[0025] In this way, based on the variance or standard deviation of the application usage quantity in the first time period in the L days before the t-th day, the stability of the user's habit of using applications in the first time period can be determined, and an accurate judgment of the stability of the user's habit of using applications can be achieved. Based on the first similarity, the stability of the user's habit of using applications in the first time period can be determined. The first similarity can be determined based on the application usage quantity in the first time period in the K days before the t-th day and the application usage quantity in the first Q time periods of the first time period of the t-th day, and thus an accurate judgment of the stability of the user's habit of using applications can be achieved.

[0026] In a possible implementation, when the t-th day is a working day, the first variance is the variance of the application usage quantity in the first time period of working days within the L days before the t-th day, and the first standard deviation is the standard deviation of the application usage quantity in the first time period of working days within the L days before the t-th day. The first similarity is related to the application usage quantity in the first time period of the K working days before the t-th day and the application usage quantity in the Q time periods before the first time period of the t-th day. When the t-th day is a holiday, the first variance is the variance of the application usage quantity in the first time period of holidays within the L days before the t-th day, and the first standard deviation is the standard deviation of the application usage quantity in the first time period of holidays within the L days before the t-th day. The first similarity is related to the application usage quantity in the first time period of the K holidays before the t-th day and the application usage quantity in the Q time periods before the first time period of the t-th day.

[0027] In this way, by distinguishing the date type of the t-th day, the accuracy of judging the habit stability of users' application usage can be further improved.

[0028] In a possible implementation, the first time period of the t-th day is the s-th time period of the t-th day. The first similarity N wts Satisfies the following formula:

[0029] N wts = max[N w,ti,s , N w,t,sj

[0030] Wherein, N w,ti,s = N ti,s - A s ,N w,t , sj = N t,sj - A s ,N ti,s is the application usage quantity in the s-th time period of the i-th target day before the t-th day, and there are i - 1 target days between the i-th target day and the t-th day. N t,sj is the application usage quantity in the j-th time period before the s-th time period of the t-th day, and there are j - 1 time periods belonging to the target days between the j-th time period and the s-th time period of the t-th day. A s is the quantity of applications used in both the s-th time period and the j-th time period of the i-th target day. i = 1, 2,..., K, and j = 1, 2,..., Q. The target days satisfy: when the t-th day is a working day, the target days are working days; when the t-th day is a holiday, the target days are holidays.

[0031] In this way, due to the application usage in the first time period during the K days before the t-th day and the application usage in the first Q time periods of the t-th day in the first time period, it will have an impact on the user's application usage in the first time period of the t-th day. In this way, based on the number of applications used in the first time period during the K days before the t-th day, the number of applications used in the first Q time periods of the t-th day in the first time period, and the number of applications that the user used in both the first time period during the K days before the t-th day and the first Q time periods of the t-th day in the first time period, a first similarity is obtained, and the first similarity is used to judge the habit stability of the user using applications in the first time period of the t-th day, which can improve the accuracy of the judgment.

[0032] In a possible implementation manner, the first time period of the t-th day is the s-th time period of the t-th day. The first variance σ tsL 2 and the first standard deviation σ tsL satisfy the following formula:

[0033]

[0034] where x ts , p is the number of applications used in the s-th time period of the p-th target day during the L days before the t-th day, n is the number of target days during the L days before the t-th day, The target day satisfies: when the t-th day is a working day, the target day is a working day; when the t-th day is a holiday, the target day is a holiday.

[0035] In this way, when using the first variance or the first standard deviation to judge the habit stability of the user using applications in the first time period, the accuracy of the judgment can be improved.

[0036] In a possible implementation, the first time period on the t-th day is the s-th time period on the t-th day, and the preset model is a multi-channel recall model. The first predicted application list for the first time period is obtained by using the preset model, including: obtaining the first data corresponding to the first time period. The first data includes first long-term data, first recent data, and first real-time data. The first long-term data includes the application usage data of the s-th time period of the target day in the L days before the t-th day. The recent data includes the application usage data of the s-th time period of the U target days before the t-th day. The real-time data includes the application usage data of the R times of using the application before the first recommended operation time. The multi-channel recall model processes the first data in the following manner to obtain the first predicted application list: The classification and regression tree (CART) algorithm in the multi-channel recall model is used to calculate the long-term data to obtain the first weights of each application corresponding to the long-term data. The nearest popular recall algorithm in the multi-channel recall model is used to calculate the recent data to obtain the second weights of each application corresponding to the recent data. The time decay algorithm in the multi-channel recall model is also used to calculate the real-time data to obtain the third weights of each application corresponding to the real-time data. The multiple applications corresponding to the first data are sorted in the order of the fourth weights of the applications to obtain the first sorted list. The fourth weight of an application is the sum of the first weight, the second weight, and the third weight of the application. The fourth weight of the application is related to the first usage probability of the application. The first N0 applications in the first sorted list are obtained to form the first predicted application list, where N0 is a preset value and N0 is greater than the target quantity.

[0037] In this way, as Figure 4 shown in S402, the multi-channel recall model can use three-way recall to separately learn the habits of users' long-term, recent, and real-time use of applications (APPs), and also use the weights of each recall to achieve the fusion sorting of multi-channel recall. The multi-channel recall model comprehensively considers the habits of users' long-term use of applications, recent use of applications, and real-time use of applications. When considering the long-term, recent, and real-time data, the accuracy of sorting the predicted application list output by the multi-channel recall model in the order of usage probability can be improved.

[0038] In a possible implementation, the t-th day contains forty-eight time periods, and there is no overlapping time between two adjacent time periods among the forty-eight time periods, and the duration of each time period among the forty-eight time periods is the same. It can be understood that if the t-th day contains forty-eight time periods, the duration of each time period can be 30 minutes (min). In this way, when receiving a recommended operation, the electronic device can obtain the application list of the time period where the recommended operation time is located, and the electronic device can use the application list to kill the applications running in the background, which can reduce the power consumption of the electronic device while ensuring accurate killing.

[0039] Second aspect, an embodiment of the present application provides an application control device, which may be an electronic device, or a chip or a chip system within the electronic device. The application control device may include a display unit and a processing unit. When the application control device is an electronic device, the processing unit may be a processor. The application control device may further include a storage unit, which may be a memory. The storage unit is used to store instructions, and the processing unit executes the instructions stored in the storage unit to enable the electronic device to implement an application control method described in the first aspect or any possible implementation manner of the first aspect. When the application control device is a chip or a chip system within the electronic device, the processing unit may be a processor. The processing unit executes the instructions stored in the storage unit to enable the electronic device to implement an application control method described in the first aspect or any possible implementation manner of the first aspect. The storage unit may be a storage unit within the chip (e.g., registers, caches, etc.), or a storage unit outside the chip within the electronic device (e.g., read-only memory, random access memory, etc.).

[0040] Third aspect, an embodiment of the present application provides an electronic device, including a processor and a memory. The memory is used to store code instructions, and the processor is used to run the code instructions to execute the method described in the first aspect or any possible implementation manner of the first aspect.

[0041] Fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which a computer program or instructions are stored. When the computer program or instructions run on a computer, the computer is enabled to execute the method described in the first aspect or any possible implementation manner of the first aspect.

[0042] Fifth aspect, an embodiment of the present application provides a computer program product including a computer program. When the computer program runs on a computer, the computer is enabled to execute the method described in the first aspect or any possible implementation manner of the first aspect.

[0043] Sixth aspect, the present application provides a chip or a chip system, which includes at least one processor and a communication interface. The communication interface and the at least one processor are interconnected by a line. The at least one processor is used to run a computer program or instructions to execute the method described in the first aspect or any possible implementation manner of the first aspect. Among them, the communication interface in the chip may be an input / output interface, a pin, a circuit, etc.

[0044] In a possible implementation, the chip or chip system described above in this application further includes at least one memory, and instructions are stored in the at least one memory. The memory may be a storage unit inside the chip, for example, registers, caches, etc., or it may be a storage unit of the chip (for example, read-only memory, random access memory, etc.).

[0045] It should be understood that the technical solutions of the second to sixth aspects of this application correspond to those of the first aspect of this application, and the beneficial effects obtained by each aspect and the corresponding feasible implementation manners are similar, and will not be elaborated herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of a scenario provided by an embodiment of this application;

[0047] Figure 2 A schematic diagram of the structure of an electronic device provided by an embodiment of this application;

[0048] Figure 3 A schematic diagram of the software architecture of an electronic device provided by an embodiment of this application;

[0049] Figure 4 A schematic diagram of the flowchart of an application control method provided by an embodiment of this application;

[0050] Figure 5 A schematic diagram of the architecture of a multi-channel recall model provided by an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] In order to facilitate a clear description of the technical solutions of the embodiments of this application, hereinafter, some terms and technologies involved in the embodiments of this application are briefly introduced:

[0052] 1. Decision tree algorithm

[0053] The decision tree algorithm is widely used in models for classification and regression tasks and is a tree structure that describes how to classify instances. It is a typical classification method. First, the data is processed, and an induction algorithm is used to generate readable rules and decision trees, and then the decision is used to analyze new data. Essentially, a decision tree is a process of classifying data through a series of rules.

[0054] Common decision tree algorithms may include the classification and regression tree (CART) algorithm, and may also include the ID3 decision tree or the C4.5 algorithm. Among them, the CART algorithm supports binary classification problems. The calculation results of the CART algorithm are all probability values. In the case of classification, the Gini index minimization criterion is often adopted.

[0055] 2. Other Terms

[0056] In the embodiments of the present application, terms such as "first" and "second" are used to distinguish identical or similar items with basically the same functions and roles. For example, the first chip and the second chip are only used to distinguish different chips, and do not limit their sequence. Those skilled in the art can understand that terms such as "first" and "second" do not limit the quantity and execution order, and "first", "second", etc. do not necessarily mean different.

[0057] It should be noted that 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 present application should not be construed as being more preferred or having more advantages than other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.

[0058] In the embodiments of the present application, "at least one" means one or more, and "a plurality" means two or more. "And / or" describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. "At least one (item)" or similar expressions refer to any combination of these items, including any combination of single item (item) or plural items (items). For example, at least one (item) of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, c can be single or multiple.

[0059] The user can open multiple applications on the electronic device. One of the opened multiple applications can run in the foreground, and the remaining applications among the opened multiple applications can run in the background.

[0060] For ease of understanding, in combination with Figure 1 a scenario where one application runs in the foreground and the remaining applications among the multiple applications run in the background is described. Figure 1 A schematic diagram of a scenario provided by the embodiments of the present application is shown.

[0061] Taking the file management application, text message application, video application, WeChat application, Douyin application, map application, and Taobao application on the electronic device being opened, with the file management application running in the foreground and the text message application, video application, WeChat application, Douyin application, map application, and Taobao application running in the background as an example, when the file management application runs in the foreground, the electronic device can display Figure 1The interface of the file management application shown in a of []. As Figure 1 In the interface shown in a of []. A virtual navigation bar 101 is displayed at the bottom of the display screen. The virtual navigation bar 101 includes multiple virtual controls. The virtual controls are such as a back control, a Home control, and a function control 102.

[0062] When the electronic device displays Figure 1 In the case of the interface shown in a of []. The user can perform a click operation on the function control 102. In response to the user's operation of clicking the function control 102, the electronic device can display as Figure 1 The recent tasks interface 103 shown in b of []. The recent tasks interface 103 may include preview interfaces of applications running in the background. Figure 1 In the recent tasks interface 103 shown in b of []. Only the preview interfaces of the file management application, the SMS application, and the video application are displayed. When the electronic device displays Figure 1 In the case of the interface shown in b of []. The user can perform a left - swipe operation or a right - swipe operation on the display screen. In this way, the preview interfaces of other applications running in the background can be displayed on the recent tasks interface 103.

[0063] When the electronic device displays Figure 1 In the case of the interface shown in b of []. The user can perform a click operation on the blank area. The electronic device can display Figure 1 The interface of the desktop application shown in c of []. As Figure 1 In the interface shown in c of []. The interface includes icons 104 of recommended applications.

[0064] When the electronic device displays Figure 1 In the case of the interface shown in c of []. The user can perform a recommendation operation. The recommendation operation is, for example, a click operation on the icon 104. In response to the recommendation operation, the electronic device can obtain a set of applications that the user may use during the time period to which the recommendation operation moment belongs.

[0065] In some implementations, in response to the recommendation operation, the electronic device uses a prediction model to predict the applications that the user may use during the time period to which the recommendation operation moment belongs, and obtains a predicted application list. The predicted application list contains the applications that the user may use. The electronic device uses the predicted application list to kill multiple applications running in the background.

[0066] Among them, the number of applications in the predicted application list is a preset fixed value. The prediction model is used to sort the applications used by the user before the recommended operation moment in the order of the magnitude of the usage probability, and then intercept a part of the applications according to the preset fixed value to obtain the predicted application list. The preset fixed value is the maximum value of the application usage quantity in each time period of a day. Therefore, the number of applications in the predicted application list corresponding to different time periods obtained by the prediction model is the same. The applications in the predicted application list corresponding to different time periods can be as shown in Table 1.

[0067] Table 1 Applications in the predicted application list corresponding to different time periods

[0068]

[0069] It should be understood that the time periods shown in Table 1 can be time periods belonging to the same day. The time periods shown in Table 1 can also be time periods belonging to different days.

[0070] Table 1 shows the applications and the number of applications in the predicted application list corresponding to the first time period, the second time period, the third time period, and the fourth time period respectively. As shown in Table 1, the electronic device can use the prediction model to obtain the predicted application list corresponding to different time periods, and the number of applications in the predicted application list corresponding to different time periods is the same.

[0071] When the electronic device uses the predicted application list for killing processes, the number of processes running in the background after killing in different time periods is the same. However, the number of applications used by the user in different time periods may be different. If the predicted application list is used to kill the processes running in the background, it will cause the applications that the user will not use not to be killed, resulting in a relatively large power consumption of the electronic device and a short battery life of the electronic device.

[0072] In view of this, the embodiment of the present application provides an application control method. In response to the recommended operation, the electronic device uses the application list corresponding to the time period to which the recommended operation moment belongs to kill the processes running in the background. Among them, the number of applications in the application list corresponding to different time periods is different. In this way, the probability that the applications that the user will not use are killed can be increased, and thus the power consumption of the electronic device can be reduced.

[0073] The electronic device according to the embodiments of the present application may include a handheld device, a vehicle-mounted device, etc. having an application recommendation function. For example, some electronic devices are: mobile phone, tablet computer, handheld computer, laptop computer, mobile internet device (MID), wearable device, virtual reality (VR) device, augmented reality (AR) device, wireless terminal in industrial control, wireless terminal in self-driving, wireless terminal in remote medical surgery, wireless terminal in smart grid, wireless terminal in transportation safety, wireless terminal in smart city, wireless terminal in smart home, cellular phone, cordless phone, session initiation protocol (SIP) phone, wireless local loop (WLL) station, personal digital assistant (PDA), handheld device with wireless communication function, computing device or other processing devices connected to a wireless modem, vehicle-mounted device, wearable device, terminal device in a 5G network, or terminal device in a future evolved public land mobile network (PLMN), etc. The embodiments of the present application are not limited thereto.

[0074] By way of example and not limitation, in the embodiments of the present application, the electronic device may also be a wearable device. A wearable device may also be referred to as a wearable intelligent device, which is a general term for devices developed by applying wearable technology to the intelligent design of daily wear, such as glasses, gloves, watches, clothing, and shoes. A wearable device is a portable device that is either directly worn on the body or integrated into the user's clothing or accessories. A wearable device is not only a hardware device, but also realizes powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable intelligent devices include those with complete functions and large sizes that can realize complete or partial functions without relying on a smart phone, such as smart watches or smart glasses, etc., and those that only focus on a certain type of application function and need to cooperate with other devices such as smart phones, such as various smart bracelets and smart jewelry for monitoring physical signs.

[0075] In addition, in the embodiments of the present application, the electronic device may also be a terminal device in an Internet of Things (IoT) system. The IoT is an important part of the future development of information technology. Its main technical feature is to connect objects to the network through communication technology, thereby realizing an intelligent network of human-machine interconnection and object-object interconnection.

[0076] The electronic device in the embodiments of the present application may also be referred to as: terminal device, user equipment (UE), mobile station (MS), mobile terminal (MT), access terminal, user unit, user station, mobile station, mobile terminal, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent or user device, etc.

[0077] In the embodiments of the present application, the electronic device or each network device includes a hardware layer, an operating system layer running on the hardware layer, and an application layer running on the operating system layer. The hardware layer includes hardware such as a central processing unit (CPU), a memory management unit (MMU), and a memory (also referred to as main memory). The operating system can be any one or more computer operating systems that implement service processing through processes. For example, Linux operating system, Unix operating system, Android operating system, iOS operating system, or Windows operating system, etc. The application layer includes applications such as a browser, an address book, a word processing software, and an instant messaging software.

[0078] Figure 2 The structure diagram of the electronic device provided by the embodiments of the present application is shown.

[0079] As Figure 2As shown in the figure, the electronic device 200 may include a processor 210, an external memory interface 220, an internal memory 221, a universal serial bus (USB) interface 230, a charging management module 240, a power management module 241, a battery 242, an antenna 1, an antenna 2, a mobile communication module 250, a wireless communication module 260, an audio module 270, a speaker 270A, a receiver 270B, a microphone 270C, a headphone interface 270D, a sensor module 280, a button 290, a motor 291, an indicator 292, a camera 293, a display screen 294, and a subscriber identification module (SIM) card interface 295, etc. The sensor module 280 may include a pressure sensor 280A, a gyroscope sensor 280B, a barometric pressure sensor 280C, a magnetic sensor 280D, an acceleration sensor 280E, a distance sensor 280F, a proximity light sensor 280G, a fingerprint sensor 280H, a temperature sensor 280J, a touch sensor 280K, an ambient light sensor 280L, a bone conduction sensor 280M, etc.

[0080] It can be understood that the structure illustrated in the embodiments of the present invention does not constitute a specific limitation on the electronic device 200. In other embodiments of the present application, the electronic device 200 may include more or fewer components than those shown in the figure, or combine certain components, or split certain components, or have different component arrangements. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0081] Exemplarily, in a first time period, M applications are running in the background of the electronic device. The processor 210 of the electronic device receives a first recommended operation through the sensor module 280. In response to the first recommended operation, the processor 210 may obtain a first application list corresponding to the first time period to which the first recommended operation belongs. The processor 210 may use the first application list to kill the M applications, so that the applications whose application identifiers are not in the first application list among the M applications are killed. In this way, the power consumption of the electronic device and the memory occupation can be reduced.

[0082] In a second time period, N applications are running in the background of the electronic device. The processor 210 of the electronic device receives a second recommended operation through the sensor module 280. In response to the second recommended operation, the processor 210 may obtain a second application list corresponding to the second time period to which the second recommended operation belongs. The processor 210 may use the second application list to kill the N applications, so that the applications whose application identifiers are not in the second application list among the N applications are killed. In this way, the power consumption of the electronic device and the memory occupation can be reduced.

[0083] The number of applications in the second application list is different from the number of applications in the first application list. The first time period is different from the second time period. M and N are both integers.

[0084] As shown in the first application list corresponding to the first time period and the second application list corresponding to the second time period, the number of applications in the application lists corresponding to different time periods is different. In this way, the probability of applications that users will not use being detected and killed can be increased, thereby reducing the power consumption of electronic devices.

[0085] The software system of the electronic device 200 may adopt a layered architecture, an event-driven architecture, a micro-core architecture, a micro-service architecture, or a cloud architecture. In the embodiment of the present invention, the Android system of the layered architecture is taken as an example to exemplify the software structure of the electronic device 100.

[0086] Figure 3 A schematic diagram of the software architecture of an electronic device provided in an embodiment of the present application is shown.

[0087] The layered architecture divides the software into several layers, each with clear roles and division of labor. The layers communicate with each other through software interfaces. In some embodiments, the Android system is divided into four layers, from top to bottom: the application layer, the application framework layer, the Android runtime and system library, and the kernel layer.

[0088] The application layer can include a series of application packages.

[0089] like Figure 3 As shown, the application package may include desktop management, perception, recommendation, camera, gallery, calendar, call, map, navigation, WLAN, Bluetooth, music, video, short message and other applications (also referred to as applications). Among them, the recommended application can be used for application recommendation. In some implementations, the recommended application is called an application recommendation application, and the application recommendation application is referred to as application recommendation.

[0090] 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.

[0091] like Figure 3 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, a location manager (Location Based Services, LBS), etc.

[0092] The location manager can be used to obtain the current location of the electronic device. For example, it can obtain the current Global Positioning System (GPS) data, (Wireless Fidelity, Wi-Fi) positioning data, and positioning data of cell base stations.

[0093] The Android Runtime includes core libraries and a virtual machine. The Android runtime is responsible for the scheduling and management of the Android system.

[0094] The core libraries consist of two parts: one part is the functional functions that need to be called by the Java language, and the other part is the core libraries of Android.

[0095] The application layer and the application framework layer run in the 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 the management of object life cycles, stack management, thread management, security and exception management, and garbage collection.

[0096] The system libraries can include multiple functional modules. For example: Surface Manager, Media Libraries, 3D graphics processing libraries (such as: OpenGL ES), 2D graphics engines (such as: SGL), etc.

[0097] The kernel layer is the layer between hardware and software. The kernel layer at least includes a display driver, a camera driver, an audio driver, and a sensor driver.

[0098] Exemplarily, in the case where the user clicks on a certain application, the desktop management application can send an application click event to the perception application. Based on the application click event, the perception application can collect the current time and the application corresponding to the application click event, and generate application usage data.

[0099] Among them, the application usage data can include the start time when the application runs in the foreground and the application identifier. The application identifier can be the application name. The start time when the application runs in the foreground can be the time when the user clicks on the application to start the application, or it can be the time when the user clicks on the application to convert the application from running in the background to running in the foreground. The application usage data includes the number of applications used. The number of applications used can be the same as the number of start time points when the application runs in the foreground. For example, if there are 5 start time points when the application runs in the foreground in the application usage data for the Xth time period, then the number of applications used in the Xth time period is 5.

[0100] At the first moment, the user can perform a first recommendation operation. The first recommendation operation is, for example, an operation where the user clicks on a recommended application. The desktop management application can send an application click event to the recommended application. The application click event includes a timestamp indicating the first moment. The recommended application can identify that the first moment belongs to a first time period. The recommended application can obtain application usage data corresponding to the first time period from a sensing application. The recommended application can transmit the application usage data corresponding to the first time period to the application framework layer.

[0101] The application framework layer can input the application usage data transmitted by the recommended application into a preset model. The preset model can output a first predicted application list corresponding to the first time period to which the first moment belongs. The application framework layer can select a target number of applications in the first time period to which the first moment belongs from the first predicted application list to obtain a first application list. The application framework layer can transmit the first application list to the recommended application. The recommended application can transmit the first application list to a killing module for killing background-running applications. The killing module can use the first application list to kill the applications running in the background.

[0102] Among them, the target number of applications in the first time period to which the first moment belongs is the predicted number of applications used in the first time period to which the first moment belongs. The killing module can be located in the application layer or in the application framework layer. The killing module is not shown in Figure 3 the figure.

[0103] At the second moment, the user can perform a second recommendation operation. The second recommendation operation is, for example, that the user clicks on a recommended application. The second moment belongs to a second time period. The first time period is different from the second time period. In response to the second recommendation operation, the recommended application can obtain a second application list corresponding to the second time period to which the second moment belongs transmitted by the application framework layer. The recommended application can transmit the second application list to the killing module. The killing module can use the second application list to kill the applications running in the background.

[0104] Among them, the specific implementation principle for the recommended application to obtain the second application list is similar to the specific implementation principle for the recommended application to obtain the first application list, and will not be elaborated here. The target number used by the application framework layer to obtain the second application list is the predicted number of applications used in the second time period to which the second moment belongs. The number of applications in the second application list is different from the number of applications in the first application list. In this way, it is possible to make the number of applications in the application lists corresponding to different time periods different. The killing module uses the application list corresponding to the time period to which the recommendation operation moment belongs to kill the applications running in the background, which can increase the probability that applications not used by the user are killed, and thus can reduce the power consumption of the electronic device.

[0105] Figure 4It shows a schematic flowchart of an application control method provided by an embodiment of the present application.

[0106] As Figure 4 shown, the method includes:

[0107] S401. The electronic device obtains the training set model features corresponding to the prediction period.

[0108] Exemplarily, taking the prediction period as the first time period as an example. The electronic device may receive a first recommendation operation. In the case of receiving the first recommendation operation, the electronic device may identify the first time period to which the first recommendation operation moment belongs, and obtain the first data corresponding to the first time period to which the first recommendation operation moment belongs. Since the feature type in the first data is the same as the training set model feature type, therefore, the first data may also be referred to as the training set model features corresponding to the first time period to which the first recommendation operation moment belongs. The first data may also be referred to as the training set model features corresponding to the first recommendation operation moment.

[0109] Exemplarily, taking the first recommendation operation moment belonging to the t-th day and the first time period belonging to the s-th time period of the t-th day as an example, the first data may include first long-term data, first recent data, and first real-time data.

[0110] Among them, the first long-term data includes the application usage data in the s-th time period in the L days before the t-th day. L may be 90, or may be other values such as 60 or 30.

[0111] The first recent data includes the application usage data in the s-th time period in the U days before the t-th day. L > U. U may be 3, or may be other values such as 5 or 7.

[0112] The first real-time data includes the application usage data of using the application R times before the first recommendation operation moment. R may be 5, or may be other values such as 7, 10, or 12. Using the application R times can be understood as having R start time points when the application runs in the foreground, or can be understood as the sum of the number of operations to start the application and the number of operations to convert the application from running in the background to running in the foreground is R.

[0113] Taking the first recommendation operation moment as 10:10 on December 14th, L as 90, U as 3, and R as 5 as an example, the first data may be as shown in Table 2.

[0114] Table 2 Schematic of the first data

[0115]

[0116] As shown in Table 2, the first long-term data may include application usage data for the s-th time period of each day in the 90 days before the t-th day. The first recent data may include application usage data for the s-th time period of each day in the 3 days before the t-th day. The first real-time data may include application usage data for the 5 times of using the application before the first recommended operation time of the t-th day (such as 10:10 on December 14th). The format of the application usage data may refer to the format of the application usage data for the 5 times of using the application before the first recommended operation time of the t-th day shown in Table 2. Among them, "12.13 24:00, Application A" indicates an application with the application name of Application A, and the start time running in the foreground is 24:00 on December 13th.

[0117] It can be understood that the (t - 1)-th day represents the day adjacent to and before the t-th day. The (t - 2)-th day represents the day adjacent to and before the (t - 1)-th day. The (t - 3)-th day represents the day adjacent to and before the (t - 2)-th day. The (t - 4)-th day represents the day adjacent to and before the (t - 3)-th day. The (t - 5)-th day represents the day adjacent to and before the (t - 4)-th day. The (t - 88)-th day represents the day adjacent to and before the (t - 87)-th day. The (t - 89)-th day represents the day adjacent to and before the (t - 88)-th day. The (t - 90)-th day represents the day adjacent to and before the (t - 89)-th day.

[0118] Optionally, still taking the first recommended operation time belonging to the t-th day and the first time period belonging to the s-th time period of the t-th day as an example. If the t-th day is a working day, the first long-term data may include application usage data for the s-th time period of working days in the L days before the t-th day. The first recent data may include application usage data for the s-th time period of working days in the U days before the t-th day. The first real-time data may include application usage data for the R times of using the application before the first recommended operation time of the t-th day. Among them, the times of using the application R times before the first recommended operation time of the t-th day all belong to the time of working days.

[0119] Optionally, still taking the first recommended operation time belonging to the t-th day and the first time period belonging to the s-th time period of the t-th day as an example. If the t-th day is a holiday, the first long-term data may include application usage data for the s-th time period of holidays in the L days before the t-th day. The first recent data may include application usage data for the s-th time period of holidays in the U days before the t-th day. The first real-time data may include application usage data for the R times of using the application before the first recommended operation time of the t-th day. Among them, the times of using the application R times before the first recommended operation time of the t-th day all belong to the time of holidays.

[0120] S402. The electronic device performs model prediction and inference to obtain a predicted application list corresponding to the prediction period.

[0121] Exemplarily, the electronic device may input the first data into a preset model, and the preset model may output a first predicted application list corresponding to the first time period.

[0122] Exemplarily, the preset model may be a multi-channel recall model. The multi-channel recall model may include a recall layer and a sorting layer. Among them, the recall layer may include the first-channel recall, the second-channel recall, and the third-channel recall. The algorithm used for the first-channel recall is the Classification and Regression Tree (CART) algorithm. The algorithm used for the second-channel recall is the recent popular recall algorithm. The algorithm used for the third-channel recall is the time decay algorithm. The recall layer is used to quickly screen a collection of tens of millions of candidate items (or features) to obtain a collection of thousands or even hundreds of features. The sorting layer is used to uniformly score and sort the results of the multi-channel recalls in the recall layer and select the top several (Topk). Figure 5 The architecture diagram of the multi-channel recall model is shown. For ease of understanding, Figure 5 it will be described later.

[0123] The electronic device may input the first long-term data into the Classification and Regression Tree (CART) of the multi-channel recall model to obtain the first probability values of each application in the first long-term data. Based on the first probability values of each application and the first-channel recall weight, the electronic device may calculate the first weight of each application.

[0124] Among them, the first-channel recall weight represents the proportion of the first-channel recall in all-channel sorting, and the first-channel recall weight may be preset. The first weight of an application may be obtained by multiplying the first probability value of the application by the first-channel recall weight. For example, the first weight of application A may be obtained by multiplying the first probability value of application A by the first-channel recall weight.

[0125] The electronic device may input the first recent data into the recent popular recall algorithm to obtain the second probability values of each application in the first recent data. Based on the second probability values of each application and the second-channel recall weight, the electronic device may calculate the second weight of each application. Exemplarily, the recent popular recall algorithm may determine the proportion of the usage times of each application in the first recent data to the sum of the usage times of the applications in the first recent data as the first probability of each application. For example, the usage times of the applications in the first recent data are 60 times. Among them, the memo is used 12 times, the video is used 6 times, the email is used 6 times, the settings are used 5 times, the weather is used 3 times, and the clock is used 3 times. It can be determined that the second probability value of the memo is 0.2, the second probability value of the video is 0.1, the second probability value of the email is 0.1, the second probability value of the settings is 0.083, the second probability value of the weather is 0.05, and the second probability value of the clock is 0.05.

[0126] Among them, the recall weight of the second path represents the proportion of the second path recall in all path rankings, and the recall weight of the second path can also be preset. The second weight applied can be obtained by multiplying the second probability value applied by the recall weight of the second path.

[0127] The electronic device can input the first real-time data into the time decay algorithm to obtain the third probability value of each application in the first real-time data. Based on the third probability value of each application and the recall weight of the third path, the electronic device can calculate and obtain the third weight of each application.

[0128] Among them, the third probability value of the y-th application in the first real-time data satisfies the formula: δ(T) y is the decay value of the y-th application with a time interval of T y . T y is the time difference between the moment when the y-th application starts running in the foreground and the first recommended operation moment. ε0 is the initial decay value. θ is the exponential decay constant. l is the translation amount representing leftward, so that the value does not start decaying from ε0 but continues to decay from any position. ε0, θ, and l can all be values obtained by the electronic device through pre-training. The recall weight of the third path represents the proportion of the third path recall in all path rankings, and the recall weight of the third path can also be preset. The third weight applied can be obtained by multiplying the third probability value applied by the recall weight of the third path.

[0129] Based on the first weight, second weight, and third weight of each application, the electronic device can obtain the fourth weight of each application. The electronic device can sort the multiple applications corresponding to the first data in the order of the fourth weight of the application to obtain the first sorted list. Among them, the fourth weight of the application is the sum of the first weight of the application, the second weight of the application, and the third weight of the application. For example, the fourth weight of application A is the sum of the first weight of application A, the second weight of application A, and the third weight of application A. The fourth weight of the application is related to the first usage probability of the application. In this way, the sorting of the applications in the first sorted list is in the order of the usage probability of the application.

[0130] The electronic device can intercept the first N0 applications in the first sorted list to form the first predicted application list. Among them, N0 is a preset value, and N0 can be the maximum value of the application usage quantities corresponding to each time period obtained through pre-statistics. N0 is greater than the target quantity corresponding to the first time period.

[0131] It should be understood that since the sorting of the applications in the first sorted list is in the order of the usage probability of the application, therefore, the multiple applications in the first predicted application list are sorted in the order of the usage probability of the application.

[0132] In this way, when the electronic device uses the first predicted application list to kill background-running applications, if it is necessary to kill background-running applications with application identifiers in the first predicted application list in order to enable the electronic device to have sufficient available memory, the electronic device may preferentially kill the applications located at the tail of the first predicted application list. It should be understood that the usage probability of the applications located at the tail of the first predicted application list is less than the usage probability of the applications located at the top of the first predicted application list, so as to reduce the occurrence probability of the applications that the user may use being killed.

[0133] It should be understood that the first predicted application list obtained in step S402 may also be referred to as the predicted application list corresponding to the first recommended operation moment. As shown in S401 - S402, the first real-time data included in the training set model features corresponding to the first recommended operation moment is associated with the first recommended operation moment.

[0134] As shown in the above first data, the first real-time data is the application usage data of using the application R times before the first recommended operation moment. The real-time data included in the training set model features corresponding to the v-th recommended operation moment may be the application usage data of using the application R times before the v-th recommended operation moment. When both the first recommended operation moment and the v-th recommended operation moment belong to the s-th time period of the t-th day, if the first real-time data is different from the real-time data included in the training set model features corresponding to the v-th recommended operation moment, then the first training data is different from the training set model features corresponding to the v-th recommended operation moment, and further the first predicted application list is different from the predicted application list corresponding to the v-th recommended operation moment.

[0135] Optionally, among the first-way recall weight, the second-way recall weight, and the third-way recall weight, each recall weight may be dynamically calculated using each recall rate.

[0136] The multi-way recall model shown above uses three-way recall to respectively learn the habits of users' long-term, recent, and real-time use of applications (APPs), and also uses each recall weight to achieve multi-way recall fusion sorting. In this way, the multi-way recall model of the embodiment of the present application comprehensively considers the habits of users' long-term use of applications, recent use of applications, and real-time use of applications. Considering the long-term, recent, and real-time data, the accuracy of sorting the predicted application list in the order of usage probability can be high.

[0137] Since the number of applications in the predicted application list for the predicted time period obtained by the multi-channel recall model is determined according to a preset value N0, and the multi-channel recall model does not consider the change in the number of applications used by the user at different time periods, it is necessary to accurately determine the application list corresponding to the predicted time based on the predicted application list for the predicted time period to kill the applications running in the background when it is determined that the user's habit of using applications is stable, so as to reduce the power consumption of the electronic device. Therefore, S403 and S404 shown later in the embodiments of the present application can determine the stability of the user's habit of using applications. Based on the judgment results of steps S403 and / or S404, the electronic device can determine whether to execute S405 or S407 subsequently.

[0138] S403. The electronic device can determine whether the dispersion of the predicted time period is greater than the dispersion threshold.

[0139] Exemplarily, the electronic device can determine whether the first dispersion corresponding to the first time period on the t-th day is greater than the dispersion threshold.

[0140] If the first dispersion is greater than the dispersion threshold, it means that the user's habit of using applications in the first time period on the t-th day is quite different from the user's habit of using applications in the first time period before the t-th day. To reduce the probability of the applications that the user will use in the first time period on the t-th day being killed, the electronic device can execute step S407.

[0141] If the first dispersion is less than or equal to the dispersion threshold, it can indicate that the user's habit of using applications in the first time period on the t-th day is less different from the user's habit of using applications in the first time period before the t-th day, or it can indicate that the user's habit of using applications in the first time period is stable. To further determine whether the user's habit of using applications in the first time period is stable, the electronic device can execute step S404.

[0142] Among them, the first dispersion can also represent the difference between the number of applications used in the first time period on the t-th day and the number of applications used in the first time period of multiple days before the t-th day.

[0143] Optionally, if the first dispersion is less than or equal to the dispersion threshold, the electronic device can execute step S405.

[0144] Exemplarily, the first dispersion can include the first variance or the first standard deviation. The dispersion threshold can include the variance threshold or the standard deviation threshold.

[0145] The first variance is the variance of the number of applications used in the first time period in the L days before the t-th day.

[0146] The first standard deviation is the standard deviation of the number of applications used in the first time period in the L days before the t-th day.

[0147] If the first variance is greater than the variance threshold, or the first standard deviation is greater than the standard deviation threshold, it can indicate that the user's usage habit of the application in the first time period on the t-th day is quite different from the user's usage habit of the application in the first time period before the t-th day. To reduce the occurrence probability of the application that the user will use in the first time period on the t-th day being killed, the electronic device can execute step S407.

[0148] If the first variance is less than or equal to the variance threshold, or the first standard deviation is less than or equal to the standard deviation threshold, it can indicate that the user's usage habit of the application in the first time period on the t-th day is less different from the user's usage habit of the application in the first time period before the t-th day, or it can indicate that the user's usage habit of the application in the first time period is stable. The electronic device can execute step S404 or S405.

[0149] Exemplarily, when the t-th day is a working day, the first variance is the variance of the application usage quantity in the first time period of working days in the L days before the t-th day, and the first standard deviation is the standard deviation of the application usage quantity in the first time period of working days in the L days before the t-th day.

[0150] When the t-th day is a holiday, the first variance is the variance of the application usage quantity in the first time period of holidays in the L days before the t-th day, and the first standard deviation is the standard deviation of the application usage quantity in the first time period of holidays in the L days before the t-th day.

[0151] In this way, when using the first variance or the first standard deviation to judge the stability of the user's usage habit of the application in the first time period, the accuracy of the judgment can be improved.

[0152] Exemplarily, the first variance σ tsL 2 and the first standard deviation σ tsL satisfy the following formula (1):

[0153]

[0154] where x ts , p is the application usage quantity in the s-th time period of the p-th target day in the L days before the t-th day, n is the number of target days in the L days before the t-th day, The target day satisfies: when the t-th day is a working day, the target day is a working day; when the t-th day is a holiday, the target day is a holiday.

[0155] S404. The electronic device can judge whether the similarity of the prediction time period is greater than the similarity threshold.

[0156] Exemplarily, the electronic device may determine whether the first similarity corresponding to the first time period on the t-th day is greater than the similarity threshold. The similarity threshold may be a value such as 0, 1, or 2.

[0157] If the first similarity is greater than the similarity threshold, it indicates that the user's habit of using the application in the first time period on the t-th day is quite different from the user's habit of using the application in the first time period before the t-th day. To reduce the occurrence probability of the application that the user will use in the first time period on the t-th day being detected and killed, the electronic device may perform step S407.

[0158] If the first similarity is less than or equal to the similarity threshold, it may indicate that the user's habit of using the application in the first time period on the t-th day is less different from the user's habit of using the application in the first time period before the t-th day, or it may indicate that the user's habit of using the application in the first time period is stable. To increase the probability of the application that the user will not use in the first time period on the t-th day being detected and killed, the electronic device may perform step S405.

[0159] Among them, the first similarity may also represent the similarity between the number of applications used in the first time period and the number of applications used in the time period before the first time period.

[0160] Exemplarily, the first similarity is related to the number of applications used in the first time period in the K days before the t-th day and the number of applications used in the first Q time periods before the first time period on the t-th day.

[0161] Among them, K may be an integer greater than or equal to 2. Q may be an integer greater than or equal to 2. K and Q may be the same or different.

[0162] Exemplarily, when the t-th day is a working day, the first similarity is related to the number of applications used in the first time period in the K working days before the t-th day and the number of applications used in the first Q time periods before the first time period on the t-th day.

[0163] When the t-th day is a holiday, the first similarity is related to the number of applications used in the first time period in the K holidays before the t-th day and the number of applications used in the first Q time periods before the first time period on the t-th day.

[0164] Exemplarily, the first similarity N wts satisfies the following formula (2):

[0165] N wts = max[N w,ti,s , N w,t,sj (2)

[0166] Among them, N w,ti,s = N ti,s - A s . Nw,t,sj = N t , sj -A s 。N ti,s is the usage quantity of the application in the s-th time period of the i-th target day before the t-th day. There are i - 1 target days between the i-th target day before the t-th day and the t-th day. N t,sj is the usage quantity of the application in the j-th time period before the s-th time period of the t-th day. There are j - 1 time periods belonging to the target day between the j-th time period before the s-th time period of the t-th day and the s-th time period of the t-th day. A s is the quantity of the application that is used in both the s-th time period of the i-th target day before the t-th day and the j-th time period before the s-th time period of the t-th day. i = 1, 2,..., K. j = 1, 2,..., Q. The target day satisfies: being a working day when the t-th day is a working day, and being a holiday when the t-th day is a holiday.

[0167] Exemplarily, assuming there are 48 time periods in a day, the s-th time period is from 10:00 to 10:30, K = Q = 2, the first time period of the t-th day or the s-th time period of the t-th day is from 10:00 to 10:30 on December 14th, December 14th is a working day and is Tuesday, and the similarity threshold is 2, N ti , s and N t , sj can be determined or composed as shown in Table 3.

[0168] Table 3 N ti , s and N t , sj Schematic of the usage quantity of the application

[0169]

[0170] As shown in Table 3, the s-th time period of the 1st target day before the t-th day is from 10:00 to 10:30 on December 13th. The s-th time period of the 2nd target day before the t-th day is from 10:00 to 10:30 on December 10th. The 1st time period before the s-th time period of the t-th day is from 09:30 to 10:00 on December 13th. The 2nd time period before the s-th time period of the t-th day is from 09:00 to 09:30 on December 13th.

[0171] It can be understood that if the first recommended operation time is 10:00, and 10:00 is the start time point of the time period 10:00 - 10:30, the electronic device can determine that the first recommended operation time belongs to the time period 10:00 - 10:30, and the first recommended operation time does not belong to the time period 09:30 - 10:00.

[0172] As shown in Table 3, the application usage quantity N in the s-th time period of the 1st target day before the t-th day t1 , s is 4. The application usage quantity N in the s-th time period of the 2nd target day before the t-th day t2 , s is 5. The application usage quantity N in the 1st time period before the s-th time period of the t-th day t , s1 is 4. The application usage quantity N in the 2nd time period before the s-th time period of the t-th day t , s2 is 3. Taking N t1 , s as an example, N t1 , s is determined as follows: Add up the number of start time points when the application runs in the foreground within the s-th time period of the 1st target day before the t-th day to obtain N t1，s .

[0173] As shown in Table 3, within the s-th time period of the i-th target day before the t-th day and the j-th time period before the s-th time period of the t-th day, there is 1 start time point when Application B runs in the foreground, 1 start time point when Application C runs in the foreground, and 1 start time point when Application D runs in the foreground, indicating that the number of applications used within the s-th time period of the i-th target day before the t-th day and the j-th time period before the s-th time period of the t-th day is 1 + 1 + 1 = 3, that is, A s = 3. Then, N w,t1,s = N t1，s - A s = 4 - 3 = 1; N w,t2,s = N t2，s - A s = 5 - 3 = 2; N w,t,s1 = N t，s1 - A s = 4 - 3 = 1; N w,t,s2 = N t，s2 - A s = 3 - 3 = 0.

[0174] Therefore, N wts = max[N w,ti,s , N w,t,sj = max[Nw,t1,s , N w,t2,s , N w,t,s1 , N w,t,s2 = max[1, 2, 1, 0] = 2.

[0175] Since the similarity threshold is 2, the electronic device can execute S405.

[0176] It can be understood that if the s-th time period of the i-th target day before the t-th day and the j-th time period before the s-th time period of the t-th day both contain 2 start time points when Application B runs in the foreground, 1 start time point when Application C runs in the foreground, and 1 start time point when Application D runs in the foreground, it means that the number of applications used in the s-th time period of the i-th target day before the t-th day and the j-th time period before the s-th time period of the t-th day can be 2 + 1 + 1 = 4.

[0177] The application usage in the first time period in the K days before the t-th day and the application usage in the first Q time periods before the first time period of the t-th day will affect the user's application usage in the first time period of the t-th day. Therefore, based on the number of applications used in the first time period in the K days before the t-th day, the number of applications used in the first Q time periods before the first time period of the t-th day, and the number of applications used in both the first time period in the K days before the t-th day and the first Q time periods before the first time period of the t-th day by the user, the first similarity is obtained, and the first similarity is used to judge the habit stability of the user's application usage in the first time period of the t-th day, which can improve the accuracy of the judgment.

[0178] The first similarity comprehensively considers

[0179] S405. The electronic device can obtain the target quantity in the prediction time period.

[0180] Exemplarily, the electronic device can obtain the target quantity in the first time period of the t-th day. Among them, the target quantity in the first time period of the t-th day is related to one or more of the following: the number of applications used in the first time period in the L days before the t-th day, the number of applications used in the first time period in the K days before the t-th day, the number of applications used in the Q time periods before the first time period of the t-th day. The Q time periods do not belong to the first time period.

[0181] Exemplarily, the target quantity in the first time period of the t-th day is positively correlated with one or more of the following: the number of applications used in the first time period in the L days before the t-th day, the number of applications used in the first time period in the K days before the t-th day, the number of applications used in the Q time periods before the first time period of the t-th day.

[0182] Exemplarily, if holidays and working days are not distinguished, the target quantity N of the first time period on the t-th day ts0 satisfies the following formula (3):

[0183]

[0184] where N t-i,s is the application usage quantity in the s-th time period on the (t - i)-th day before the t-th day. N t,s-j is the application usage quantity in the (s - j)-th time period before the s-th time period on the t-th day. There are j - 1 time periods between the (s - j)-th time period before the s-th time period on the t-th day and the s-th time period on the t-th day. The duration of the (s - j)-th time period and the s-th time period may be the same. γ1 is a weight coefficient. γ2 is a weight coefficient. γ3 is a weight coefficient, and γ3 can be 0.005. σ tsL0 2 is the variance of the application usage quantity in the s-th time period in the L days before the t-th day, σ tsL0 2 *γ3 represents the weighting of the variance σ tsL0 2 . γ1 can be greater than γ3, and γ2 can be greater than γ3.

[0185] Since the situation of the user using the application in the first time period on the t-th day may be affected by the user's habit of using the application in the first time period, the application usage situation in the s-th time period in multiple days before the t-th day, and the application usage situation in multiple time periods before the s-th time period on the t-th day. Therefore, as shown in formula (3), based on the application usage situation in the s-th time period in the K days before the t-th day (such as ), the application usage situation in the Q time periods before the s-th time period on the t-th day (such as ) and the habit stability of the user using the application in the s-th time period (such as σ tsL0 2 ), predicting the target quantity N of the first time period on the t-th day ts0 can improve the accuracy of the predicted target quantity of the first time period on the t-th day.

[0186] Exemplarily, if holidays and working days are distinguished, when the t-th day is a working day, the target quantity of the first time period on the t-th day is related to one or more of the following: the application usage quantity in the first time period of working days in the L days before the t-th day, the application usage quantity in the first time period in the K working days before the t-th day, and the application usage quantity in the Q time periods belonging to working days before the first time period on the t-th day.

[0187] In the case where the t-th day is a holiday, the target quantity in the first time period of the t-th day is related to one or more of the following: the application usage quantity in the first time period of holidays in the L days before the t-th day, the application usage quantity in the first time period of the K holidays before the t-th day, and the application usage quantity in the Q time periods belonging to holidays before the first time period of the t-th day.

[0188] Exemplarily, the target quantity N in the first time period of the t-th day ts satisfies the following formula (4):

[0189]

[0190] where N ti,s is the application usage quantity in the s-th time period of the i-th target day before the t-th day. There are i - 1 target days between the i-th target day before the t-th day and the t-th day. N t,sj is the application usage quantity in the j-th time period before the s-th time period of the t-th day. There are j - 1 time periods belonging to the target day between the j-th time period before the s-th time period of the t-th day and the s-th time period of the t-th day. The duration of the j-th time period and the duration of the s-th time period may be the same. γ1 is a weight coefficient. γ2 is a weight coefficient. γ3 is a weight coefficient, and γ3 can be 0.005. σ tsL 2 is the variance of the application usage quantity in the s-th time period of the target days in the L days before the t-th day. σ tsL 2 *γ3 represents the weighting of the variance σ tsL 2 The target day satisfies: being a working day when the t-th day is a working day, and being a holiday when the t-th day is a holiday. * is the multiplication sign. For example, σ tsL 2 *γ3 represents σ tsL 2 ×γ3.

[0191] Since the habits of users using applications on working days may be different from those on holidays, therefore, as shown in formula (4), based on the application usage situation in the s-th time period of the K target days before the t-th day (such as ), the application usage situation in the Q time periods belonging to the target day before the s-th time period of the t-th day (such as ), and the habit stability of users using applications in the s-th time period of the target day (such as σ tsL 2 ), predicting the target quantity N in the first time period of the t-th day ts can further improve the accuracy of the predicted target quantity in the first time period of the t-th day.

[0192] S406. Obtain a target application list corresponding to a prediction period based on a prediction application list corresponding to the prediction period and a target quantity of the prediction period, and use the target application list corresponding to the prediction period to kill applications running in the background.

[0193] Exemplarily, based on the first prediction application list and the target quantity N of the first time period of the t-th day ts , obtain a first target application list corresponding to the first time period of the t-th day, and use the first target application list to kill applications running in the background. The first target application list can also be referred to as the target application list corresponding to the first recommended operation moment.

[0194] For example, in the case of obtaining the first prediction application list and the target quantity of the first time period of the t-th day, the electronic device can intercept the first N ts applications from the first prediction application list to form the first target application list, and use the first target application list to kill applications running in the background. Since N ts < N0, when the user's habit of using applications in the first time period is stable, compared with using the first prediction application list to kill applications running in the background, using the first target application list to kill applications running in the background can increase the probability that applications not used by the user in the first time period are killed, and thus can reduce the power consumption of the electronic device.

[0195] Optionally, the target quantity of the first time period of the t-th day can also be N ts0 , and N ts0 < N0. The specific implementation principle and technical effect in the case where the target quantity of the first time period of the t-th day is N ts0 are similar to the specific implementation principle and technical effect in the case where the target quantity of the first time period of the t-th day is N ts , and will not be elaborated here.

[0196] S407. Use the prediction application list corresponding to the prediction period to kill applications running in the background.

[0197] Exemplarily, in the case where the first variance is greater than the variance threshold or the first standard deviation is greater than the standard deviation threshold, using the first prediction application list to kill applications running in the background can reduce the occurrence probability that applications used by the user in the first time period are killed, and can improve the user experience.

[0198] Optionally, when the first variance is greater than the variance threshold and the first similarity is greater than the similarity threshold, or when the first standard deviation is greater than the standard deviation threshold and the first similarity is greater than the similarity threshold, the first predicted application list is used to kill the applications running in the background, which can reduce the probability that the applications that the user will use in the first time period are killed, and can improve the user experience.

[0199] In the application control method provided by the embodiments of the present application, when receiving a first recommendation operation, the electronic device can obtain first data corresponding to a first time period to which the first recommendation operation time belongs, and input the first data into a multi-channel recall model to obtain a first predicted application list. When it is determined that the user's habit of using applications in the first time period is stable, the electronic device can predict the number of applications that the user may use within the first time period to which the first recommendation operation time belongs, and obtain the target number (such as N ts ) of the first time period to which the first recommendation operation time belongs. The electronic device can intercept the first N ts applications in the first predicted application list to form a first target application list. The electronic device can use the first target application list to kill the applications running in the background, which can increase the probability that the applications that the user will not use within the first time period to which the first recommendation operation time belongs are killed, and thus can reduce the power consumption of the electronic device and reduce the probability of the short battery life of the electronic device. When it is determined that the user's habit of using applications in the first time period is unstable or not stable, the electronic device can use the first predicted application list to kill the applications running in the background, which can reduce the probability that the applications that the user will use in the first time period to which the first recommendation operation time belongs are killed, and thus can improve the user experience.

[0200] Figure 5 FIG. shows a schematic structural diagram of a multi-channel recall model provided by the embodiments of the present application.

[0201] As Figure 5 shown, the multi-channel recall model includes a recall layer and a sorting layer. The recall layer may include a first-channel recall using a classification and regression tree (CART) algorithm, a second-channel recall using a recent popular recall algorithm, and a third-channel recall using a time decay algorithm. The first-channel recall may be referred to as CART decision tree recall. The second-channel recall may be referred to as recent popular recall. The third-channel recall may be referred to as recent usage recall.

[0202] In a possible implementation manner, the electronic device may be in a cold start stage or may be in a normal processing stage.

[0203] Exemplarily, still taking the time when the first recommended operation moment belongs to the first time period of the t-th day as an example, if the sum β of the application usage quantities in the first time period in the L days before the t-th day satisfies 201 ≤ β ≤ 800, it indicates that the electronic device is in the cold start phase. Exemplarily, the number of days from the date when the user first uses the electronic device to the t-th day is less than or equal to U, which can also indicate that the electronic device is in the cold start phase.

[0204] If the sum β of the application usage quantities in the first time period in the L days before the t-th day satisfies 800 < β, it indicates that the electronic device is in the cold start phase. Exemplarily, the number of days from the date when the user first uses the electronic device to the t-th day is greater than U, which can also indicate that the electronic device is in the normal phase.

[0205] Optionally, the manner in which the electronic device in the normal phase obtains the predicted application list for the prediction period may be different from the manner in which the electronic device in the cold start phase obtains the predicted application list for the prediction period.

[0206] Exemplarily, still taking the time when the first recommended operation moment belongs to the first time period of the t-th day as an example.

[0207] For the scenario where the electronic device is in the normal phase, when the electronic device receives the first recommended operation, as Figure 4 shown in S402 in the embodiment, the electronic device can process the first long-term data, the first recent data, and the first real-time data using the three-way recall in the multi-way recall model, and process and sort the results of the three-way recall to obtain the first predicted application list. Exemplarily, the first data obtained by the electronic device may include a data set representing the first long-term data, a data set representing the first recent data, and a data set representing the first real-time data. When the electronic device obtains the first data, if the electronic device identifies that the date corresponding to the first long-term data in the first data is different from the date corresponding to the first recent data, or the number of dates corresponding to the first long-term data is more than the number of dates corresponding to the first recent data, it can indicate that the electronic device is in the normal phase. The electronic device can mark the data set representing the first long-term data as the first identifier, mark the data set representing the first recent data as the second identifier, and mark the data set representing the first real-time data as the third identifier. When the electronic device inputs the first data into the preset model, the first-way recall can process the data set with the first identifier, the second-way recall can process the data set with the second identifier, and the third-way recall can process the data set with the third identifier.

[0208] For the scenario where the electronic device is in the cold start phase, when the electronic device receives the first recommendation operation, since the number of days from the first time the user uses the electronic device to the t-th day is less than or equal to U. When the electronic device obtains the first data, if the electronic device identifies that the date corresponding to the first long-term data in the first data is the same as the date corresponding to the first recent data, or the number of dates corresponding to the first long-term data in the first data is the same as the number of dates corresponding to the first recent data, it can indicate that the electronic device is in the cold start phase, and it can also indicate that the first long-term data and the first recent data in the first data are the same. The electronic device can clear the data set representing the first long-term data, mark the data set representing the first recent data as the second identifier, and mark the data set representing the first real-time data as the third identifier.

[0209] When the electronic device inputs the first data into the preset model, the second recall can process the data set with the second identifier to obtain the second weights of each application corresponding to the first recent data. The third recall can process the data set with the third identifier to obtain the third weights of each application corresponding to the first real-time data.

[0210] The sorting layer in the multi-channel recall model obtains the fifth weights of each application based on the second weights and the third weights of each application. The sorting layer in the multi-channel recall model sorts the multiple applications corresponding to the first data in the order of the fifth weights of the applications to obtain the first sorted list. The electronic device can intercept the first N0 applications in the first sorted list to form the first predicted application list.

[0211] Among them, the specific implementation principle of the second weight of the application can refer to Figure 4 the specific implementation principle of the second weight of the application in S402 in Figure 4 the specific implementation principle of the third weight of the application in S402 in

[0212] Since the first recall has requirements for the amount of data to be processed, if the amount of data processed by the first recall is less than or equal to the amount of data processed by the second recall, it may lead to poor accuracy of the processing results obtained by the first recall. The amount of the first long-term data in the cold start stage of the electronic device is the same as the amount of the first recent data, and the amount of the first long-term data does not meet the data volume requirement of the first recall for the data to be processed. Therefore, in the scenario where the electronic device is in the cold start stage, the first long-term data is cleared, and the multi-channel recall model processes and sorts the first recent data and the first real-time data, which can reduce the impact of the processing results with low accuracy obtained by the first recall on the accuracy of the first prediction application list, and thus an accurate first prediction application list can be obtained.

[0213] An embodiment of the present application further provides an application control method, which is applied to an electronic device. The method includes:

[0214] In the first time period of the t-th day, M applications are running in the background of the electronic device, and the M applications are killed using the first application list.

[0215] In the second time period of the t-th day, N applications are running in the background of the electronic device, and the N applications are killed using the second application list. The number of applications in the second application list is different from the number of applications in the first application list. The first time period is different from the second time period, and both M and N are integers.

[0216] Exemplarily, taking the first application list as the one obtained by the electronic device in response to the first recommendation operation, and the first recommendation operation moment belonging to the s-th time period of the t-th day as an example, the first application list can be the first target application list. For example, in response to the first recommendation operation, the electronic device can execute Figure 4 S401 - S406 in, or execute Figure 4 S401 - S403, S405 and S406 in, or execute Figure 4 S401 - S402 and S404 - S406 in, so as to realize that in the first time period, the electronic device uses the first target application list to kill the M applications, which can increase the probability of killing the applications that the user will not use in the first time period of the t-th day, and thus reduce the power consumption of the electronic device.

[0217] It can be understood that the first application list can also be the first prediction application list. For example, in response to the first recommendation operation, the electronic device can execute Figure 4 S401 - S403 and S407 in, so as to realize that in the first time period, the electronic device uses the first prediction application list to kill the M applications, which can reduce the occurrence probability of killing the applications that the user will use in the first time period of the t-th day.

[0218] Exemplarily, taking the second application list as the one obtained by the electronic device in response to the second recommendation operation, and the moment of the second recommendation operation belonging to the second time period of the t-th day as an example, the second application list can be a second predicted application list. For example, in response to the second recommendation operation, the electronic device can use a multi-channel recall model to process the training set model features corresponding to the second time period of the t-th day to obtain the second predicted application list. In the case where it is determined that the user's habit of using applications in the second time period is unstable, the electronic device uses the second predicted application list to kill N applications. Among them, the specific implementation principle and technical effect of the electronic device using the multi-channel recall model to process the training set model features corresponding to the second time period of the t-th day to obtain the second predicted application list are similar to those of the electronic device using the multi-channel recall model to process the first data to obtain the first predicted application list, and will not be elaborated here.

[0219] The specific implementation principle and technical effect of the electronic device using the second predicted application list to kill N applications in the second time period of the t-th day are similar to those of the electronic device using the first predicted application list to kill M applications in the first time period of the t-th day, and will not be elaborated here.

[0220] It can be understood that the second application list can also be a second target application list. For example, in response to the second recommendation operation, the electronic device can obtain the second predicted application list. In the case where it is determined that the user's habit of using applications in the second time period is stable, the electronic device can intercept the target number of applications in the second time period of the t-th day from the second predicted application list to form the second target application list, and use the second target application list to kill N applications.

[0221] Among them, the specific implementation principle of the electronic device to obtain the target number of applications in the second time period of the t-th day is similar to that of the electronic device to obtain the target number of applications in the first time period of the t-th day. The specific implementation principle of the electronic device to determine whether the user's habit of using applications in the second time period is stable is similar to that of the electronic device to determine whether the user's habit of using applications in the first time period is stable, and will not be elaborated here.

[0222] It can be understood that the specific implementation principle and technical effect of the electronic device using the second target application list to kill N applications in the second time period of the t-th day are similar to those of the electronic device using the first target application list to kill M applications in the first time period of the t-th day, and will not be elaborated here.

[0223] Exemplarily, as shown in Table 4, the application control method provided in the embodiments of the present application can obtain application lists for killing background applications in different time periods.

[0224] Applications in the application list corresponding to different time periods in Table 4

[0225]

[0226] It should be understood that the time periods shown in Table 4 may be time periods belonging to the same day. The time periods shown in Table 4 may also belong to time periods of different days. The application list for each time period shown in Table 4 is the application list corresponding to a moment in each time period. The application lists corresponding to different moments within a time period may be different. For example, the application list corresponding to the first time period is the application list corresponding to a moment in the first time period.

[0227] As shown in Table 4, the application control method provided by the embodiments of the present application can achieve obtaining the application lists corresponding to different time periods, and the number of applications in the application lists corresponding to different time periods is different. Using the corresponding application list to kill the applications running in the background in different time periods can not only increase the probability that applications not used by the user are killed, thereby reducing the power consumption of the electronic device, but also reduce the probability of occurrence that applications used by the user are killed, improving the user experience.

[0228] Next, taking the first application list as the first target application list and the second application list as the second predicted application list as an example, the application control method provided by the embodiments of the present application will be described.

[0229] Optionally, the first application list is obtained through the following method:

[0230] When reaching the first time period of the t-th day, use a preset model to predict the first predicted application list for the first time period of the t-th day. Select a target number of applications from the first predicted application list to obtain the first application list.

[0231] Wherein, the preset model is used to output the first predicted application list when inputting the application usage data in the first time period in the L days before the t-th day and the application usage data of using the application R times before the first recommended operation moment. The first recommended operation moment is the moment when reaching the first time period of the t-th day. The application usage data includes the start time when the application runs in the foreground and the application identifier. The target number is related to one or more of the following: the number of applications used in the first time period in the L days before the t-th day, the number of applications used in the first time period in the K days before the t-th day, the number of applications used in Q time periods before the first time period of the t-th day, and the Q time periods do not belong to the first time period. It can be understood that the first application list may be the application list corresponding to the first recommended operation moment.

[0232] Exemplarily, the preset model may be Figure 3The multi-channel recall model shown in the embodiments. In the embodiments of the present application, a preset model is used to predict a first predicted application list for a first time period, and a target number of applications are selected from the first predicted application list to obtain the specific implementation principle of the first application list, which is the same as that of Figure 3 the specific implementation principle of the electronic device obtaining the first target application list in the embodiments, and will not be elaborated here.

[0233] In this way, the preset model considers the order of magnitude of the possible application usage probabilities of the user in the first time period. The first application list formed by selecting a target number of applications from the first predicted application list output by the preset model also considers the possible number of application usages of the user in the first time period, improving the accuracy of the first application list, and can increase the probability of killing applications that the user will not use in the first time period, so as to reduce the power consumption of the electronic device.

[0234] Optionally, the target number of the first time period is positively correlated with one or more of the following: the number of application usages in the first time period in the L days before the t-th day, the number of application usages in the first time period in the K days before the t-th day, and the number of application usages in the Q time periods before the first time period of the t-th day.

[0235] Exemplarily, the target number of the first time period is positively correlated with one or more of the following, and the target number of the first time period can satisfy Figure 4 Formula (3) in the embodiments.

[0236] In this way, predicting the target number of the first time period based on the number of application usages in the first time period in the L days before the t-th day, the number of application usages in the first time period in the K days before the t-th day, and the number of application usages in the Q time periods before the first time period of the t-th day can improve the prediction accuracy of the target number of the first time period.

[0237] Optionally, when the t-th day is a working day, the target number is related to one or more of the following: the number of application usages in the first time period of working days in the L days before the t-th day, the number of application usages in the first time period in the K working days before the t-th day, and the number of application usages in the Q time periods belonging to working days before the first time period of the t-th day.

[0238] When the t-th day is a holiday, the target number is related to one or more of the following: the number of application usages in the first time period of holidays in the L days before the t-th day, the number of application usages in the first time period in the K holidays before the t-th day, and the number of application usages in the Q time periods belonging to holidays before the first time period of the t-th day.

[0239] In this way, since the usage habits of users when using the application on weekdays may be different from those when using the application on holidays, differentiating the date type of the t-th day can further improve the prediction accuracy of the target quantity in the first time period.

[0240] Optionally, the first time period of the t-th day is the s-th time period of the t-th day. The target quantity N of the first time period ts satisfies formula (4).

[0241] In this way, based on the application usage in the s-th time period among the K target days before the t-th day (such as ), the application usage in the Q time periods belonging to the target days before the s-th time period of the t-th day (such as ), and the habit stability of the user using the application in the s-th time period of the target day (such as σ tsL 2 ), predicting the target quantity N ts of the first time period of the t-th day can further improve the accuracy of the predicted target quantity of the first time period of the t-th day.

[0242] Optionally, the multiple applications in the first prediction list are arranged in the order of the magnitude of the application usage probability.

[0243] Selecting the applications with the target quantity in the first prediction application list to obtain the first application list, including: intercepting the first N ts applications from the first prediction application list to form the first application list. N ts is the target quantity of the first time period of the t-th day.

[0244] In this way, it can be realized that the applications in the first application list are also arranged in the order of the magnitude of the application usage probability. When the electronic device uses the first application list to kill the background-running applications, if it is necessary to kill the background-running applications with the application identifiers in the first application list in order to make the electronic device have sufficient available memory, the electronic device can preferentially kill the applications at the end of the first application list. It should be understood that the usage probability of the applications at the end of the first application list is less than that of the applications at the top of the first application list, so as to reduce the occurrence probability of the applications that the user will use being killed.

[0245] Optionally, before using the first application list to kill M applications, the method further includes:

[0246] Determine whether the first similarity corresponding to the first time period on the t-th day is greater than the similarity threshold, and / or whether the first dispersion corresponding to the first time period on the t-th day is greater than the dispersion threshold. The first similarity represents the similarity between the number of application usages in the first time period on the t-th day and the number of application usages in the time period before the first time period on the t-th day. The first dispersion represents the difference between the number of application usages in the first time period on the t-th day and the number of application usages in the first time period on multiple days before the t-th day.

[0247] Use the first application list to detect and kill M applications, including:

[0248] In the case where the first similarity is less than or equal to the similarity threshold, and / or the first dispersion is less than or equal to the dispersion threshold, use the first application list to detect and kill M applications.

[0249] In this way, if the first similarity is less than or equal to the similarity threshold, and / or the first dispersion is less than or equal to the dispersion threshold, it can indicate that the user's habit of using applications in the first time period is stable. Using the first application list for detection and killing can increase the probability of detecting and killing applications that the user will not use in the first time period on the t-th day, and reduce the power consumption of the electronic device.

[0250] Optionally, the second time period on the t-th day belongs to the time period on the t-th day. The second application list includes a second predicted application list, which is predicted using a preset model when reaching the second time period on the t-th day. The preset model is also used to output the second predicted application list when inputting the application usage data in the second time period in the L days before the t-th day and the application usage data of the R times of using applications before the second recommended operation time. The second recommended operation time is the time when reaching the second time period on the t-th day.

[0251] Before using the second application list to detect and kill N applications, the method further includes:

[0252] Determine whether the second similarity corresponding to the second time period on the t-th day is greater than the similarity threshold, and / or whether the second dispersion corresponding to the second time period on the t-th day is greater than the dispersion threshold. The second similarity represents the similarity between the number of application usages in the second time period on the t-th day and the number of application usages in the time period before the second time period on the t-th day. The second dispersion represents the difference between the number of application usages in the second time period on the t-th day and the number of application usages in the second time period on multiple days before the t-th day.

[0253] Use the second application list to detect and kill N applications, including:

[0254] When the second similarity is greater than the similarity threshold and / or the second dispersion is greater than the dispersion threshold, the second predicted application list is used to detect and kill N applications.

[0255] In this way, if the second similarity is greater than the similarity threshold and / or the second dispersion is greater than the dispersion threshold, it can indicate that the user's habit of using applications in the second time period is unstable. Using the second predicted application list for detection and killing can reduce the probability that the applications the user will use in the second time period on the t-th day are detected and killed, thus improving the user experience.

[0256] Optionally, the first dispersion includes the first variance or the first standard deviation. The first variance is the variance of the application usage quantity in the first time period in the L days before the t-th day. The first standard deviation is the standard deviation of the application usage quantity in the first time period in the L days before the t-th day.

[0257] The first similarity is related to the application usage quantity in the first time period in the K days before the t-th day and the application usage quantity in the first Q time periods before the first time period on the t-th day.

[0258] In this way, based on the variance or standard deviation of the application usage quantity in the first time period in the L days before the t-th day, the stability of the user's habit of using applications in the first time period can be judged, and an accurate judgment of the stability of the user's habit of using applications can be achieved. Based on the first similarity, the stability of the user's habit of using applications in the first time period can be judged. Based on the application usage quantity in the first time period in the K days before the t-th day and the application usage quantity in the first Q time periods before the first time period on the t-th day, the first similarity can be determined, and thus an accurate judgment of the stability of the user's habit of using applications can be achieved.

[0259] Optionally, when the t-th day is a working day, the first variance is the variance of the application usage quantity in the first time period of working days in the L days before the t-th day, and the first standard deviation is the standard deviation of the application usage quantity in the first time period of working days in the L days before the t-th day. The first similarity is related to the application usage quantity in the first time period in the K working days before the t-th day and the application usage quantity in the first Q time periods before the first time period on the t-th day.

[0260] When the t-th day is a holiday, the first variance is the variance of the application usage quantity in the first time period of holidays in the L days before the t-th day, and the first standard deviation is the standard deviation of the application usage quantity in the first time period of holidays in the L days before the t-th day. The first similarity is related to the application usage quantity in the first time period in the K holidays before the t-th day and the application usage quantity in the first Q time periods before the first time period on the t-th day.

[0261] In this way, by distinguishing the date type of the t-th day, the accuracy of judging the stability of the user's habit of using applications can be further improved.

[0262] Optionally, the first time period of the t-th day is the s-th time period of the t-th day. The first similarity N wts Satisfies formula (2).

[0263] Since the application usage in the first time period during the K days before the t-th day and the application usage in the first Q time periods of the first time period of the t-th day will affect the application usage of the user in the first time period of the t-th day. Thus, based on the number of application usages in the first time period during the K days before the t-th day, the number of application usages in the first Q time periods of the first time period of the t-th day, and the number of applications that the user has used both in the first time period during the K days before the t-th day and in the first Q time periods of the first time period of the t-th day, the first similarity is obtained, and the first similarity is used to judge the habit stability of the user's application usage in the first time period of the t-th day, which can improve the accuracy of the judgment.

[0264] Optionally, the first time period of the t-th day is the s-th time period of the t-th day. The first variance σ tsL 2 And the first standard deviation σ tsL Satisfies formula (1).

[0265] Thus, when using the first variance or the first standard deviation to judge the habit stability of the user's application usage in the first time period, the accuracy of the judgment can be improved.

[0266] Optionally, the first time period of the t-th day is the s-th time period of the t-th day, and the preset model is a multi-channel recall model.

[0267] Using the preset model to predict and obtain the first predicted application list for the first time period, including:

[0268] Obtaining the first data corresponding to the first time period. The first data includes first long-term data, first recent data, and first real-time data. The first long-term data includes the application usage data in the s-th time period of the target day during the L days before the t-th day, the recent data includes the application usage data in the s-th time period of the U target days before the t-th day, and the real-time data includes the application usage data of the R times of using the application before the first recommendation operation time.

[0269] Using the multi-channel recall model to process the first data in the following manner to obtain the first predicted application list:

[0270] The classification and regression tree CART algorithm in the multi-channel recall model is used to calculate the long-term data, and the first weight of each application corresponding to the long-term data is obtained. The most recently popular recall algorithm in the multi-channel recall model is used to calculate the recent data, and the second weight of each application corresponding to the recent data is obtained. The time decay algorithm in the multi-channel recall model is used to calculate the real-time data, and the third weight of each application corresponding to the real-time data is obtained.

[0271] The multiple applications corresponding to the first data are sorted in the order of the fourth weight of the application, and a first sorted list is obtained. The fourth weight of the application is the sum of the first weight, the second weight, and the third weight of the application. The fourth weight of the application is related to the first usage probability of the application.

[0272] The first N0 applications in the first sorted list are obtained to form a first predicted application list, where N0 is a preset value and N0 is greater than the target quantity.

[0273] Exemplarily, the specific implementation principle of the embodiment of the present application can be referred to Figure 4 the specific implementation principle of S402 in

[0274] In this way, the multi-channel recall model is implemented to adopt three-way recall to separately learn the habits of users using applications (APPs) in the long term, recently, and in real time, and also adopt the weights of each recall to achieve the fusion sorting of multi-channel recall. The multi-channel recall model comprehensively considers the habits of users using applications in the long term, recently, and in real time. When considering the long-term, recent, and real-time data, the accuracy of sorting the predicted application list output by the multi-channel recall model in the order of usage probability can be improved.

[0275] Optionally, the t-th day includes forty-eight time periods, and there is no overlapping time between two adjacent time periods among the forty-eight time periods, and the duration of each time period among the forty-eight time periods is the same.

[0276] It can be understood that if the t-th day includes forty-eight time periods, the duration of each time period can be 30 minutes (min). In this way, when receiving a recommendation operation, the electronic device can obtain the application list of the time period where the recommendation operation moment is located, and the electronic device can use the application list to kill the applications running in the background, which can reduce the power consumption of the electronic device while ensuring the accuracy of killing.

[0277] It should be noted that the module names involved in the embodiments of the present application can all be defined as other names, as long as the functions of each module can be realized, and no specific restrictions are imposed on the module names.

[0278] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the embodiments of the present application are all information and data authorized by the user or fully authorized by all parties. And the collection, use, and processing of relevant data need to comply with the relevant laws, regulations, and standards of relevant countries and regions, and corresponding operation entrances are provided for users to choose to authorize or refuse.

[0279] The application control method of the embodiments of the present application has been described above. Next, the device for executing the above method provided by the embodiments of the present application will be described. Those skilled in the art can understand that the method and the device can be combined and referenced with each other, and the relevant device provided by the embodiments of the present application can execute the steps in the above application control method.

[0280] The application control method provided by the embodiments of the present application can be applied to an electronic device with communication functions. The electronic device includes a terminal device, and the specific device form of the terminal device and the like can refer to the above relevant description, which will not be elaborated here.

[0281] The embodiments of the present application provide an electronic device, which includes: a processor and a memory. The memory stores computer-executable instructions. The processor executes the computer-executable instructions stored in the memory, so that the electronic device executes the above method.

[0282] The embodiments of the present application provide a chip. The chip includes a processor, and the processor is used to call a computer program in the memory to execute the technical solutions in the above embodiments. Its implementation principle and technical effects are similar to those of the above relevant embodiments, which will not be elaborated here.

[0283] The embodiments of the present application provide a chip system. The chip system includes at least one processor and a communication interface. The communication interface and the at least one processor are interconnected by a line, and the at least one processor is used to run a computer program or instruction to execute the above method.

[0284] The embodiments of the present application also provide a computer-readable storage medium. The computer-readable storage medium stores a computer program. When the computer program is executed by a processor, the above method is implemented. The methods described in the above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. If implemented in software, the functions can be stored as one or more instructions or codes on a computer-readable medium or transmitted on a computer-readable medium. The computer-readable medium can include a computer storage medium and a communication medium, and can also include any medium that can transfer a computer program from one place to another. The storage medium can be any target medium accessible by a computer.

[0285] In one possible implementation, the computer-readable medium may include RAM, ROM, compact disc read-only memory (CD-ROM), or other optical disc storage, magnetic disk storage, or any other medium targeted to carry or store the required program code in the form of instructions or data structures and accessible by a computer. Moreover, any connection is properly termed a computer-readable medium. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL), or wireless technologies such as infrared, radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of the medium. As used herein, disk and optical disc include optical disc, laser disc, optical disc, Digital Versatile Disc (DVD), floppy disk, and Blu-ray disc, where disks typically reproduce data magnetically, while optical discs use lasers to optically reproduce data. Combinations of the above should also be included within the scope of computer-readable media.

[0286] An embodiment of the present application provides a computer program product. The computer program product includes a computer program that, when run, causes a computer to execute the above method.

[0287] Embodiments of the present application are described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present application. It should be understood that each process and / or block in the flowcharts and / or block diagrams, and the combination of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processing unit of a general-purpose computer, special-purpose computer, embedded processor, or other programmable device to generate a machine such that the instructions executed by the processing unit of the computer or other programmable data processing device generate means for implementing the functions specified in one Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0288] The above specific implementation manners further elaborate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are only specific implementation manners of the present invention and are not used to limit the protection scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solution of the present invention should be included within the protection scope of the present invention.

Claims

1. An application control method, characterized in that, Applied to an electronic device, the method includes: In the first time period of the t-th day, M applications are running in the background of the electronic device, and the M applications are killed using a first application list; In the second time period of the t-th day, N applications are running in the background of the electronic device, and the N applications are killed using a second application list. The number of applications in the second application list is different from the number of applications in the first application list. The first time period is different from the second time period. Both M and N are integers.

2. The method according to claim 1, wherein The first application list is obtained through the following method: When reaching the first time period of the t-th day, a first predicted application list for the first time period of the t-th day is predicted using a preset model; Select a target number of applications from the first predicted application list to obtain the first application list; Wherein, the preset model is used to output the first predicted application list when inputting application usage data in the first time period in the L days before the t-th day and application usage data of using applications R times before the first recommendation operation time. The first recommendation operation time is the time when reaching the first time period of the t-th day. The application usage data includes the start time when the application runs in the foreground and the application identifier. The target number is related to one or more of the following: the number of applications used in the first time period in the L days before the t-th day, the number of applications used in the first time period in the K days before the t-th day, the number of applications used in Q time periods before the first time period of the t-th day, and the Q time periods do not belong to the first time period.

3. The method according to claim 2, wherein The target number is positively correlated with one or more of the following: the number of applications used in the first time period in the L days before the t-th day, the number of applications used in the first time period in the K days before the t-th day, the number of applications used in Q time periods before the first time period of the t-th day.

4. The method according to claim 2 or 3, wherein When the t-th day is a working day, the target number is related to one or more of the following: the number of applications used in the first time period of working days in the L days before the t-th day, the number of applications used in the first time period in the K working days before the t-th day, the number of applications used in Q time periods belonging to working days before the first time period of the t-th day; When the t-th day is a holiday, the target number is related to one or more of the following: the number of applications used in the first time period of holidays in the L days before the t-th day, the number of applications used in the first time period in the K holidays before the t-th day, the number of applications used in Q time periods belonging to holidays before the first time period of the t-th day.

5. The method according to any one of claims 2 - 4, characterized in that, The first time period of the t-th day is the s-th time period of the t-th day; the target quantity N ts satisfies the following formula: where N ti,s is the application usage quantity in the s-th time period of the i-th target day before the t-th day, and there are i - 1 target days between the i-th target day before the t-th day and the t-th day. N t,sj is the application usage quantity in the j-th time period before the s-th time period of the t-th day, and there are j - 1 time periods belonging to target days between the j-th time period before the s-th time period of the t-th day and the s-th time period of the t-th day. γ1 is a weight coefficient, γ2 is a weight coefficient, γ3 is a weight coefficient, σ tsL 2 is the variance of the application usage quantity in the s-th time period of target days in the L days before the t-th day. σ tsL 2 *γ3 represents the weighting of the variance σ tsL 2 The target days satisfy: when the t-th day is a working day, the target days are working days; when the t-th day is a holiday, the target days are holidays.

6. The method according to any one of claims 2-5, characterized in that, The multiple applications in the first predicted list are arranged in descending order of application usage probability; The selecting a target number of applications from the first predicted application list to obtain the first application list includes: Intercept the first N applications from the first predicted application list ts to form the first application list; The said N ts is the said target quantity.

7. The method according to any one of claims 2-6, characterized in that, Before killing the M applications using the first application list, the method further includes: Determining whether the first similarity corresponding to the first time period on the t-th day is greater than a similarity threshold, and / or whether the first dispersion corresponding to the first time period on the t-th day is greater than a dispersion threshold; the first similarity represents the similarity between the number of applications used in the first time period on the t-th day and the number of applications used in the time period before the first time period on the t-th day, and the first dispersion represents the difference between the number of applications used in the first time period on the t-th day and the number of applications used in the first time period on multiple days before the t-th day; Killing the M applications using the first application list includes: When the first similarity is less than or equal to the similarity threshold, and / or the first dispersion is less than or equal to the dispersion threshold, killing the M applications using the first application list.

8. The method according to claim 7, characterized in that, The second time period on the t-th day belongs to the time period on the t-th day; the second application list includes a second predicted application list, which is predicted using the preset model when reaching the second time period on the t-th day; the preset model is further configured to output the second predicted application list when inputting the application usage data in the second time period in the L days before the t-th day and the application usage data of the applications used R times before the second recommended operation time; the second recommended operation time is the time when reaching the second time period on the t-th day; Before killing the N applications using the second application list, the method further includes: Determining whether the second similarity corresponding to the second time period on the t-th day is greater than the similarity threshold, and / or whether the second dispersion corresponding to the second time period on the t-th day is greater than the dispersion threshold; the second similarity represents the similarity between the number of applications used in the second time period on the t-th day and the number of applications used in the time period before the second time period on the t-th day, and the second dispersion represents the difference between the number of applications used in the second time period on the t-th day and the number of applications used in the second time period on multiple days before the t-th day; Killing the N applications using the second application list includes: When the second similarity is greater than the similarity threshold, and / or the second dispersion is greater than the dispersion threshold, killing the N applications using the second predicted application list.

9. The method according to claim 7 or 8, characterized in that, The first dispersion includes a first variance or a first standard deviation; the first variance is the variance of the number of applications used in the first time period in the L days before the t-th day; the first standard deviation is the standard deviation of the number of applications used in the first time period in the L days before the t-th day; The first similarity is related to the number of applications used in the first time period in the K days before the t-th day and the number of applications used in the first Q time periods before the first time period on the t-th day.

10. The method according to claim 9, wherein when the t-th day is a working day, the first variance is the variance of the application usage quantities in the first time period of the working days in the L days before the t-th day, and the first standard deviation is the standard deviation of the application usage quantities in the first time period of the working days in the L days before the t-th day; the first similarity is related to the application usage quantities in the first time period of the K working days before the t-th day and the application usage quantities in the Q time periods before the first time period of the t-th day; when the t-th day is a holiday, the first variance is the variance of the application usage quantities in the first time period of the holidays in the L days before the t-th day, and the first standard deviation is the standard deviation of the application usage quantities in the first time period of the holidays in the L days before the t-th day; the first similarity is related to the application usage quantities in the first time period of the K holidays before the t-th day and the application usage quantities in the Q time periods before the first time period of the t-th day.

11. The method according to claim 10, wherein The first time period of the t-th day is the s-th time period of the t-th day; the first similarity N wts satisfies the following formula: N wts = max[N w,ti,s , N w,t,sj ​ Among them, N w,ti,s = N ti,s - A s , N w,t , sj = N t,sj - A s , N ti,s is the usage quantity of the application in the s-th time period of the i-th target day before the t-th day. There are i - 1 target days between the i-th target day and the t-th day. N t,sj is the usage quantity of the application in the j-th time period before the s-th time period of the t-th day. There are j - 1 time periods belonging to the target days between the j-th time period and the s-th time period of the t-th day. A s is the quantity of the application used in both the s-th time period and the j-th time period of the i-th target day. i = 1, 2,..., K, and j = 1, 2,..., Q. The target days satisfy that when the t-th day is a working day, the target days are working days, and when the t-th day is a holiday, the target days are holidays.

12. The method according to any one of claims 9-11, characterized in that, The first time period of the t-th day is the s-th time period of the t-th day; the first variance σ tsL 2 and the first standard deviation σ tsL satisfy the following formula: where x ts , p is the usage quantity of the s-th time period of the p-th target day in the L days before the t-th day, n is the number of target days in the L days before the t-th day, the target days satisfy that: when the t-th day is a working day, the target days are working days; when the t-th day is a holiday, the target days are holidays.

13. The method according to any one of claims 2-12, characterized in that, The first time period of the t-th day is the s-th time period of the t-th day, and the preset model is a multi-channel recall model; The first predicted application list for the first time period obtained by predicting using the preset model includes: obtaining first data corresponding to the first time period; the first data includes first long-term data, first recent data, and first real-time data, the first long-term data includes the application usage data in the s-th time period of the target days in the L days before the t-th day, the recent data includes the application usage data in the s-th time period of the U target days before the t-th day, and the real-time data includes the application usage data of the applications used R times before the first recommended operation moment; processing the first data by using the multi-channel recall model in the following manner to obtain the first predicted application list: calculating, by using the Classification and Regression Tree (CART) algorithm in the multi-channel recall model, the first weights of the respective applications corresponding to the long-term data; and calculating, by using the nearest popular recall algorithm in the multi-channel recall model, the second weights of the respective applications corresponding to the recent data; and also calculating, by using the time decay algorithm in the multi-channel recall model, the third weights of the respective applications corresponding to the real-time data; sorting the multiple applications corresponding to the first data in the order of the magnitudes of the fourth weights of the applications to obtain a first sorted list; the fourth weight of an application is the sum of the first weight, the second weight, and the third weight of the application; the fourth weight of the application is related to the first usage probability of the application; acquiring the first N0 applications in the first sorted list to form the first predicted application list, where N0 is a preset value and N0 is greater than the target quantity.

14. The method according to any one of claims 1-13, characterized in that, The t-th day includes forty-eight time periods, and there is no overlapping time between two adjacent time periods among the forty-eight time periods, and the durations of the respective time periods among the forty-eight time periods are the same.

15. An electronic device, characterized in that, including: A processor and a memory; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, such that the electronic device executes the method according to any one of claims 1-14.

16. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, the method according to any one of claims 1-14 is implemented.

17. A chip system, characterized in that, Comprising at least one processor and a communication interface, the communication interface and the at least one processor are interconnected by a line, and the at least one processor is configured to run a computer program or instructions to execute the method according to any one of claims 1-14.

18. A computer program product, characterized in that, Comprising a computer program, when the computer program is run, the computer is caused to execute the method according to any one of claims 1-14.

Images