CN107632697B - Application processing method and device, storage medium and electronic equipment - Google Patents

Abstract

The embodiment of the present application discloses a method, device, storage medium and electronic device for processing an application. The method for processing the application obtains usage information of a sample application at each sampling time point in a historical time period, generates training samples according to the sampling time point and usage information, and then trains a preset mixed Gaussian model according to the training samples, and processes the background application in the electronic device based on the trained mixed Gaussian model and the preset Bayesian model. This solution can reduce the occupation of terminal resources of the electronic device, improve the running smoothness of the electronic device, and reduce the power consumption of the electronic device.

Description

Application processing method, device, storage medium and electronic device

technical field

The present application relates to the technical field of electronic devices, and in particular, to a method, device, storage medium and electronic device for processing an application program.

Background technique

With the development of the Internet and the development of mobile communication networks, as well as the rapid development of the processing and storage capabilities of electronic devices, a large number of applications have been rapidly disseminated and used; commonly used applications are convenient for users to work and live. Many newly developed applications have also entered the daily life of users, improving the quality of life of users, the frequency of using the terminal, and the sense of entertainment in use.

When an electronic device has multiple application programs turned on, the application programs running in the background will seriously occupy the resources of the electronic device, reduce the running smoothness of the electronic device, and at the same time lead to high power consumption of the electronic device.

SUMMARY OF THE INVENTION

The embodiments of the present application provide an application processing method, apparatus, storage medium and electronic device, which can intelligently control the application and reduce the power consumption of the electronic device.

In a first aspect, an embodiment of the present application provides a method for processing an application program, which is applied to an electronic device, and the method includes:

Obtain the usage information of the sample application at each sampling time point in the historical time period;

generating training samples according to the sampling time point and the usage information;

training a preset Gaussian mixture model according to the training sample;

The background application in the electronic device is processed based on the trained mixture Gaussian model and the preset Bayesian model.

In a second aspect, an embodiment of the present application provides an apparatus for processing an application program, which is applied to an electronic device, and the apparatus includes:

The acquisition module is used to acquire the usage information of the sample application at each sampling time point in the historical time period;

a generating module, configured to generate training samples according to the sampling time point and the usage information;

a training module for training a preset Gaussian mixture model according to the training sample;

The processing module is used for processing the background application in the electronic device based on the trained mixture Gaussian model and the preset Bayesian model.

In a third aspect, an embodiment of the present application further provides a storage medium, where a plurality of instructions are stored in the storage medium, and the instructions are suitable for being loaded by a processor to execute the above processing method of an application program.

In a fourth aspect, an embodiment of the present application further provides an electronic device, including a processor and a memory, the processor is electrically connected to the memory, and the memory is used to store instructions and data; the processor is used to execute the above-mentioned The processing method of the application.

The embodiments of the present application disclose an application processing method, apparatus, storage medium, and electronic device. The processing method of the application program is to obtain the usage information of the sample application program at each sampling time point in the historical time period, generate training samples according to the sampling time point and the usage information, and then train the preset Gaussian mixture model according to the training samples. Based on the trained mixture Gaussian model and the preset Bayesian model, the background application in the electronic device is processed. The solution can reduce the occupation of final resources of the electronic device, improve the running smoothness of the electronic device, and reduce the power consumption of the electronic device.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present application more clearly, the following briefly introduces the drawings that are used in the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

FIG. 1 is a schematic diagram of a scenario architecture of an application processing method provided by an embodiment of the present application.

FIG. 2 is a schematic flowchart of a method for processing an application program provided by an embodiment of the present application.

FIG. 3 is another schematic flowchart of a method for processing an application program provided by an embodiment of the present application.

FIG. 4 is a schematic diagram of a Gaussian model provided by an embodiment of the present application.

FIG. 5 is a schematic diagram of training a Gaussian mixture model provided by an embodiment of the present application.

FIG. 6 is a schematic diagram of a Gaussian mixture model provided by an embodiment of the present application.

FIG. 7 is a schematic structural diagram of an apparatus for processing an application program provided by an embodiment of the present application.

FIG. 8 is another schematic structural diagram of an apparatus for processing an application program provided by an embodiment of the present application.

FIG. 9 is another schematic structural diagram of an apparatus for processing an application program provided by an embodiment of the present application.

FIG. 10 is another schematic structural diagram of an apparatus for processing an application program provided by an embodiment of the present application

FIG. 11 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

FIG. 12 is another schematic structural diagram of an electronic device provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present application.

Embodiments of the present application provide an application processing method, apparatus, storage medium, and electronic device. The detailed descriptions will be given below.

Please refer to FIG. 1. FIG. 1 is a schematic diagram of a scene architecture of an application processing method provided by an embodiment of the present application.

As shown in the figure, take the processing of applications A to E running in the background as an example. First, data collection is performed to record the usage information of electronic devices in various applications, such as recording the time when each application is opened within a month. Then, according to the collected usage records of the application programs, the usage probability of each application program at different times is counted, and the usage time and the corresponding usage probability are used as training samples to train the preset Gaussian mixture model. The sample adjusts the parameter information in the Gaussian mixture model to obtain the trained Gaussian mixture model corresponding to each application. Based on the trained mixture Gaussian model corresponding to each application and combined with a preset Bayesian model, the usage of the background application is predicted, and the usage probability of each background application at time T is calculated. Then, a target background application program whose usage probability is lower than the preset probability P is determined from the plurality of background application programs A to E, and the target background application program is closed. Thereby, the management and control of the background application program is realized based on the user's usage habits, and the resource occupation of the electronic device by the application program is reduced.

The electronic device may be a mobile terminal, such as a mobile phone, a tablet computer, a notebook computer, etc., which is not limited in this embodiment of the present application.

In one embodiment, a method for processing an application program is provided, which is applied to an electronic device, and the electronic device may be a mobile terminal such as a smart phone, a tablet computer, and a notebook computer. As shown in Figure 2, the process can be as follows:

101. Acquire usage information of the sample application at each sampling time point in the historical time period.

The application program mentioned in this embodiment may be any application program installed on the electronic device, such as an office application, a social application, a game application, a shopping application, and the like.

The sample application programs may be multiple or all installed application programs in the electronic device. The usage information of the application may be a usage record of the application, such as a record of the opening time of each application. The sampling time point can be set according to the actual needs. If you want to get higher accuracy results, you can set the sampling time point more densely, for example, every 1min is a sampling time point; if you want to save the resources of electronic equipment, If the accuracy of the result is not required, the sampling time point can be set loosely, for example, every 10min is a sampling time point.

In some embodiments, since the application is installed, the usage information of each installed application may be recorded, converted into corresponding data, and stored in a preset storage area. When the usage information of one or some applications needs to be used, the data corresponding to the one or some applications can be retrieved from the storage area, and the acquired data can be parsed to obtain the corresponding information, which can be used as The usage information of the one or some application programs, and the one or some application programs are used as a sample application program, and the usage information in the required time period can be selected from the acquired usage information.

In some embodiments, in order to reduce the power consumption of the electronic device and save the terminal resources of the electronic device, the required recording time period can be directly set, and then the usage information of the application program can be sampled at each sampling time point during the time period. Just record it for later use.

102. Generate a training sample according to the sampling time point and the usage information.

Specifically, the obtained usage information of the sample applications can be preprocessed, the usage probability of each sample application at different sampling time points can be calculated, and the probability distribution of the usage of each sample application over time can be obtained. , and generate training samples by one-to-one correspondence between sampling time points and usage probabilities.

103. Train a preset Gaussian mixture model according to the training sample.

Specifically, the above-generated training samples are input into the preset Gaussian mixture model, and the relevant parameters in the preset Gaussian mixture model are continuously modified according to the input training samples, so that the trained Gaussian mixture model can be applied to All training samples, and finally a Gaussian mixture model is trained for each sample application. Wherein, each Gaussian mixture model consists of multiple sub-Gaussian models.

104. Based on the trained mixture Gaussian model and the preset Bayesian model, process the background application in the electronic device.

In the embodiment of the present application, if there are N number of sample application programs, there are correspondingly N trained mixture Gaussian models. Obtain the identity information of each background application (such as application name, application identification, etc.), and select the target mixture Gaussian model from the N trained mixture Gaussian models (that is, for the background application according to the identity information of the background application). trained mixture Gaussian model), and process the background application based on the target mixture Gaussian model.

In some embodiments, the usage probability of each background application at this time may be calculated based on the respective trained mixture Gaussian models corresponding to different background applications and in combination with the current time. According to the calculated usage probability corresponding to each application program, the background application program whose usage probability satisfies a certain condition is cleaned or closed, so as to reduce the occupation of electronic device resources by the application program.

As can be seen from the above, the present application is a processing method of the application program provided by the embodiment. By obtaining the usage information of the sample application program at each sampling time point in the historical time period, a training sample is generated according to the sampling time point and the usage information, and then according to the training sample. The preset mixture Gaussian model is trained, and the background application in the electronic device is processed based on the trained mixture Gaussian model and the preset Bayesian model. The solution can reduce the occupation of final resources of the electronic device, improve the running smoothness of the electronic device, and reduce the power consumption of the electronic device.

In an embodiment, another method for processing an application program is also provided, which is applied to an electronic device, and the electronic device may be a mobile terminal such as a smart phone, a tablet computer, and a notebook computer. As shown in Figure 3, the process can be as follows:

201. Acquire usage information of the sample application at each sampling time point in the historical time period.

The sample applications may be multiple or all of the installed applications in the electronic device. The sampling time point can be set according to the actual needs. If you want to get higher accuracy results, you can set the sampling time point more densely, for example, every 1min is a sampling time point; if you want to save the resources of electronic equipment, you can There is no requirement for the accuracy of the sampling time point, and the sampling time point can be set loosely, for example, every 10min is a sampling time point. The usage information of the application can be related information during the use of the application.

For example, the historical period can be the past month, and each time point can be the timestamp of the current time. The usage parameters can be extracted from the database, which can store the opening records of the applications in the electronic device in the past month, as shown in Table 1 below:

application package name Timestamp of opening this app com.tencent.mobileqq 1457550655465 com.android.settings 1457605107522 ... ...

Table 1

Afterwards, the opening records of these applications are used as the usage information of each sample application at each sampling time point.

202. Determine a time period and a sampling period corresponding to each sampling time point, wherein the sampling time points in each time period are in one-to-one correspondence with the sampling period.

In some embodiments, the historical time period includes multiple time periods. If the historical time period is the past month, the time period may be every day of the past month. Each time period can be divided into multiple sampling periods, such as every minute of the day. Specifically, the time period to which it belongs and the specific sampling period may be determined based on the timestamp corresponding to the sampling time point, for example, it may be xx month xx day xx minutes. Taking September 9th as an example of 481 points, September is the historical time period, 9th is the time period, and 481 points are the sampling period.

Among them, the collection of samples can be completed on terminal devices such as smart phones and tablet computers, and the application information currently being used on the terminal device is obtained every 1 minute and stored in the database of the terminal device. Monthly usage records, tens of thousands of usage information samples can be extracted.

203. Process the usage information corresponding to the same sampling period in different time periods to obtain the sample usage probability corresponding to each sampling period of the sample application.

In some embodiments, the step of "processing the usage information corresponding to the same sampling period in different time periods to obtain the sample usage probability corresponding to the sample application in each sampling period" may include the following process:

Determine whether the usage information satisfies the preset conditions;

Determine the number of sampling time points at which the corresponding usage information of each sample application in the same sampling period satisfies the preset condition;

Obtain the total number of sampling time points where the usage information of each sample application meets the preset conditions in multiple time periods;

According to the number of sampling time points and the total number of sampling time points, calculate the sample usage probability corresponding to each sample application in each sampling period.

Specifically, N sample application programs most commonly used by the user may be counted according to the collected application program records of the user, where N is configurable. In order to reasonably allocate the resources of the electronic equipment and reduce the amount of computation, usually N=5.

In some embodiments, the usage information may be the running status information of the sample application; then the step "judging whether the usage information satisfies the preset condition" may include the following process:

Determine whether the running status is foreground running;

If so, it is determined that the usage information satisfies the preset condition;

If not, it is determined that the usage information does not meet the preset condition.

Among them, the running state is running in the foreground, which means that the current user is using the sample application. Then for these N sample applications, count the same time period in each day of each sample application in the past month (for example, a day can include 1440 minutes, then the 481st minute on September 1 and the 481st minute on September 31) The 481st minute is the same time period; the 1440th minute on September 1st and the 1440th minute on September 31st are the same time period) the number of sampling time points running in the foreground, denoted as X=[x ₁ , x ₂ , x ₃ ... x _i ..., x _n ], where x _i represents the number of times the application is used at the i-th minute of each day in September.

For example, taking the 30 days from September 1st to September 30th as an example of a historical time period, if within these 30 days, users used WeChat from 8:01 am to 8:10 am on 25 days, while other Time does not need WeChat. Then the statistical method is: convert 8:01 into a time slot as the 481st minute (8*60+1=481), and convert 8:10 into a time slot as the 490th minute (8*60+10=490) . Then the statistical results of the user's WeChat usage information can be shown in Table 2 below:

application x₁ ... x₄₈₁ ... x₄₉₀ ... x₁₄₄₀ WeChat 0 0 0 0 QQ ... ... ... ... ... ... ... … …

Table 2

In the embodiment of the present application, the probability of the sample usage probability corresponding to each sample application in each sampling period can be defined as P _i , and the specific algorithm of the probability P _i can refer to the following formula:

Among them, x _j and x _i have the same definitions, and both represent the number of times the application is used at the ith or jth minute in a day. n is a positive integer greater than 1. Based on the above data and probability algorithm, the probability distribution of the sample usage probability corresponding to each sample application in each sampling period can be obtained, as shown in Table 3 below:

application P₁ ... P₄₈₁ ... P₄₉₀ ... P₁₄₄₀ WeChat 0 0 0.1 0.1 0.1 0 0 QQ ... ... ... ... ... ... ... … …

table 3

204. Generate a training sample based on the sampling period and the corresponding sample usage probability.

Specifically, according to the probability distribution of the sample usage probability of each sample application program in Table 2 above, the training samples are generated by one-to-one correspondence between the sampling time point and the sample usage probability.

In some embodiments, if the sampling period is denoted as t, the sampling period includes [t ₁ , t ₂ ···t _m ], the sample usage probability is denoted as P, and the sample usage probability includes [P ₁ , P ₂ ···P _m ] ]. Specifically, the generated training sample may be recorded as (t _m , P _m ), for example, the training sample corresponding to the 481st minute is (481, 0.1).

205. Input the training samples into the first preset formula to train the first preset formula to obtain a plurality of trained sub-Gaussian models.

In the embodiment of the present application, the essence of the first preset formula is the probability spectral density function of the Gaussian mixture model, specifically as follows:

where A _i represents the sample application i, t represents the sampling period, k represents the number of sub-Gaussian models, μ _k represents the mathematical expectation, σ _k represents the variance, ω _k represents the weight, and N(t|μ _k ,σ _k ) represents the The random variable t obeys a normal distribution with a mathematical expectation of μ _k and a variance of σ _k , and P(t|A _i ) can represent the probability that the sample application i is running in the foreground when the sampling period is t.

is a Gaussian distribution probability model.

Referring to Figure 4, an initialized Gaussian model can be constructed. Then, based on the input sampling period t and the sample usage probability P, the first preset formula is trained to obtain a plurality of trained sub-Gaussian models. Referring to Figure 5, the collected data is first preprocessed to obtain the probability distribution used by each application, and then the probability distribution is used as input to train the preset Gaussian mixture model, and finally a suitable Gaussian mixture model is obtained.

For example, the Gaussian mixture model can be modeled when the training samples corresponding to the first minute are read; then the training samples corresponding to the second minute are read to update the parameters of the Gaussian model; the training samples corresponding to the third minute are read again, and the update is continued. Mixture Gaussian model parameters... and so on, until all training samples have been read, update the Gaussian model parameters to obtain the final trained Gaussian mixture model.

The mixture Gaussian model is generally composed of 3 to 5 sub-Gaussian models. In the modeling process, it is necessary to initialize some parameters such as variance σ _k , mathematical expectation μ _k , and weight ω _k in the Gaussian mixture model, and obtain the data required for modeling through these parameters. In the initialization process, the variance can be set as large as possible, and the weight (ie, ω _k ) can be set as small as possible (eg, 0.001). This setting is because the initialized Gaussian model is an inaccurate model, and it is necessary to continuously narrow its range and update its parameter values to obtain the most probable Gaussian model. Setting the variance larger is to include as many pixels as possible into a model, to find the parameter k, all the corresponding weights ω _k , and the corresponding parameters μ _k and σ _k in all sub-Gaussian models.

In some embodiments, the maximum likelihood estimation method can be used to determine these model parameters ω _k , μ _k and σ _k . Among them, the likelihood function of the mixture Gaussian model is:

The expectation maximization (EM) algorithm is used to maximize the likelihood function of (μ _k ,σ _k ). Then ω _k , μ _k and σ _k corresponding to the maximum value are our estimates. Finally [(ω ₁ , μ ₁ ,σ ₁ ), (ω ₁ , μ ₁ ,σ ₁ ),…(ω _k , μ _k ,σ _k )].

206. Superimpose multiple trained sub-Gaussian models to obtain a trained mixture Gaussian model.

Specifically, after each sub-Gaussian model is weighted according to the estimated weight ω _k , the weighted k sub-Gaussian models are superimposed to obtain a trained mixture Gaussian model. Referring to Fig. 6, the resulting mixture Gaussian model consists of 4 sub-Gaussian models.

Assuming that the user has N sample applications, there are N mixture Gaussian models, ie [P(t|A ₁ ), P(t|A ₂ ), ... P(t|A _N )].

207. Determine a background application in the electronic device.

In some embodiments, when a central processing unit (CPU, central processing unit) of the electronic device occupies a large amount, a running memory resource occupies a large amount, and/or the remaining power of the electronic device is insufficient, the application processing instruction may be triggered. The electronic device acquires the application processing instruction, and then determines a background application running in the background according to the application processing instruction, so as to process the background application subsequently.

208. Based on the trained mixture Gaussian model corresponding to each application, use a second preset formula to calculate the usage probability of each background application at the target time.

In the embodiment of the present application, each application program corresponds to a unique trained Gaussian mixture model. Based on the trained Gaussian mixture model, the corresponding usage probability of the application at different times can be accurately estimated. In the embodiment of the present application, the expression of the preset Bayesian model is the second preset formula, and the second preset formula is as follows:

Among them, T represents time, N represents the number of Gaussian mixture models after training, P(A _i |T) represents the probability that the application running in the foreground is application i when the sampling period is T, and P(T|A _i ) represents The running state of the sample application i is the probability that the sampling period is T when the running state is in the foreground, P(T|A _j ) represents the probability that the running state of the application j is the sampling period T when the running state is the foreground running, and P(A _i ) represents the application The application usage probability of the program i in the historical time period, P(A _j ) represents the application usage probability of the application program j in the historical time period.

Specifically, first, based on the trained Gaussian mixture model, the initial usage probability (ie, P(T|A _i )) corresponding to each application program at the target time is estimated. Next, the application usage probability (ie, P(A _i )) of the target background application i in the historical time period is calculated. Among them, P(A _i ) can be obtained from the data preprocessing period, which can be obtained by the ratio of the usage times of application i in the historical time period to the sum of the usage times of all sample applications in the historical time period, namely P(A _i ) The calculation formula is as follows:

Among them, S(A _i ) is the total usage times of application i in the historical time period, and S is the sum of the usage times of all sample applications in the historical time period.

Similarly, according to the algorithm of the initial use probability corresponding to the target background application i, using the Gaussian mixture model corresponding to each application, calculate the initial use probability of each application; using the calculation formula of P(A _i ), calculate Probability of application usage for each sample application over a historical time period.

Finally, the data obtained above are substituted into the second preset formula (that is, the preset Bayesian model), and the second preset formula is used to calculate the corresponding usage probability of the target background application under the target time, so as to improve the Use the precision of probabilities.

209. Process the background application according to the usage probability.

In some embodiments, a probability threshold may be set as a benchmark for processing an application. That is, the step "processing the background application according to the usage probability" may include the following flow:

Determine the target background application whose usage probability is less than the preset threshold from the current background applications;

Close the target background application.

Wherein, the preset threshold can be set by those skilled in the art or product manufacturers. For example, if the preset threshold is set to 0.5, then if the probability P(T|A _i ) of opening the background application A _i in the future period T is less than 0.5, the background application A _i will be cleaned up, if not less than 0.5, then Keep the background application A _i running in the background.

As can be seen from the above, the application processing method provided by the embodiment of the present application obtains the usage information of the sample application program at each sampling time point in the historical time period, and then determines the time period and sampling period corresponding to each sampling time point, and then determines the time period and sampling period corresponding to each sampling time point. The usage information corresponding to the same sampling period in different time periods is processed to obtain the sample usage probability corresponding to each sampling period of the sample application. A training sample is generated based on the sampling period and the corresponding sample usage probability, and is input into the preset Gaussian mixture model for model training, and a new Gaussian mixture model composed of multiple trained sub-Gaussian models is obtained. Finally, the new Gaussian mixture model and the preset Bayesian model are used to estimate the usage probability of each background application at the target time, and the corresponding background application is processed according to the obtained probability. The solution can reduce the occupation of final resources of the electronic device, improve the running smoothness of the electronic device, and reduce the power consumption of the electronic device.

In yet another embodiment of the present application, a processing device for an application program is also provided. The processing device for an application program can be integrated into an electronic device in the form of software or hardware, and the electronic device can specifically include a mobile phone, a tablet computer, a notebook computer and other equipment. As shown in FIG. 7 , the processing device 30 of the application may include an acquisition module 31, a generation module 32, a training module 33 and a processing module 34, wherein:

The acquisition module 31 is used to acquire the usage information of the sample application program at each sampling time point in the historical time period;

A generation module 32 is used to generate training samples according to the sampling time point and the usage information;

The training module 33 is used for training the preset Gaussian mixture model according to the training samples;

The processing module 34 is configured to process the background application in the electronic device based on the trained Gaussian mixture model and the preset Bayesian model.

In some embodiments, the historical time period includes a plurality of time periods, and each time period is divided into a plurality of sampling periods. 8, the generation module 32 may include:

The first determination sub-module 321 is used to determine the time period and the sampling period corresponding to each sampling time point, wherein the sampling time points in each time period correspond to the sampling period one-to-one;

The information processing sub-module 322 is used to process the usage information corresponding to the same sampling period in different time periods to obtain the sample usage probability corresponding to each sampling period of the sample application;

The generating sub-module 323 is configured to generate training samples based on the sampling period and the corresponding sample usage probability.

In some embodiments, the processing sub-module 322 may include:

a judging unit for judging whether the usage information satisfies a preset condition;

a first determining unit, configured to determine the number of sampling time points at which the usage information corresponding to each sample application in the same sampling period satisfies a preset condition;

an acquisition unit, configured to acquire the total number of sampling time points where the usage information of each sample application satisfies a preset condition in multiple time periods;

The calculation unit is configured to calculate the sample usage probability corresponding to each sample application program in each sampling period according to the number of sampling time points and the total number of sampling time points.

In some embodiments, the usage information is the running state information of the sample application; the judgment unit can be used to:

Determine whether the running status is foreground running;

If so, it is determined that the usage information satisfies the preset condition;

If not, it is determined that the usage information does not meet the preset conditions

In some embodiments, the sampling period includes [t ₁ , t ₂ . . . t _m ], and the sample usage probability includes [P ₁ , P ₂ . . . P _m ]; with reference to FIG. 9 , the training module 33 may include:

The input sub-module 331 is used to input the sampling period and the corresponding sample usage probability into the first preset formula, and the first preset formula is:

where A _i represents the sample application i, t represents the sampling period, k represents the number of sub-Gaussian models, μ _k represents the mathematical expectation, σ _k represents the variance, ω _k represents the weight, and N(t|μ _k ,σ _k ) represents the The random variable t obeys a normal distribution with a mathematical expectation of μ _k and a variance of σ _k , and P(t|A _i ) represents the probability that the sampling period is t when the running state of the sample application i is foreground running;

The training sub-module 332 is used for training the first preset formula based on the input sampling period t and the sample usage probability P to obtain a plurality of trained sub-Gaussian models;

The stacking sub-module 333 is used for stacking multiple trained sub-Gaussian models to obtain a trained mixture Gaussian model.

In some embodiments, each application corresponds to a unique trained Gaussian mixture model; with reference to FIG. 10 , the processing module 34 may include:

Obtaining sub-module 341 for obtaining application processing instructions;

The second determination submodule 342 is configured to determine the background application in the electronic device according to the application processing instruction;

The calculation sub-module 343 is used to calculate the use probability of each background application at the target time based on the Gaussian mixture model after training corresponding to each application program, and the second preset formula is:

Among them, T represents time, N represents the number of Gaussian mixture models after training, P(A _i |T) represents the probability that the application running in the foreground is application i when the sampling period is T, and P(T|A _i ) represents The running state of the sample application i is the probability that the sampling period is T when the running state is in the foreground, P(T|A _j ) represents the probability that the running state of the application j is the sampling period T when the running state is the foreground running, and P(A _i ) represents the application The application usage probability of program i in the historical time period, P(A _j ) represents the application usage probability of application program j in the historical time period;

The application processing sub-module 344 is used to process the background application according to the usage probability.

In some embodiments, the application processing sub-module 344 may include:

a second determining unit, configured to determine, from the current background application, a target background application whose usage probability is less than a preset threshold;

Shutdown unit, used to close the target background application.

As can be seen from the above, the application processing device provided by the embodiment of the present application generates training samples according to the sampling time point and the use information by acquiring the usage information of the sample application program at each sampling time point in the historical time period, and then performs the training samples according to the training samples. The preset mixture Gaussian model is trained, and the background application in the electronic device is processed based on the trained mixture Gaussian model and the preset Bayesian model. The solution can reduce the occupation of final resources of the electronic device, improve the running smoothness of the electronic device, and reduce the power consumption of the electronic device.

Another embodiment of the present application further provides an electronic device, where the electronic device may be a smartphone, a tablet computer, or the like. As shown in FIG. 11 , the electronic device 400 includes a processor 401 and a memory 402 . The processor 401 is electrically connected to the memory 402 .

The processor 401 is the control center of the electronic device 400, uses various interfaces and lines to connect various parts of the entire electronic device, executes the electronic device by running or loading the application stored in the memory 402, and calling the data stored in the memory 402. The various functions and processing data of the device are used to monitor the electronic equipment as a whole.

In this embodiment, the processor 401 in the electronic device 400 loads the instructions corresponding to the processes of one or more applications into the memory 402 according to the following steps, and is executed by the processor 401 and stored in the memory 402 application to achieve various functions:

Obtain the usage information of the sample application at each sampling time point in the historical time period;

Generate training samples based on sampling time points and usage information;

Train the preset Gaussian mixture model according to the training samples;

Based on the trained mixture Gaussian model and the preset Bayesian model, the background application in the electronic device is processed.

In some embodiments, the historical time period includes multiple time periods, and each time period is divided into multiple sampling periods; the processor 401 is further configured to perform the following steps:

Determine the time period and the sampling period corresponding to each sampling time point, wherein the sampling time point in each time period corresponds to the sampling period one-to-one;

Process the usage information corresponding to the same sampling period in different time periods to obtain the sample usage probability corresponding to each sampling period of the sample application;

Training samples are generated based on the sampling period and the corresponding sample usage probabilities.

In some embodiments, the processor 401 is further configured to perform the following steps:

Determine whether the usage information meets the preset conditions;

Determine the number of sampling time points at which the corresponding usage information of each sample application in the same sampling period satisfies the preset condition;

Obtain the total number of sampling time points where the usage information of each sample application meets the preset conditions in multiple time periods;

According to the number of sampling time points and the total number of sampling time points, calculate the sample usage probability corresponding to each sample application in each sampling period.

In some embodiments, the usage information is running state information of the sample application, and the processor 401 is further configured to perform the following steps:

Determine whether the running status is foreground running;

If so, it is determined that the usage information satisfies the preset condition;

If not, it is determined that the usage information does not meet the preset condition.

In some embodiments, the sampling period includes [t ₁ , t ₂ . . . t _m ], and the sample usage probability includes [P ₁ , P ₂ . . . P _m ]; the processor 401 is further configured to perform the following steps:

The sampling period and the corresponding sample usage probability are input into the first preset formula, and the first preset formula is:

where A _i represents the sample application i, t represents the sampling period, k represents the number of sub-Gaussian models, μ _k represents the mathematical expectation, σ _k represents the variance, ω _k represents the weight, and N(t|μ _k ,σ _k ) represents the The random variable t obeys a normal distribution with a mathematical expectation of μ _k and a variance of σ _k , and P(t|A _i ) represents the probability that the sampling period is t when the running state of the sample application i is foreground running;

Based on the input sampling period t and the sample usage probability P, the first preset formula is trained to obtain a plurality of trained sub-Gaussian models;

Stack multiple trained sub-Gaussian models to get a trained mixture Gaussian model.

In some embodiments, each application corresponds to a unique trained Gaussian mixture model; the processor 401 is further configured to perform the following steps:

Get application processing instructions;

Determine the background application in the electronic device according to the application processing instruction;

Based on the trained mixture Gaussian model corresponding to each application, use the second preset formula to calculate the usage probability of each background application at the target time. The second preset formula is:

Among them, T represents time, N represents the number of Gaussian mixture models after training, P(A _i |T) represents the probability that the application running in the foreground is application i when the sampling period is T, and P(T|A _i ) represents The running state of the sample application i is the probability that the sampling period is T when the running state is in the foreground, P(T|A _j ) represents the probability that the running state of the application j is the sampling period T when the running state is the foreground running, and P(A _i ) represents the application The application usage probability of program i in the historical time period, P(A _j ) represents the application usage probability of application program j in the historical time period;

Background applications are processed according to the probability of use.

In some embodiments, the processor 401 is further configured to perform the following steps:

Determine the target background application whose usage probability is less than the preset threshold from the current background applications;

Close the target background application.

Memory 402 may be used to store applications and data. The application stored in memory 402 contains instructions executable in the processor. Applications can be composed of various functional modules. The processor 401 executes various functional applications and data processing by executing applications stored in the memory 402 .

In some embodiments, as shown in FIG. 12 , the electronic device 400 further includes: a display screen 403 , a control circuit 404 , a radio frequency circuit 405 , an input unit 406 , an audio circuit 407 , a sensor 408 and a power supply 409 . The processor 401 is electrically connected to the display screen 403 , the control circuit 404 , the radio frequency circuit 405 , the input unit 406 , the audio circuit 407 , the sensor 408 and the power supply 409 respectively.

The display screen 403 may be used to display information input by or provided to the user and various graphical user interfaces of the electronic device, which may be composed of images, text, icons, videos, and any combination thereof.

The control circuit 404 is electrically connected to the display screen 403 for controlling the display screen 403 to display information.

The radio frequency circuit 405 is used for transmitting and receiving radio frequency signals, so as to establish wireless communication with network equipment or other electronic equipment through wireless communication, and send and receive signals with the network equipment or other electronic equipment.

Input unit 406 may be used to receive input numbers, character information, or user characteristic information (eg, fingerprints), and generate keyboard, mouse, joystick, optical, or trackball signal input related to user settings and function control. The input unit 406 may include a fingerprint identification module.

The audio circuit 407 can provide an audio interface between the user and the electronic device through speakers and microphones.

The sensor 408 is used to collect external environment information. Sensors 408 may include ambient brightness sensors, acceleration sensors, light sensors, motion sensors, and other sensors.

Power supply 409 is used to power various components of electronic device 400 . In some embodiments, the power supply 409 may be logically connected to the processor 401 through a power management system, so as to implement functions such as managing charging, discharging, and power consumption through the power management system.

Although not shown in FIG. 12 , the electronic device 400 may further include a camera, a Bluetooth module, and the like, which will not be repeated here.

It can be seen from the above that the electronic device provided by the embodiment of the present application generates training samples according to the sampling time point and the use information by acquiring the usage information of the sample application program at each sampling time point in the historical time period, and then performs the presetting of the training samples according to the training samples. The mixture Gaussian model is trained, and the background application in the electronic device is processed based on the trained mixture Gaussian model. The solution can reduce the occupation of final resources of the electronic device, improve the running smoothness of the electronic device, and reduce the power consumption of the electronic device.

In some embodiments, a storage medium is also provided, and a plurality of instructions are stored in the storage medium, and the instructions are suitable for being loaded by a processor to execute the processing method of any of the above application programs.

Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium can include: Read Only Memory (ROM, Read Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk, etc.

The terms "a" and "said" and similar words are used in describing concepts of the present application (particularly in the appended claims) and should be construed to encompass both the singular and the plural. Furthermore, unless otherwise indicated herein, recitation of numerical ranges herein is merely by way of shorthand to refer to each separate value belonging to the relevant range, and each separate value is incorporated into the specification as if These values are individually stated in this document. Additionally, the steps of all methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. Variations of the present application are not limited to the order of steps described. Unless otherwise claimed, any and all examples or exemplary language provided herein (eg, "for example") are used merely to better illustrate the concepts of the present application and are not intended to limit the scope of the concepts of the present application. Various modifications and adaptations will be readily apparent to those skilled in the art without departing from the spirit and scope.

The method, device, storage medium, and electronic device for processing the application provided by the embodiments of the present application have been described in detail above. The principles and implementations of the present application are described with specific examples. The descriptions of the above embodiments are only It is used to help understand the method and the core idea of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there will be changes in the specific embodiments and application scope. In summary, this specification The content should not be construed as a limitation on this application.

Claims (14)

1. a kind of processing method of application program is applied to electronic equipment, which is characterized in that the described method includes:

Obtain the use information of each sampling time point sample application program in historical time section；

Training sample is generated according to the sampling time point and the use information；

Preset mixed Gauss model is trained according to the training sample, wherein each application program is corresponding with uniquely Mixed Gauss model after training；

Obtain application program process instruction；

The background application in the electronic equipment is determined according to the application program process instruction；

Based on corresponding to each application program training after mixed Gauss model, using preset Bayesian model calculate it is each after Use probability of the platform application program in the object time, the expression formula of the Bayesian model are the second preset formula, described second Preset formula are as follows:

Wherein, T indicates the time, and N indicates the quantity of the mixed Gauss model after training, P (A_i| T) indicate sampling periods be T when before The probability that the application program of platform operation is application program i, and P (T | A_i) indicate the operating status of sample application program i for foreground fortune When row sampling periods be T probability, P (T | A_j) indicate that sampling periods are T's when the operating status of application program j is front stage operation Probability, P (A_i) indicate application program i in historical time section using probability, P (A_j) indicate application program j in history In period using probability；

The background application in the electronic equipment is handled using probability according to described.

2. the processing method of application program as described in claim 1, which is characterized in that when the historical time section includes multiple Between the period, every a period of time is divided into multiple sampling periods；

The step of generating training sample according to the sampling time point and the use information, comprising:

Determine each sampling time point corresponding time cycle and sampling periods, wherein sampling time point in every a period of time It is corresponded with sampling periods；

The corresponding use information of sampling periods identical in the different time period is handled, obtains sample application program each The corresponding sample of sampling periods uses probability；

Training sample is generated using probability based on sampling periods and corresponding sample.

3. the processing method of application program as claimed in claim 2, which is characterized in that by sampling identical in the different time period Period, corresponding use information was handled, and is obtained sample application program in the corresponding sample of each sampling periods and is used probability Step, comprising:

Judge whether the use information meets preset condition；

Determine that each sample applies the sampling time that corresponding use information meets preset condition in identical sampling periods to count Amount；

Obtain the sampling time point total quantity that each sample applies within multiple time cycles use information to meet preset condition；

According to the sampling time point quantity and the sampling time point total quantity, calculates each sample application program and adopted each Sample period corresponding sample uses probability.

4. the processing method of application program as claimed in claim 3, which is characterized in that the use information is sample application journey The running state information of sequence；The step of whether use information meets preset condition judged, comprising:

Judge whether the operating status is front stage operation；

If so, determining that the use information meets preset condition；

If it is not, then determining that the use information is unsatisfactory for preset condition.

5. the processing method of application program as claimed in claim 2, which is characterized in that the sampling periods include [t₁,t₂… t_m], the sample includes [P using probability₁,P₂…P_m]；

The step of preset mixed Gauss model is trained according to the training sample, comprising:

The sampling periods and corresponding sample are input in the first preset formula using probability, first preset formula Are as follows:

Wherein, A_iIndicate that sample application program i, t indicate that sampling periods, k indicate sub- Gauss model quantity, μ_kIndicate mathematic expectaion, σ_kIndicate variance, ω_kExpression weight, and N (t | μ_k,σ_k) indicate that stochastic variable t obeys a mathematic expectaion as μ_k, variance σ_kJust State distribution, and P (t | A_i) indicate the probability that sampling periods are t when the operating status of sample application program i is front stage operation；

Probability is used based on sampling periods, the sample inputted, first preset formula is trained, multiple training are obtained Sub- Gauss model afterwards；

By the sub- Gauss model superposition after multiple training, with the mixed Gauss model after being trained.

6. such as the processing method of application program described in any one of claim 1 to 5, which is characterized in that according to described using general The step of rate handles the background application in the electronic equipment, comprising:

The target background application for being less than preset threshold using probability is determined from current background application program；

Close the target background application.

7. a kind of processing unit of application program, it is applied to electronic equipment, which is characterized in that described device includes:

Module is obtained, for obtaining the use information of each sampling time point sample application program in historical time section；

Generation module, for generating training sample according to the sampling time point and the use information；

Training module, for being trained according to the training sample to preset mixed Gauss model, wherein each to apply journey Ordered pair should have the mixed Gauss model after unique training；

Processing module, including acquisition submodule, the second determining submodule, computational submodule and application processing submodule；

The acquisition submodule, for obtaining application program process instruction；

Described second determines submodule, for determining that the backstage in the electronic equipment is answered according to the application program process instruction Use program；

The computational submodule, for based on the mixed Gauss model after training corresponding to each application program, utilization to be preset Bayesian model calculates each background application in the probability that uses of object time, and the expression formula of the Bayesian model is the Two preset formulas, second preset formula are as follows:

Wherein, T indicates the time, and N indicates the quantity of the mixed Gauss model after training, P (A_i| T) indicate sampling periods be T when before The probability that the application program of platform operation is application program i, and P (T | A_i) indicate the operating status of sample application program i for foreground fortune When row sampling periods be T probability, P (T | A_j) indicate that sampling periods are T when the operating status of application program j is front stage operation Probability, P (A_i) indicate application program i in historical time section using probability, P (A_j) indicate going through for application program j In the history period using probability；

The application handles submodule, for being carried out using probability to the background application in the electronic equipment according to described Processing.

8. the processing unit of program the use as claimed in claim 7, which is characterized in that when the historical time section includes multiple Between the period, every a period of time is divided into multiple sampling periods；

The generation module includes:

First determines submodule, for determining each sampling time point corresponding time cycle and sampling periods, wherein per a period of time Between in the period sampling time point and sampling periods correspond；

Information processing submodule is obtained for handling the corresponding use information of sampling periods identical in the different time period Probability is used in the corresponding sample of each sampling periods to sample application program；

Submodule is generated, for generating training sample using probability based on sampling periods and corresponding sample.

9. the processing unit of application program as claimed in claim 8, which is characterized in that the processing submodule includes:

Judging unit, for judging whether the use information meets preset condition；

First determination unit, for determining that each sample applies the corresponding use information in identical sampling periods to meet default item The sampling time point quantity of part；

Acquiring unit, when applying that use information meets the sampling of preset condition within multiple time cycles for obtaining each sample Between put total quantity；

Computing unit, for calculating each sample and answering according to the sampling time point quantity and the sampling time point total quantity Probability is used in the corresponding sample of each sampling periods with program.

10. the processing unit of application program as claimed in claim 9, which is characterized in that the use information is sample application The running state information of program；The judging unit is used for:

Judge whether the operating status is front stage operation；

If so, determining that the use information meets preset condition；

If it is not, then determining that the use information is unsatisfactory for preset condition.

11. the processing unit of application program as claimed in claim 10, which is characterized in that the sampling periods include [t₁, t₂…t_m], the sample includes [P using probability₁,P₂…P_m]；The training module includes:

Input submodule, for the sampling periods and corresponding sample to be input in the first preset formula using probability, institute State the first preset formula are as follows:

Wherein, A_iIndicate that sample application program i, t indicate that sampling periods, k indicate sub- Gauss model quantity, μ_kIndicate mathematic expectaion, σ_kIndicate variance, ω_kExpression weight, and N (t | μ_k,σ_k) indicate that stochastic variable t obeys a mathematic expectaion as μ_k, variance σ_kJust State distribution, and P (t | A_i) indicate the probability that sampling periods are t when the operating status of sample application program i is front stage operation；

Training submodule instructs first preset formula for using probability based on the sampling periods, the sample that are inputted Practice, the sub- Gauss model after obtaining multiple training；

It is superimposed submodule, for the sub- Gauss model after multiple training to be superimposed, with the mixed Gauss model after being trained.

12. the processing unit of application program as claimed in claim 11, which is characterized in that the application handles submodule packet It includes:

Second determination unit, for determining that the target backstage for being less than preset threshold using probability is answered from current background application program Use program；

Closing unit, for closing the target background application.

13. a kind of storage medium, which is characterized in that be stored with a plurality of instruction in the storage medium, described instruction be suitable for by Device load is managed to execute the processing method such as application program of any of claims 1-6.

14. a kind of electronic equipment, which is characterized in that including processor and memory, the processor and the memory are electrical Connection, the memory is for storing instruction and data；The processor is for executing as described in any one of claim 1-6 Application program processing method.