WO2019128598A1 - Application processing method, electronic device, and computer readable storage medium - Google Patents

Abstract

An application processing method, comprising: acquiring first feature data of each feature, the first feature data being feature data of a corresponding feature during a moment of prediction; acquiring a decision tree model for use in predicting whether a user will use a target application during a preset time period; using the first feature data as an input for the decision tree model, and outputting the prediction result; when the prediction result is that the user will not use the target application during the preset time period, freezing the target application.

Description

Application processing method, electronic device, computer readable storage medium

Cross-reference to related applications

This application claims the priority of the Chinese Patent Application filed on Dec. 29, 2017, the Chinese Patent Application No. 2017114844409, entitled "Application Processing Method and Apparatus, Electronic Apparatus, Computer Readable Storage Medium", the entire contents of which are hereby incorporated by reference. This is incorporated herein by reference.

Technical field

The present application relates to the field of data processing, and in particular, to an application processing method, an electronic device, and a computer readable storage medium.

Background technique

With the development of mobile communication technologies, mobile operating systems provide a method for resource limitation of background applications. In the traditional operating system, when the system resource usage rate is too high, part of the background application is forced to exit processing to recover some resources for the foreground application.

The traditional method is to force the exit of the corresponding background application. The traditional method is to select the length of the background in the background, the frequency of use, the duration of the application, etc., and select the longer duration of the background station and the application frequency. A background application that is low or has a short duration, forcibly exits the selected background application. For the traditional method, there will still be a corresponding background application that has just been checked and killed. In a short time, the user has to use the application again, so the system needs to load the recovered resources again, resulting in low accuracy of recycling.

Summary of the invention

An application processing method, an electronic device, and a computer readable storage medium are provided according to various embodiments of the present application.

An application processing method includes: acquiring first feature data of each feature, wherein the first feature data is feature data of a corresponding feature at a predicted time; and acquiring is used to predict whether the user will use the preset time length a decision tree model of the target application, the start time of the preset duration is a predicted time; the first feature data is used as an input of the decision tree model, and the predicted result is output; and when the predicted result is not The target application is frozen when the target application is used within a preset duration.

An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following operations:

Acquiring first feature data of each feature, the first feature data is feature data of the corresponding feature at the predicted time; acquiring a decision tree model for predicting whether the user will use the target application within a preset duration, The starting time of the preset duration is a predicted time; the first feature data is used as an input of the decision tree model, and the predicted result is output; and when the predicted result is not used within a preset duration When the target application is described, the target application is frozen.

A computer readable storage medium having stored thereon a computer program that is executed by a processor to:

Acquiring first feature data of each feature, the first feature data is feature data of the corresponding feature at the predicted time; acquiring a decision tree model for predicting whether the user will use the target application within a preset duration, The starting time of the preset duration is a predicted time; the first feature data is used as an input of the decision tree model, and the predicted result is output; and when the predicted result is not used within a preset duration When the target application is described, the target application is frozen.

In the embodiment of the present application, the decision tree model of the target application is preset, and the first feature data of each feature and the decision tree model of the target application are obtained, and the first feature data is used as an input of the decision tree model to obtain Whether the user output by the decision tree model uses the prediction result of the target application within a preset duration, and when the prediction result is that the target application is not used within the preset duration, the target application is frozen to limit the target. The application's occupation of resources improves the accuracy of freezing the target application, which in turn increases the effectiveness of system resource release.

Details of one or more embodiments of the present application are set forth in the accompanying drawings and description below. Other features, objects, and advantages of the invention will be apparent from the description and appended claims.

DRAWINGS

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings to be used in the embodiments or the prior art description will be briefly described below. Obviously, the drawings in the following description are only It is a certain embodiment of the present application, and other drawings can be obtained according to the drawings without any creative work for those skilled in the art.

FIG. 1 is a schematic diagram showing the internal structure of an electronic device in an embodiment.

2 is a partial block diagram of a system in an electronic device in an embodiment.

FIG. 3 is an application environment diagram of an application processing method in an embodiment.

4 is a flow chart of an application processing method in one embodiment.

FIG. 5 is a flow chart of sample classification of a sample set according to an information gain rate classified by a feature according to a feature in one embodiment, and a decision tree model for predicting whether a user will use a target application within a preset duration.

Figure 6A is a schematic diagram of a decision tree in one embodiment.

Figure 6B is a schematic diagram of a decision tree in another embodiment.

Figure 6C is a schematic diagram of a decision tree in yet another embodiment.

7 is a flow diagram of obtaining an information gain rate for a feature classification for a target sample set in one embodiment.

Figure 8 is a flow chart of an application processing method in another embodiment.

Figure 9 is a block diagram showing the structure of an application processing device in an embodiment.

Figure 10 is a block diagram showing the structure of an application processing device in another embodiment.

Figure 11 is a block diagram showing the structure of an application processing device in still another embodiment.

Figure 12 is a block diagram showing a portion of the structure of a handset in an embodiment.

Detailed ways

In order to make the objects, technical solutions, and advantages of the present application more comprehensible, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the application and are not intended to be limiting.

It will be understood that the terms "first", "second" and the like, as used herein, may be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish one element from another. For example, the first feature data may be referred to as second feature data without departing from the scope of the present invention, and similarly, the second feature data may be referred to as first feature data. Both the first feature data and the second feature data are feature data, but they are not the same feature data.

In one embodiment, as shown in FIG. 1, an internal structure diagram of an electronic device is provided. The electronic device includes a processor, memory, and display screen connected by a system bus. The processor is used to provide computing and control capabilities to support the operation of the entire electronic device. The memory is used to store data, programs, and/or instruction codes, etc., and the memory stores at least one computer program, which can be executed by the processor to implement an application processing method suitable for an electronic device provided in the embodiments of the present application. The memory may include a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random storage memory (Random-Access-Memory, RAM). For example, in one embodiment, the memory includes a non-volatile storage medium and an internal memory. Non-volatile storage media stores operating systems, databases, and computer programs. The database stores data related to an application processing method provided by the above various embodiments, for example, information such as the name of each process or application may be stored. The computer program can be executed by a processor for implementing an application processing method provided by various embodiments of the present application. The internal memory provides a cached operating environment for operating systems, databases, and computer programs in non-volatile storage media. The display screen can be a touch screen, such as a capacitive screen or an electronic screen, for displaying interface information of an application corresponding to the first process, and can also be used for detecting a touch operation applied to the display screen, and generating corresponding instructions, such as before execution. Switching instructions for background applications, etc.

A person skilled in the art can understand that the structure shown in FIG. 1 is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the electronic device to which the solution of the present application is applied. The specific electronic device may be It includes more or fewer components than those shown in the figures, or some components are combined, or have different component arrangements. For example, the electronic device further includes a network interface connected through a system bus, and the network interface may be an Ethernet card or a wireless network card, etc., for communicating with an external electronic device, for example, for communicating with a server. For example, there is no display connected through the system bus on the electronic device, or an external display device can be connected.

In one embodiment, as shown in FIG. 2, a partial architectural diagram of an electronic device is provided. The architecture of the electronic device includes a JAVA spatial layer 210, a local framework layer 220, and a kernel space layer 230. The JAVA spatial layer 210 can include a freeze management application 212, by which the electronic device can implement a freeze policy for each application, and perform freezing and thawing management operations on the related applications of the background power consumption. The resource priority and restriction management module 222 and the platform freeze management module 224 are included in the local framework layer 220. The electronic device can maintain different applications in different priorities and different resource organizations through the resource priority and limit management module 222, and adjust the resource group of the application according to the requirements of the upper layer to achieve optimized performance and save power. effect. The electronic device can allocate the tasks that can be frozen in the background to the frozen layer corresponding to the preset different levels according to the length of the entering freeze time through the platform freeze management module 224. Optionally, the frozen layer can include three, respectively: the CPU Limit sleep mode, CPU freeze sleep mode, process deep freeze mode. The CPU restricts the sleep mode to limit the CPU resources occupied by the related processes, so that the related processes occupy less CPU resources, and the free CPU resources are tilted to other unfrozen processes, thereby limiting the occupation of CPU resources. It also limits the occupation of network resources and I/O interface resources by the process; CPU freeze sleep mode refers to prohibiting related processes from using the CPU, while preserving the occupation of memory, when prohibiting the use of CPU resources, the corresponding network resources and I The /O interface resource is also forbidden; the process deep freeze mode means that in addition to prohibiting the use of CPU resources, the memory resources occupied by the related processes are further recovered, and the recovered memory can be used by other processes. The kernel space layer 230 includes a UID management module 231, a Cgroup module 232, a Binder management module 233, a process memory recovery module 234, and a freeze timeout exit module 235. The UID management module 231 is configured to implement an application-based User Identifier (UID) to manage resources of a third-party application or freeze. Compared with the Process Identifier (PID) for process management and control, it is easier to uniformly manage the resources of a user's application through UID. The Cgroup module 232 is used to provide a complete set of Central Processing Unit (CPU), CPUSET, memory, input/output (I/O), and Net related resource restriction mechanisms. The Binder management module 233 is used to implement the priority control of the background binder communication. The interface module of the local framework layer 220 includes a binder interface developed to the upper layer, and the upper layer framework or application sends a resource restriction or frozen instruction to the resource priority and restriction management module 222 and the platform freeze management module 224 through the provided binder interface. . The process memory recovery module 234 is configured to implement the process deep freeze mode, so that when a third-party application is in a frozen state for a long time, the file area of the process is mainly released, thereby saving the memory module and speeding up the application next time. The speed at startup. The freeze timeout exit module 235 is configured to resolve an exception generated by the freeze timeout scenario. Through the above architecture, the application processing method in various embodiments of the present application can be implemented.

In an embodiment, as shown in FIG. 3, which is a schematic diagram of an application scenario in which an application processing method is used in an embodiment, an electronic device may collect feature data of a preset feature for embodying a user behavior habit as a sample under different time periods. Forming a sample set, classifying the sample set according to the information gain rate of the feature classification for the sample, to construct a decision tree model for predicting whether the target application will be used within a preset duration; and then predicting the preset feature at the moment The feature data is used as an input to the decision tree model to obtain a prediction result. When the prediction result is that the target application is not used within the preset duration when the prediction result is that the target application is not frozen, the target application is not frozen.

In an embodiment, as shown in FIG. 4, an application processing method is provided. The embodiment is applied to the electronic device shown in FIG. 1 as an example for description. The method includes:

Operation 402: Acquire first feature data of each feature, where the first feature data is feature data of the corresponding feature at the predicted time.

The feature data is data for characterizing the operational habits of the user's application to the electronic device. The feature may be one or more dimensions of the device hardware and software features of the electronic device, the device usage state, and the user operation time duration. The electronic device can bury the point in the preset path of the system, and detect the feature data of each feature in real time according to the preset sampling frequency. The electronic device may also record the feature data of the relevant feature and the application identifier of the application at the startup time when the application is activated, and associate the application identifier with the recorded feature data.

Optionally, the feature includes a current time period when the corresponding data is recorded, a current date category, an application identifier of the previous foreground application, an application identifier of the previous foreground application, a current wireless fidelity WiFi (WIreless Fidelity) connection status, The WiFi ID of the connected WiFi, the duration of the application staying in the background, the duration of the application during the background pause, the current plug-in status of the headset, the current charging status, the current battery level, the way the application is switched, the category of the application, and the application One or more of the characteristics such as the number of times to switch to the foreground. The date category may include a working day and a rest day, and the WiFi connection status includes the unconnected WiFi and the connected WiFi, and the WiFi identifier may be a Service Set Identifier (SSID) or a Basic Service Set Identifier (BSSID). Any of the types of information that can uniquely identify the WiFi; the category of the application includes a social application, a payment application, a game application, a tool application, and the like, and the classification of the application may include multiple types, and is not limited thereto. The application switching means that the application switches from the foreground to the background or from the background to the foreground. The switching manner may include an application switching formed by directly opening an application according to an application startup icon, an application switching formed by clicking an application notification message, and directly exiting the application. Application switching, etc. The previous foreground application indicates the application that was running in the foreground at the moment when the feature was recorded. The previous application indicates the application that was last running in the foreground at the time of recording the feature.

The first feature data refers to the feature data of the corresponding feature at the predicted time. When the feature relates to the application, the target application, for example, the first feature data may include the current time period and the current date category at the predicted time. , the previous foreground application name, the previous foreground application name, the current WiFi connection status, the WiFi identity of the connected WiFi, the duration of the target application staying in the background, the duration of the target application during the background pause, and the current headset plugging and unplugging One or more characteristics of the status, the current state of charge, the current battery level, the manner in which the target application is switched, the category of the target application, the number of times the target application switches to the foreground, and the like. The predicted time may be the current time.

In one embodiment, the electronic device may initiate a freeze instruction for one or more applications running in the background by freezing the management application 212, and the freeze instruction may begin to acquire the first feature data of each feature according to the instruction. . The electronic device may trigger the application freeze instruction when detecting that the available resource is lower than a preset threshold, or may initiate a freeze instruction to the application when detecting that an application is switched to the background. The application for which the freeze instruction is directed is the target application.

Operation 404, obtaining a decision tree model for predicting whether the user will use the target application within a preset duration; the starting time of the preset duration is the predicted time.

The decision tree model is a model for predicting whether the target application will be used within the preset duration. The starting time of the preset duration is the predicted time, that is, whether the predicted target application will be used within the preset duration from the current time. The preset duration can be any suitable length of time, and can be set according to an empirical value, such as 1 minute, 5 minutes, 10 minutes, or 30 minutes. The target application is the application of the benefit measurement, wherein the target application may be one or more applications in the electronic device that are currently running in the background.

In one embodiment, the electronic device presets a decision tree model corresponding to the target application, and obtains a decision tree model corresponding to the target application to predict whether the target application will be used by the user within a preset duration. For different applications, different decision tree models can be set correspondingly. For example, a corresponding decision tree model can be set for each application, or a corresponding decision tree model can be set for each type of application.

Operation 406, the first feature data is used as an input of the decision tree model, and the prediction result is output.

The electronic device may use the collected first feature data of the respective features at the predicted time as the data of the decision tree model, and run the decision tree model to obtain an output result of the decision tree. The result of this data is the predicted result. Among them, the prediction result includes that the target application will not be used within the preset duration or will be used within the preset duration.

Operation 408, when the predicted result is that the target application is not used within the preset duration, the target application is frozen.

When the predicted result is that it is not used within the preset duration, the electronic device can freeze the target application. Optionally, the electronic device may send the prediction result to the platform freeze management module 224 as shown in FIG. 2, and when the platform freeze management module 224 receives the prediction result that does not use the target application within the preset duration, the The target application initiates a freeze operation, thereby limiting the resources that the target application can use, such as adopting any one of a CPU-limited sleep mode, a CPU freeze sleep mode, or a process deep freeze mode to freeze the target application.

The application processing method described above, by setting a decision tree model of the target application in advance, and acquiring the first feature data of each feature and the decision tree model of the target application, using the first feature data as an input of the decision tree model, Whether the user output by the decision tree model uses the prediction result of the target application within a preset duration, and when the prediction result is that the target application is not used within the preset duration, the target application is frozen to limit The target application's occupation of resources. Since it is necessary to consume a certain amount of system resources when freezing the application, whether the target application is used within the preset duration by performing the prediction, and freezing when not in use, thereby reducing the user to freeze after the target application is frozen. The application is used within the preset duration, and the application needs to be thawed, which causes unnecessary freezing operations, improves the accuracy of freezing the target application, and thus improves the effectiveness of releasing system resources.

In an embodiment, before the operation 402, the method further includes: acquiring preset second feature data of each feature as a sample, and generating a sample set, where the second feature data is a correspondence when the reference application is started before the predicted time Feature data of the feature, the reference application includes a target application; when the data volume of the sample set exceeds a preset threshold, the sample set is sample-classified according to the information gain rate of the feature classification for the sample set, and is generated for predicting whether the user is preset The decision tree model of the target application is used within the duration.

The second feature data is feature data of each feature recorded before the predicted time, and the second feature data has an association relationship with the application identifier of the corresponding reference application, that is, before the predicted time, the reference application is detected At startup, the feature data of each feature (ie, the second feature data) is recorded at the startup time, and the second feature data is associated with the application identifier of the reference application.

The electronic device may collect the second feature data of each feature according to a preset frequency through a historical time period, and use the second feature recorded each time as a sample to form a sample set. When the amount of data of the accumulated sample set exceeds a preset threshold, then a decision tree model of each application requiring prediction is constructed based on the accumulated sample set. The historical time period may be, for example, the past 7 days or 10 days; the preset frequency may be, for example, collected every 10 minutes and collected every half hour. It can be understood that the multi-dimensional feature data collected at one time constitutes one sample, and multiple samples collected multiple times constitute a sample set.

After constituting the sample set, each sample in the sample set can be marked, the application identification of the sample is marked, and the sample label of each sample is obtained. The tag for each sample is the application ID of the corresponding association. Since the target application is targeted, the sample tag can be classified into a target application and a non-target application, that is, the sample category includes a "target application" and a "non-target application." When the target application is started, the sample tag of the second feature data of the recorded feature is set as the "target application", and the sample tag of the second feature data of the recorded feature is activated when the non-target application is started. Set to "non-target app". Alternatively, the value "1" may be used to indicate "target application", and the value "0" may be used to indicate "non-target application", and vice versa.

The preset threshold may be any suitable value preset, for example, may be 10000, that is, when the number of samples recorded exceeds 10000, the decision tree model is started. Alternatively, the larger the number of samples, the more accurate the decision tree model is constructed.

In an embodiment, to facilitate sample classification, the electronic device may quantize the feature information in the feature data that is not directly represented by the value by a specific value. For example, for the feature of the current WiFi connection state, the value 1 may be used to indicate that Connect WiFi, use the value 0 to indicate that WiFi is not connected (or vice versa).

The embodiment of the present application may perform sample classification on the sample set based on the information gain rate of the feature classification for the sample set to construct a decision tree model for predicting whether the user will use the target application within a preset duration. For example, a decision tree model can be constructed based on the C4.5 algorithm.

Among them, the decision tree is a tree built on the basis of decision-making. In machine learning, a decision tree is a predictive model that represents a mapping between object attributes and object values. Each node represents an object, and each forked path in the tree represents a certain Possible attribute values, and each leaf node corresponds to the value of the object represented by the path from the root node to the leaf node. The decision tree has only a single output. If there are multiple outputs, separate decision trees can be created to handle different outputs.

Among them, the C4.5 algorithm is a kind of decision tree. It is a series of algorithms used in the classification problem of machine learning and data mining. It is an important algorithm improved by ID3. Its goal is to supervise learning: Given a data set, each of these tuples can be described by a set of attribute values, each of which belongs to a class in a mutually exclusive category. The goal of C4.5 is to find a mapping from attribute values to categories by learning, and this mapping can be used to classify entities with unknown new categories.

ID3 (Iterative Dichotomiser 3, iterative binary tree 3 generation) is based on the Occam razor principle, that is, to do more with as few things as possible. In information theory, the smaller the expected information, the greater the information gain and the higher the purity. The core idea of the ID3 algorithm is to measure the choice of attributes with information gain, and select the attribute with the largest information gain after splitting to split. The algorithm uses a top-down greedy search to traverse possible decision spaces.

In the embodiment of the present application, the information gain rate may be defined as the ratio of the information gain of the feature to the sample classification and the split information of the feature to the sample classification. The specific information gain rate acquisition method is described below.

The information gain is for one feature. It is to look at a feature t. What is the amount of information when the system has it and without it? The difference between the two is the amount of information that the feature brings to the system, that is, the information gain. .

The split information is used to measure the breadth and uniformity of the feature split data (such as the sample set), and the split information can be the entropy of the feature.

In one embodiment, as shown in FIG. 5, the sample set is sample-classified according to the information gain rate of the feature classification for the sample set, and a decision tree model for predicting whether the user will use the target application within a preset duration is generated. include:

Operation 501, the sample set is used as the node information of the root node of the decision tree; and the node information of the root node is determined as the target sample set to be classified currently.

That is, in the case of starting construction, the sample set is taken as the target sample set to be classified currently. The classification of sample sets includes "target applications" and "non-target applications."

Operation 502, obtaining an information gain rate of the feature classification for the target sample set.

Operation 503, selecting a sample from the target sample set as the division feature according to the information gain rate.

The partitioning feature is a feature selected from the features according to the information gain rate of each feature for the sample set classification, and is used to classify the sample set. Among them, there are various ways to select the feature according to the information gain rate. For example, in order to improve the accuracy of the sample classification, the feature corresponding to the maximum information gain rate may be selected as the division feature.

In an embodiment, in order to improve the decision accuracy of the decision tree model, a gain rate threshold may also be set; when the maximum information gain rate is greater than the threshold, the electronic device selects a feature corresponding to the information gain rate as a division feature. The maximum target information gain rate may be selected from the information gain rate; whether the target information gain rate is greater than a preset threshold; and when the target information gain rate is greater than the preset threshold, the feature corresponding to the target information gain rate is selected as the current division feature. .

When the target information gain rate is not greater than the preset threshold, the current node may be used as a leaf node, and the sample with the largest number of samples may be selected as the output of the leaf node. The sample category includes “target application” and “non-target application”. When the output is “target application”, it means that the target application will be used within the preset duration. When the output is “non-target application”, This means that the target application will not be used within the preset duration. The preset threshold can be set according to actual needs, such as 0.9, 0.8, and the like. For example, when the information gain rate of feature 1 for the sample classification is 0.9, the preset gain rate threshold is 0.8, since the maximum information gain rate is greater than the preset threshold, feature 1 can be used as the division feature. For another example, when the preset threshold is 1, then the maximum information gain rate is less than the preset threshold. At this time, the current node may be used as a leaf node.

Operation 504, dividing the target sample set according to the dividing feature, and generating at least one sub-sample set.

The electronic device classifies the samples according to the division features in various ways. For example, the sample set may be divided based on the feature values of the divided features. The electronic device may also acquire feature values of the feature set in the target sample set; and divide the target sample set according to the feature value. For example, an electronic device may divide a sample with the same feature value in a sample set into the same subsample set. For example, the feature values of the divided features include: 0, 1, 2, then, at this time, the samples whose feature values are 0 can be classified into one class, and the samples with the feature value 1 are classified into one class, and the feature values are The samples of 2 are classified into one category.

Operation 505, removing the dividing feature of the sample in each sub-sample set; generating a child node of the current node, and removing the sub-sample set after dividing the feature as the node information of the child node.

Operation 506, determining whether the child node satisfies the preset classification termination condition; when satisfied, performing operation 507, when not satisfied, updating the target sample set to the subsample set after removing the division feature, and returning to performing operation 502.

In operation 507, the child node is used as a leaf node, and the output of the leaf node is set according to the category of the subsample set after the division feature is removed.

When the child node satisfies the preset classification termination condition, the electronic device may use the child node as a leaf node, that is, stop the sample set classification of the child node, and set the output of the leaf node based on the category of the sample in the removed sub-sample set. There are several ways to set the output of a leaf node based on the category of the sample. For example, the category with the largest number of samples in the sample set can be removed as the output of the leaf node.

The preset classification termination condition may be set according to actual requirements. When the child node satisfies the preset classification termination condition, the electronic device uses the current child node as a leaf node, and stops classifying the sample set corresponding to the child node; When the preset classification termination condition is not met, the electronic device may continue to classify the sample set corresponding to the child node. For example, the preset classification termination condition may include: whether the number of categories of the samples in the removed sub-sample set of the child node is a preset number, and when yes, determining that the child node meets the preset classification termination condition.

Through the above method, a decision tree model of the target application can be constructed, so that it is predicted according to the decision tree model whether the target application will be used within a preset duration.

For example, for sample set D{sample 1, sample 2...sample i...sample n}, where the sample includes several features A.

First, all samples in the sample set are initialized, then a root node d is generated, and the sample set D is taken as the node information of the root node d, as described with reference to FIG. 6A.

Calculate the information gain rate g _R (D, A)1, g _R (D, A) 2, ... g _R (D, A) m of each feature such as feature A for the sample set; select the maximum information gain rate g _R (D, A) max.

When the maximum information gain rate g _R (D, A)max is less than the preset threshold ε, the current node acts as a leaf node, and the sample category with the largest number of samples is selected as the output of the leaf node.

When the maximum information gain rate g _R (D, A)max is greater than the preset threshold ε, the feature corresponding to the information gain g _R (D, A) max may be selected as the partitioning feature A _g , and the sample set D according to the feature A _g {sample 1, sample 2 sample I ...... ...... sample n} be divided, in particular, for each value of a _g a _i, in accordance with the a _g = a _i to D is divided into several non-empty set D _I, As a child of the current node. For example, the sample set is divided into two subsample sets D ₁ {sample 1, sample 2 ... sample k} and D ₂ {sample k+1 ... sample n}.

The division feature A _g in the subsample sets D1 and D2 is removed, that is, AA _g . The child nodes d ₁ and d _{2 of the} root node d are generated with reference to FIG. 6A, and the sub-sample set D _{1 is taken} as the node information of the child node d ₁ and the sub-sample set D2 is taken as the node information of the child node d2.

Next, for each child node, for each child node, AA _{g is taken} as the feature, and the D _{i of the} child node is used as the data set, and the above steps are recursively called to construct the subtree until the preset classification termination condition is satisfied.

Taking the child node d ₁ as an example, determining whether the child node satisfies the preset classification termination condition, and if so, using the current child node d ₁ as a leaf node, and setting the leaf according to the category of the sample in the subsample set corresponding to the child node d ₁ Node output.

When the child node does not satisfy the preset classification termination condition, the above-mentioned sub-sample set corresponding to the child node is continued to be classified according to the information gain classification method. For example, the child node d ₂ can be used as an example to calculate the relative features of the A ₂ sample set. For the information gain rate g _R (D, A) of the sample classification, the maximum information gain rate g _R (D, A) max is selected, when the maximum information gain rate g _R (D, A) max is greater than the preset threshold ε The feature corresponding to the information gain rate g _R (D, A) may be selected as a division feature A _g (such as feature Ai+1), and D _{2 is} divided into several sub-sample sets based on the division feature A _g , such as D ₂ Divided into subsample sets D ₂₁ , D ₂₂ , D ₂₃ , and then the divided features A _g in the subsample sets D ₂₁ , D ₂₂ , D ₂₃ are removed, and the child nodes d ₂₁ , d _{22 of the} current node d ₂ are generated, d ₂₃ , the sample sets D ₂₁ , D ₂₂ , and D ₂₃ after the division feature A _g are removed as the node information of the child nodes d ₂₁ , d ₂₂ , and d ₂₃ , respectively.

By analogy, the above-mentioned information gain rate classification based manner can be used to construct a decision tree as shown in FIG. 6B, and the output of the leaf node of the decision tree includes using the target application within a preset duration or at a preset duration. The target application will not be used within. Among them, the output results "yes" and "no" can correspond to "target application" and "non-target application", which means "the target application will be used within the preset duration" and "will not be used within the preset duration" Target application."

In one embodiment, in order to improve the speed and efficiency of prediction using the decision tree, it is also possible to mark the feature values of the corresponding divided features on the path between the nodes. For example, in the above process based on information gain classification, the feature values of the corresponding divided features may be marked on the path of the current node and its child nodes.

For example, the feature values of the partition feature A _g include: 0, 1 may mark 1 on the path between d ₂ and d, mark 0 on the path between d ₁ and d, and so on, at each division Then, the corresponding division feature value such as 0 or 1 can be marked on the path of the current node and its child nodes, and the decision tree as shown in FIG. 6C can be obtained. Similarly, the output results "yes" and "no" can correspond to "target application" and "non-target application", which means "the target application will be used within the preset duration" and "will not be within the preset duration" Use the target app."

In one embodiment, as shown in FIG. 7, the information gain rate of the feature acquired for the target sample set classification includes:

Operation 702, obtaining an information gain of the feature classification for the target sample set.

The information gain represents the degree of uncertainty in the information of a class of a feature ("target application" or "non-target application").

Operation 704, acquiring split information of the feature classification for the target sample set.

The split information is used to measure the breadth and uniformity of the feature split data (such as the sample set), and the split information can be the entropy of the feature.

Operation 706, acquiring an information gain rate of the feature classification for the target sample set according to the information gain and the split information.

The information gain rate can be the ratio of the information gain of the feature to the sample set classification and the split information of the feature to the sample classification. The information gain rate can be obtained by dividing the obtained information gain by the corresponding split information.

The information gain of the feature classification for the target sample set is also the difference between empirical entropy and conditional entropy. The electronic device can obtain the empirical entropy of the target sample classification; acquire the conditional entropy of the feature for the classification result of the target sample set; and obtain the information gain of the feature for the target sample set classification according to the conditional entropy and the empirical entropy. Wherein, the electronic device can obtain the first probability that the positive sample appears in the sample set and the second probability that the negative sample appears in the sample set, the positive sample is the sample whose sample category is “target application”, and the negative sample is the sample category is “non- A sample of the target application; the empirical entropy of the sample is obtained from the first probability and the second probability.

In one embodiment, for example, for a sample set D{sample 1, sample 2...sample i...sample n}, the sample includes a multi-dimensional feature, such as feature A. The information gain rate of feature A for sample classification can be obtained by the following formula:

Where g _R (D, A) is the information gain rate of feature A for sample set D classification, g(D, A) is the information gain of feature A for sample classification, and H _A (D) is the split information of feature A, That is, the entropy of feature A.

Where g _R (D, A) can be obtained by the following formula:

H(D) is the empirical entropy of the sample set D classification, and H(D|A) is the conditional entropy of the feature A for the sample set D classification.

If the sample size of the sample category is "target application" is j, the number of samples of "non-target application" is nj; at this time, the probability of occurrence of the positive sample in sample set D is p ₁ = j / n, and the sample of negative sample is in the sample set The probability of occurrence in D is p ₂ =nj/n. Then, based on the following empirical entropy calculation formula, the empirical entropy H(D) of the sample classification is calculated:

In the decision tree classification problem, the information gain is the difference between the information of the decision tree before and after the attribute selection. In this implementation, the empirical entropy H(D) of the sample classification is:

H(D)=p ₁ log p ₁ +p ₂ log p ₂

In an embodiment, the electronic device may divide the sample set into several sub-sample sets according to the feature A, and then obtain the information entropy of each sub-sample set classification, and the probability that each feature value of the feature A appears in the sample set, According to the information entropy and the probability, the divided information entropy, that is, the conditional entropy of the feature A _i for the sample set classification result can be obtained.

For example, for sample feature A, the conditional entropy of the sample feature A for the sample set D classification result can be calculated by the following formula:

Where n is the number of values of the feature A, that is, the number of feature value types. At this time, p _i is the probability that the sample whose A eigenvalue is the ith value appears in the sample set D, and A _i is the ith value of A. (D|A=A _i ) is the empirical entropy of the subsample set D _i classification, and the A eigenvalues of the samples in the subsample set D _i are all the ith values.

For example, taking the value of the feature A as 3, that is, A ₁ , A ₂ , and A ₃ as an example, at this time, the feature set A can be a sample set D {sample 1, sample 2, ... sample i ... sample n} is divided into three sub-sample set, wherein the value of {D a _{1 1} sample 1 sample 2 ...... sample d}, wherein a value of ₂ a D ₂ {d + 1 ...... sample sample e}, wherein A3 is D3 {sample e+1...sample n}. d, e are positive integers and less than n.

At this time, the conditional entropy of feature A for the classification result of sample set D is:

H(D|A)=p ₁ H(D|A=A ₁ )+p ₂ H(D|A=A ₂ )+p ₃ H(D|A=A ₃ )

Wherein p ₁ = D ₁ /D, p ₂ = D2 / D, p ₂ = D ₃ / D;

H(D|A ₁ ) is the information entropy of the subsample set D ₁ classification, that is, the empirical entropy, which can be calculated by the above formula of empirical entropy.

After obtaining the empirical entropy H(D) of the sample classification and the conditional entropy H(D|A) of the feature A for the sample set D classification result, the information gain of the feature A for the sample set D classification can be calculated, for example, by The formula is calculated:

That is, the information gain of the feature A for the sample set D classification is: the difference between the empirical entropy H(D) and the conditional entropy H(D|A) of the feature A for the sample set D classification result.

The split information of the feature classification for the sample set is the entropy of the feature. The probability of the distribution of the features can be obtained based on the probability of distribution of the samples in the target sample set. For example, H _A (D) can be obtained by the following formula:

Is the value category of feature A, or the number of species.

Where D _i is a sample set in which the feature set D of the sample set D is the i-th kind.

In one embodiment, before freezing the target application, the method further includes: detecting whether the target application belongs to the whitelist application, and if so, freezing the target application; otherwise, performing freezing on the target application.

Wherein, detecting whether the target application belongs to the whitelist application may be performed in any process before the target application is frozen. The whitelist of the freeze-free application is preset in the electronic device, and the application in the whitelist can be customized for the user, or can be set as the system default. The whitelist records application information of an application that can be freed from freezing, such as an application identifier of a recordable application. When the target application is in the freeze-free list, the electronic device does not freeze the target application, and when the target application is not in the white list, and the predicted result is that the target application is not used within the preset duration, the electronic device applies to the target application. Freeze. By further setting up the whitelist, the accuracy of the application freeze is further improved.

In one embodiment, as shown in FIG. 8, another application processing method is provided, the method comprising:

Operation 801: Acquire second preset feature data of each feature as a sample, and generate a sample set.

The acquired features may include a plurality of features, as shown in Table 1 below, for the 14 dimensions acquired by the electronic device. In practice, the number of feature information included in one sample may be more than the number of information shown in Table 1, or may be less than the number of information shown in Table 1, and the specific feature information may also be as shown in Table 1. Different, it is not specifically limited herein.

Dimension	Characteristic information
1	Current time period
2	Current date category
3	a foreground app name
4	Previous foreground application name
5	How long the app stays in the background
6	The length of time the app is off during the background pause
7	Current plug-in status
8	Current state of charge
9	Current battery level
10	How the app is switched
11	App category
12	The number of times the app switches to the foreground
13	Current WiFi connection status
14	Connected WiFi SSID or BSSID

Table 1

The electronic device may acquire the feature information of the plurality of features described above as a sample according to a preset frequency in the latest time period. The multi-dimensional feature data collected at one time constitutes one sample, and multiple samples collected multiple times constitute a sample set. The electronic device may bury a point in the preset path of the system, and when detecting that an application is activated, record the feature data of the relevant feature and the application identifier of the application at the startup time of the application, and identify the application identifier. Associated with the recorded feature data.

After constituting the sample set, the electronic device may mark each sample in the sample set to obtain a sample label of each sample, and the sample label may be an application identifier of the corresponding application, or an application category to which the corresponding application belongs. The sample tag may be classified into a "target application" or a "non-target application" with respect to the target application to be detected, or may be classified into "the same application type as the target application" or "different from the application type of the target application". .

Operation 802, when the data volume of the sample set exceeds a preset threshold, using the sample set as the node information of the root node of the decision tree; determining the node information of the root node as the current target sample set to be classified.

That is, the sample set of the root node is determined as the target sample set to be classified currently. For example, referring to FIG. 6A, for the sample set D{sample 1, sample 2, ... sample i ... sample n}, the electronic device can be made into the root node d of the decision tree, and the sample set D is taken as the node of the root node d information.

Operation 803, obtaining an information gain rate of the feature classification for the target sample set, and determining a maximum information gain rate from the information gain rate.

In one embodiment, it may first pass the formula H(D|A)=p ₁ H(D|A=A ₁ )+p ₂ H(D|A=A ₂ )+p ₃ H(D|A=A ₃ ) Calculate the conditional entropy H(D|A) of the feature A for the classification result of the sample set D, and then pass

Calculate the information gain g(D,A) of feature A for sample set D classification, and then follow

The information gain rate g _R (D, A) of the feature classification for the target sample set is calculated.

Operation 804, detecting whether the maximum information gain rate is greater than a preset threshold, and when the maximum information gain rate is greater than the preset threshold, performing operation 805; otherwise, performing operation 806.

For example, the electronic device can determine whether the maximum information gain g _R (D, A) max is greater than a preset threshold ε, which can be set according to actual needs.

In operation 805, the sample corresponding to the maximum information gain rate is selected as the partitioning feature, and the target sample set is divided according to the partitioning feature to generate at least one subsample set.

Operation 806, the current node is taken as a leaf node, and the sample with the largest number of samples is selected as the output of the leaf node.

For example, when the feature corresponding to the maximum information gain g _R (D, A) max is the feature A _g , the electronic device may select the feature A _g as the dividing feature.

Specifically, the electronic device may divide the sample set into several sub-sample sets according to the number of feature values of the divided features, and the number of the sub-sample sets is the same as the number of feature values. For example, the electronic device may divide the samples in the sample set with the same feature value into the same subsample set. For example, the feature values of the divided features include: 0, 1, 2, then, at this time, the samples whose feature values are 0 can be classified into one class, and the samples with the feature value 1 are classified into one class, and the feature values are The samples of 2 are classified into one category.

For example, when there are two values of the partitioning feature i, the sample set D can be divided into D ₁ {sample 1, sample 2, ... sample k} and D ₂ {sample k+1 ... sample n}. Then, a sub-division of sample sets features D ₁ and D ₂ A _g may be removed, i.e., AA _g.

Operation 807, the division feature of the sample in each sub-sample set is removed, the child node of the current node is generated, and the sub-sample set after the division feature is removed is used as the node information of the child node.

Among them, one subsample set corresponds to one child node. For example, the test map 6A generates the child nodes d1 and d2 of the root node d, and uses the subsample set D ₁ as the node information of the child node d _{1 and} the subsample set D ₂ as the node information of the child node d ₂ .

In an embodiment, the electronic device may further set the divided feature values corresponding to the child nodes on the path of the child node and the current node, so as to facilitate subsequent prediction whether the application is used, refer to FIG. 6C.

In operation 808, it is determined whether the number of categories of the sample in the subsample set after removing the partition feature corresponding to the child node is a preset number, and if yes, performing operation 809; otherwise, updating the target sample set to the subsample after removing the partition feature Set and return to perform operation 803.

When the child node does not meet the preset classification termination condition, the electronic device may continue to classify the sub-sample set corresponding to the child node by using the above-mentioned information gain classification method, for example, the child node d ₂ may be used as an example to calculate the A ₂ sample set. The information gain rate g _R (D, A) of each feature relative to the sample classification, the maximum information gain rate g _R (D, A) max is selected, when the maximum information gain rate g _R (D, A) max is greater than the preset For the threshold ε, the feature corresponding to the information gain rate g _R (D, A) may be selected as the partitioning feature A _g (such as the feature A _i+1 ), and the D2 is divided into several sub-sample sets based on the partitioning feature Ag, if D2 is divided into subsample sets D ₂₁ , D ₂₂ , D ₂₃ , and then the partition features A _g in the subsample sets D ₂₁ , D ₂₂ , D ₂₃ are removed, and the child nodes d ₂₁ , d _{22 of the} current node d ₂ are generated. And d ₂₃ , the sample sets D ₂₁ , D ₂₂ , and D ₂₃ after the division feature A _g are removed as the node information of the child nodes d ₂₁ , d ₂₂ , and d23 , respectively.

If the child node satisfies the preset classification termination condition, the electronic device may use the category of the sample in the subsample set as the output of the leaf node. If, after the removal, the subsample set has only the sample of the category "target application", then the electronic device can use the "target application" as the output of the leaf node.

In operation 809, the child node is used as a leaf node, and the output of the leaf node is set according to the category of the subsample set after the division feature is removed.

For example, when the sub-sample set D _{1 of the} child node d ₁ is classified, if the maximum information gain is small and the preset threshold, the electronic device may use the sample category with the largest number of samples in the sub-sample set D ₁ as the leaf node. Output. The maximum number of samples "target application", the electronic device may be an output of leaf nodes d ₁ "target application" as.

Operation 810, acquiring first feature data of each feature, where the first feature data is feature data of the corresponding feature at the predicted time.

After the decision tree model is constructed, the electronic device can acquire the feature data of the feature at the predicted time when the time at which the target application needs to be predicted is used.

Operation 811, obtaining a decision tree model for predicting whether the user will use the target application within a preset duration, and the starting time of the preset duration is the predicted time.

Operation 812, the first feature data is used as an input of the decision tree model, and the prediction result is output. When the prediction result is that the target application is not used within the preset duration, the operation 813 is performed; otherwise, the operation 815 is performed.

The electronic device can obtain a pre-built decision tree model, and use the first feature data as an input of the model to obtain a corresponding output result.

In operation 813, it is detected whether the target application belongs to the whitelist application. When the target application belongs to the whitelist application, operation 814 is performed. When the target application does not belong to the whitelist application, operation 815 is performed.

Operation 814, freeing the target application from freezing.

Operation 815, freezing the target application.

When the predicted result is that the target application is not used within the preset duration, and the target application is not in the whitelist, the platform freeze management module 224 may initiate a freeze operation on the target application to perform resources for the target application. Restrictions, such as the CPU limit sleep mode, CPU freeze sleep mode or process deep freeze mode, etc., can freeze the target application.

The application processing method described above, by constructing a decision tree model for predicting whether the target application is within a preset duration, and acquiring first feature data of each feature and a decision tree model of the target application, the first feature The data is used as an input of the decision tree model to obtain whether the user output by the decision tree model uses the prediction result of the target application within a preset duration, when the prediction result is that the target application is not used within the preset duration , the target application is frozen to limit the target application's occupation of resources. Since it is necessary to consume a certain amount of system resources when freezing the application, whether the target application is used within the preset duration by performing the prediction, and freezing when not in use, thereby reducing the user to freeze after the target application is frozen. The application is used within the preset duration, and the application needs to be thawed, which causes unnecessary freezing operations, improves the accuracy of freezing the target application, and thus improves the effectiveness of releasing system resources.

It should be understood that although the operations in the flowcharts of FIGS. 4, 5, 7, and 8 are sequentially displayed in accordance with the indication of the arrows, these operations are not necessarily performed in the order indicated by the arrows. Except as explicitly stated herein, the execution of these operations is not strictly limited, and the operations may be performed in other sequences. Moreover, at least some of the operations in FIGS. 4, 5, 7, and 8 may include multiple sub-operations or multiple stages, which are not necessarily performed at the same time, but may be at different times. Execution, the order of execution of these sub-operations or phases is not necessarily performed sequentially, but may be performed alternately or alternately with at least a portion of sub-operations or phases of other operations or other operations.

In one embodiment, as shown in FIG. 9, an application processing apparatus is provided, which includes a feature data acquisition module 902, a decision tree model acquisition module 904, a prediction module 906, and an application processing module 908. The feature data obtaining module 902 is configured to acquire first feature data of each feature, where the first feature data is feature data of the corresponding feature at the predicted time; the decision tree model obtaining module 904 is configured to obtain whether the user is predicted to be in the The decision tree model of the target application is used within the preset duration, and the starting time of the preset duration is the predicted time; the prediction module 906 is configured to use the first feature data as an input of the decision tree model, and output the predicted result; the application processing module 908 The target application is frozen when the predicted result is that the target application will not be used within the preset duration.

In one embodiment, another application processing device is provided as shown in FIG. 10, the device further comprising:

The decision tree construction module 910 is configured to acquire the second feature data of each feature as a sample, and generate a sample set, where the second feature data is feature data of the corresponding feature when the reference application is started before the predicted time, and the reference The application includes a target application; when the data volume of the sample set exceeds a preset threshold, the sample set is sample-classified according to the information gain rate of the feature classification for the sample set, and the generated sample is used to predict whether the user will use the target application within the preset duration. Decision tree model.

In an embodiment, the decision tree construction module 910 is further configured to use the sample set as the node information of the root node of the decision tree; determine the node information of the root node as the target sample set to be classified; and acquire the feature for the target sample set. Information gain rate; according to the information gain rate, the sample is selected from the target sample set as a partition feature; the target sample set is divided according to the partition feature to generate at least one subsample set; the partition feature of each subsample set is removed; and the current node is generated. a child node, and removing the subsample set after dividing the feature as the node information of the child node; determining whether the child node satisfies the preset classification termination condition; when the child node does not satisfy the preset classification termination condition, updating the target sample set to The sub-sample set after the feature is removed is removed, and the information gain rate of the acquired feature classification for the target sample set is returned. When the child node satisfies the preset classification termination condition, the child node is used as the leaf node, and the sub-node is removed according to the de-divided feature. The category of the sample set sets the output of the leaf node.

In an embodiment, the decision tree construction module 910 is further configured to determine a maximum information gain rate from the information gain rate; when the maximum information gain rate is greater than a preset threshold, select a sample corresponding to the maximum information gain rate as the partition feature. When the maximum information gain rate is not greater than the preset threshold, the current node is taken as the leaf node, and the sample with the largest number of samples is selected as the output of the leaf node.

In an embodiment, the decision tree construction module 910 is further configured to determine whether the number of categories of the sample in the subsample set after the partition feature is removed by the child node is a preset number; and the subsample set after the child node corresponding to remove the partition feature When the number of categories of the sample is a preset number, it is determined that the child node satisfies the preset classification termination condition.

In an embodiment, the decision tree construction module 910 is further configured to acquire an information gain of the feature classification for the target sample set; acquire split information of the feature classification for the target sample set; and classify the target sample set according to the information gain and the split information acquisition feature. Information gain rate.

In one embodiment, the decision tree building module 910 is also used to pass

Calculating the information gain rate of the feature classification for the target sample set; where D represents the sample set, g(D, A) is the information gain of feature A for the sample set D classification, and H _A (D) is the split information of feature A, g _R (D, A) is the information gain rate of feature A for sample set D; g(D, A) is passed

Calculated; where H(D) is the empirical entropy of the sample set D classification, H(D|A) is the conditional entropy of feature A for the sample set D classification, and n is the number of sample values of feature A, p _i is The probability that the feature A takes the sample of the i-th value in the sample set, n and i are positive integers greater than zero.

In one embodiment, as shown in FIG. 11, yet another application processing device is provided, the device further comprising:

The application detection module 912 is configured to detect whether the target application belongs to a whitelist application.

The application processing module 908 is further configured to freeze the target application when the prediction result is that the target application is not used within the preset duration; when the prediction result is that the target application is used within the preset duration, or the target application belongs to white When the list is applied, the target application is free from freezing.

The division of each module in the application processing device is for illustrative purposes only. In other embodiments, the application processing device may be divided into different modules as needed to complete all or part of the functions of the application processing device.

For specific definitions of the application processing device, reference may be made to the above definition of the application processing method, and details are not described herein again. The various modules in the application processing device described above may be implemented in whole or in part by software, hardware, and combinations thereof. Each of the above modules may be embedded in or independent of the processor in the electronic device, or may be stored in a memory in the electronic device in a software format, so that the processor calls to perform operations corresponding to the above modules.

The implementation of each module in the application processing apparatus provided in the embodiments of the present application may be in the form of a computer program. The computer program can run on an electronic device such as a terminal or a server. The program module of the computer program can be stored on a memory of the electronic device. When the computer program is executed by the processor, the operation of the application processing method described in the embodiment of the present application is implemented.

The various modules in the application processing device described above may be implemented in whole or in part by software, hardware, and combinations thereof. The above modules may be embedded in the hardware in the processor or in the memory in the server, or may be stored in the memory in the server, so that the processor calls the corresponding operations of the above modules. As used in this application, the terms "module" and the like are intended to mean a computer-related entity, which may be hardware, a combination of hardware and software, software, or software in execution. For example, a module can be, but is not limited to being, a process running on a processor, a processor, an object, an executable, a thread of execution, a program, and/or a computer. By way of illustration, both an application running on a server and a server can be a module. One or more modules can reside within a process and/or a thread of execution, and a module can be located in a computer and/or distributed between two or more computers.

In one embodiment, an electronic device is provided, including a memory, a processor, and a computer program stored on the memory and operable on the processor, and the application processing provided by the above embodiments is implemented when the processor executes the computer program The operation of the method.

In one embodiment, there is also provided a computer readable storage medium having stored thereon a computer program for performing application processing as described in various embodiments of the present application when executed by a processor The operation of the method.

In one embodiment, a computer program product comprising instructions, when executed on a computer, causes the computer to perform the application processing methods described in the various embodiments of the present application.

The embodiment of the present application also provides a computer device. As shown in FIG. 12, for the convenience of description, only the parts related to the embodiments of the present application are shown. If the specific technical details are not disclosed, please refer to the method part of the embodiment of the present application. The computer device may be any terminal device including a mobile phone, a tablet computer, a PDA (Personal Digital Assistant), a POS (Point of Sales), a vehicle-mounted computer, a wearable device, and the like, taking a computer device as a mobile phone as an example. :

FIG. 12 is a block diagram showing a part of a structure of a mobile phone related to a computer device according to an embodiment of the present application. Referring to FIG. 12, the mobile phone includes: a radio frequency (RF) circuit 1210, a memory 1220, an input unit 1230, a display unit 1240, a sensor 1250, an audio circuit 1260, a wireless fidelity (WiFi) module 1270, and a processor 1280. And power supply 1290 and other components. It will be understood by those skilled in the art that the structure of the handset shown in FIG. 12 does not constitute a limitation to the handset, and may include more or less components than those illustrated, or some components may be combined, or different components may be arranged.

The RF circuit 1210 can be used for receiving and transmitting information during the transmission and reception of information or during the call. The downlink information of the base station can be received and processed by the processor 1280. The uplink data can also be sent to the base station. Generally, RF circuits include, but are not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a Low Noise Amplifier (LNA), a duplexer, and the like. In addition, RF circuitry 1210 can also communicate with the network and other devices via wireless communication. The above wireless communication may use any communication standard or protocol, including but not limited to Global System of Mobile communication (GSM), General Packet Radio Service (GPRS), Code Division Multiple Access (Code Division). Multiple Access (CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (LTE), e-mail, Short Messaging Service (SMS), and the like.

The memory 1220 can be used to store software programs and modules, and the processor 1280 executes various functional applications and data processing of the mobile phone by running software programs and modules stored in the memory 1220. The memory 1220 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as an application of a sound playing function, an application of an image playing function, etc.); The data storage area can store data (such as audio data, address book, etc.) created according to the use of the mobile phone. Moreover, memory 1220 can include high speed random access memory, and can also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.

The input unit 1230 can be configured to receive input numeric or character information and to generate key signal inputs related to user settings and function control of the handset 1200. Specifically, the input unit 1230 may include a touch panel 1231 and other input devices 1232. The touch panel 1231, which may also be referred to as a touch screen, can collect touch operations on or near the user (such as a user using a finger, a stylus, or the like on the touch panel 1231 or near the touch panel 1231. Operation) and drive the corresponding connection device according to a preset program. In one embodiment, the touch panel 1231 may include two parts of a touch detection device and a touch controller. Wherein, the touch detection device detects the touch orientation of the user, and detects a signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts the touch information into contact coordinates, and sends the touch information. The processor 1280 is provided and can receive commands from the processor 1280 and execute them. In addition, the touch panel 1231 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic waves. In addition to the touch panel 1231, the input unit 1230 may also include other input devices 1232. Specifically, other input devices 1232 may include, but are not limited to, one or more of a physical keyboard, function keys (such as volume control buttons, switch buttons, etc.).

The display unit 1240 can be used to display information input by the user or information provided to the user as well as various menus of the mobile phone. The display unit 1240 may include a display panel 1241. In one embodiment, the display panel 1241 may be configured in the form of a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like. In one embodiment, the touch panel 1231 can cover the display panel 1241. When the touch panel 1231 detects a touch operation thereon or nearby, the touch panel 1231 transmits to the processor 1280 to determine the type of the touch event, and then the processor 1280 is The type of touch event provides a corresponding visual output on display panel 1241. Although in FIG. 12, the touch panel 1231 and the display panel 1241 are used as two independent components to implement the input and input functions of the mobile phone, in some embodiments, the touch panel 1231 and the display panel 1241 may be integrated. Realize the input and output functions of the phone.

The handset 1200 can also include at least one type of sensor 1250, such as a light sensor, motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 1241 according to the brightness of the ambient light, and the proximity sensor may close the display panel 1241 and/or when the mobile phone moves to the ear. Or backlight. The motion sensor may include an acceleration sensor, and the acceleration sensor can detect the magnitude of the acceleration in each direction, and the magnitude and direction of the gravity can be detected at rest, and can be used to identify the gesture of the mobile phone (such as horizontal and vertical screen switching), and vibration recognition related functions (such as Pedometer, tapping, etc.; in addition, the phone can also be equipped with gyroscopes, barometers, hygrometers, thermometers, infrared sensors and other sensors.

Audio circuitry 1260, speaker 1261, and microphone 1262 can provide an audio interface between the user and the handset. The audio circuit 1260 can transmit the converted electrical data of the received audio data to the speaker 1261, and convert it into a sound signal output by the speaker 1261; on the other hand, the microphone 1262 converts the collected sound signal into an electrical signal, by the audio circuit 1260. After receiving, it is converted into audio data, and then processed by the audio data output processor 1280, transmitted to another mobile phone via the RF circuit 1210, or outputted to the memory 1220 for subsequent processing.

WiFi is a short-range wireless transmission technology. The mobile phone through the WiFi module 1270 can help users to send and receive e-mail, browse the web and access streaming media, etc. It provides users with wireless broadband Internet access. Although FIG. 12 shows the WiFi module 1270, it will be understood that it does not belong to the essential configuration of the handset 1200 and may be omitted as needed.

The processor 1280 is a control center for the handset that connects various portions of the entire handset using various interfaces and lines, by executing or executing software programs and/or modules stored in the memory 1220, and invoking data stored in the memory 1220, The phone's various functions and processing data, so that the overall monitoring of the phone. In one embodiment, processor 1280 can include one or more processing units. In one embodiment, the processor 1280 can integrate an application processor and a modem, wherein the application processor primarily processes an operating system, a user interface, an application, etc.; the modem primarily processes wireless communications. It will be appreciated that the above described modem may also not be integrated into the processor 1280. For example, the processor 1280 can integrate an application processor and a baseband processor, and the baseband processor and other peripheral chips can form a modem. The handset 1200 also includes a power source 1290 (such as a battery) that powers the various components. Preferably, the power source can be logically coupled to the processor 1280 via a power management system to manage functions such as charging, discharging, and power management through the power management system.

In one embodiment, the handset 1200 can also include a camera, a Bluetooth module, and the like.

In the embodiment of the present application, the processor included in the mobile phone implements the application processing method described above when executing a computer program stored in the memory.

Any reference to a memory, storage, database or other medium used herein may include non-volatile and/or volatile memory. Suitable non-volatile memories can include read only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM), which acts as an external cache. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronization. Link (Synchlink) DRAM (SLDRAM), Memory Bus (Rambus) Direct RAM (RDRAM), Direct Memory Bus Dynamic RAM (DRDRAM), and Memory Bus Dynamic RAM (RDRAM).

The above embodiments are merely illustrative of several embodiments of the present application, and the description thereof is more specific and detailed, but is not to be construed as limiting the scope of the claims. It should be noted that a number of variations and modifications may be made by those skilled in the art without departing from the spirit and scope of the present application. Therefore, the scope of the invention should be determined by the appended claims.

Claims (24)

1. An application processing method includes:

   Obtaining first feature data of each feature, where the first feature data is feature data of the corresponding feature at the predicted time;

   Obtaining a decision tree model for predicting whether the user will use the target application within a preset duration, where the starting time of the preset duration is a predicted moment;

   Using the first feature data as an input of the decision tree model, and outputting a prediction result; and

   When the predicted result is that the target application is not used within a preset duration, the target application is frozen.
2. The method according to claim 1, wherein before the acquiring the first feature data of each feature preset, the method further comprises:

   Obtaining a second feature data of each feature as a sample, and generating a sample set, where the second feature data is feature data of a corresponding feature when the reference application is started before the predicted time, the reference application includes the Target application; and

   When the data volume of the sample set exceeds a preset threshold, classifying the sample set according to the information gain rate of the feature classification for the sample set, and generating a method for predicting whether the user will use the target application within a preset duration Decision tree model.
3. The method according to claim 2, wherein the sample grouping is performed on the sample set according to the information gain rate of the feature classification for the sample set, and generating is used to predict whether the user will use the target application within a preset time duration. Decision tree model, including:

   Using the sample set as node information of a root node of the decision tree;

   Determining the node information of the root node as a target sample set to be classified currently;

   Obtaining the information gain rate of the feature classification for the target sample set;

   Selecting a sample from the target sample set as a division feature according to the information gain rate;

   And dividing the target sample set according to the dividing feature to generate at least one sub-sample set;

   Removing the partitioning features of the samples in each of the subsample sets;

   Generating a child node of the current node, and removing the subsample set after dividing the feature as node information of the child node;

   Determining whether the child node meets a preset classification termination condition;

   And when the child node does not satisfy the preset classification termination condition, updating the target sample set to the sub-sample set after the dividing feature is removed, and returning an information gain rate of performing the acquiring feature for the target sample set classification;

   When the child node satisfies the preset classification termination condition, the child node is used as a leaf node, and the output of the leaf node is set according to the category of the subsample set after the division feature is removed.
4. The method according to claim 3, wherein the selecting a sample from the target sample set according to the information gain rate as a dividing feature comprises:

   Determining a maximum information gain rate from the information gain rate;

   When the maximum information gain rate is greater than a preset threshold, selecting a sample corresponding to the maximum information gain rate as a partitioning feature; and

   When the maximum information gain rate is not greater than a preset threshold, the current node is taken as a leaf node, and the sample with the largest number of samples is selected as the output of the leaf node.
5. The method according to claim 3, wherein the determining whether the child node satisfies a preset classification termination condition comprises:

   Determining whether the number of categories of samples in the subsample set after removing the partition feature corresponding to the child node is a preset number; and

   And determining, when the number of categories of the sample of the subsample set in the sub-node corresponding to the de-divided feature is the preset quantity, determining that the sub-node satisfies the preset classification termination condition.
6. The method according to any one of claims 3 to 5, wherein the information gain rate of the acquisition feature for the target sample set classification comprises:

   Acquiring the information gain of the feature classification for the target sample set;

   Obtaining split information of the feature classification for the target sample set; and

   And an information gain rate classified by the information gain and the split information acquisition feature for the target sample set according to the information gain.
7. The method according to claim 6, wherein the information gain rate classified according to the information gain and the split information acquisition feature for the target sample set comprises:

   by

   

   Calculating the information gain rate of the feature classification for the target sample set;

   Where D represents the sample set, g(D, A) is the information gain of feature A for sample set D classification, H _A (D) is the split information of feature A, and g _R (D, A) is feature A for the sample set Information gain rate of the D classification;

   The g(D, A) passes

   

   Calculated

   Where H(D) is the empirical entropy of the sample set D classification, H(D|A) is the conditional entropy of the feature A for the sample set D classification, n is the number of sample values of the feature A, and p _i is the feature A The probability that the sample of the i-th value appears in the sample set, n and i are positive integers greater than zero.
8. The method according to claim 1, wherein before the freezing of the target application, the method further comprises:

   Detecting whether the target application belongs to a whitelist application, and when the target application belongs to a whitelist application, freezes the target application; and when the target application does not belong to the whitelist application, executing the opposite The target application is frozen.
9. An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following operations:

   Obtaining first feature data of each feature, where the first feature data is feature data of the corresponding feature at the predicted time;

   Obtaining a decision tree model for predicting whether the user will use the target application within a preset duration, where the starting time of the preset duration is a predicted moment;

   Using the first feature data as an input of the decision tree model, and outputting a prediction result; and

   When the predicted result is that the target application is not used within a preset duration, the target application is frozen.
10. The electronic device according to claim 9, wherein the processor further performs the following operations before performing the acquiring the first feature data of each feature preset:

    Obtaining a second feature data of each feature as a sample, and generating a sample set, where the second feature data is feature data of a corresponding feature when the reference application is started before the predicted time, the reference application includes the Target application; and

    When the data volume of the sample set exceeds a preset threshold, classifying the sample set according to the information gain rate of the feature classification for the sample set, and generating a method for predicting whether the user will use the target application within a preset duration Decision tree model.
11. The electronic device according to claim 10, wherein the processor performs sample classification on the sample set according to an information gain rate of the feature classification for the sample set, and generates a method for predicting whether the user will be at a preset duration When the decision tree model of the target application is used, the following operations are also performed:

    Using the sample set as node information of a root node of the decision tree;

    Determining the node information of the root node as a target sample set to be classified currently;

    Obtaining the information gain rate of the feature classification for the target sample set;

    Selecting a sample from the target sample set as a division feature according to the information gain rate;

    And dividing the target sample set according to the dividing feature to generate at least one sub-sample set;

    Removing the partitioning features of the samples in each of the subsample sets;

    Generating a child node of the current node, and removing the subsample set after dividing the feature as node information of the child node;

    Determining whether the child node meets a preset classification termination condition;

    And when the child node does not satisfy the preset classification termination condition, updating the target sample set to the sub-sample set after the dividing feature is removed, and returning an information gain rate of performing the acquiring feature for the target sample set classification;

    When the child node satisfies the preset classification termination condition, the child node is used as a leaf node, and the output of the leaf node is set according to the category of the subsample set after the division feature is removed.
12. The electronic device according to claim 11, wherein when the processor performs the selecting a sample from the target sample set according to the information gain rate as a dividing feature, the processor further performs the following operations:

    Determining a maximum information gain rate from the information gain rate;

    When the maximum information gain rate is greater than a preset threshold, selecting a sample corresponding to the maximum information gain rate as a partitioning feature; and

    When the maximum information gain rate is not greater than a preset threshold, the current node is taken as a leaf node, and the sample with the largest number of samples is selected as the output of the leaf node.
13. The electronic device according to claim 11, wherein when the processor performs the determining whether the child node satisfies a preset classification termination condition, the processor further performs the following operations:

    Determining whether the number of categories of samples in the subsample set after removing the partition feature corresponding to the child node is a preset number; and

    And determining, when the number of categories of the sample of the subsample set in the sub-node corresponding to the de-divided feature is the preset quantity, determining that the sub-node satisfies the preset classification termination condition.
14. The electronic device according to any one of claims 11 to 13, wherein when the processor performs the information gain rate of the acquisition feature for the target sample set classification, the following operations are also performed:

    Acquiring the information gain of the feature classification for the target sample set;

    Obtaining split information of the feature classification for the target sample set; and

    And an information gain rate classified by the information gain and the split information acquisition feature for the target sample set according to the information gain.
15. The electronic device according to claim 14, wherein the processor performs the following operations when performing the information gain rate according to the information gain and the split information acquisition feature for the target sample set:

    by

    

    Calculating the information gain rate of the feature classification for the target sample set;

    Where D is the sample set, g(D, A) is the information gain of feature A for the sample set D, HA(D) is the split information of feature A, gR(D, A) is the feature A, and sample D is classified. Information gain rate;

    The g(D, A) passes

    

    Calculated

    Where H(D) is the empirical entropy of the sample set D classification, H(D|A) is the conditional entropy of the feature A for the sample set D classification, n is the number of sample values of the feature A, and pi is the feature A. The probability that a sample of i values appears in the sample set, n and i are positive integers greater than zero.
16. The electronic device according to claim 9, wherein the processor further performs the following operations before performing the freezing of the target application:

    Detecting whether the target application belongs to a whitelist application, and when the target application belongs to a whitelist application, freezes the target application; and when the target application does not belong to the whitelist application, executing the opposite The target application is frozen.
17. A computer readable storage medium having stored thereon a computer program, wherein the computer program is executed by the processor as follows:

    Obtaining first feature data of each feature, where the first feature data is feature data of the corresponding feature at the predicted time;

    Obtaining a decision tree model for predicting whether the user will use the target application within a preset duration, where the starting time of the preset duration is a predicted moment;

    Using the first feature data as an input of the decision tree model, and outputting a prediction result; and

    When the predicted result is that the target application is not used within a preset duration, the target application is frozen.
18. The computer readable storage medium according to claim 17, wherein the processor further performs the following operations before performing the acquiring the first feature data of each feature preset:

    Obtaining a second feature data of each feature as a sample, and generating a sample set, where the second feature data is feature data of a corresponding feature when the reference application is started before the predicted time, the reference application includes the Target application; and

    When the data volume of the sample set exceeds a preset threshold, classifying the sample set according to the information gain rate of the feature classification for the sample set, and generating a method for predicting whether the user will use the target application within a preset duration Decision tree model.
19. The computer readable storage medium according to claim 18, wherein the processor performs sample classification on the sample set according to an information gain rate of the feature classification for the sample set, and generates a prediction for whether the user is predicting When using the decision tree model of the target application within the duration, the following operations are also performed:

    Using the sample set as node information of a root node of the decision tree;

    Determining the node information of the root node as a target sample set to be classified currently;

    Obtaining the information gain rate of the feature classification for the target sample set;

    Selecting a sample from the target sample set as a division feature according to the information gain rate;

    And dividing the target sample set according to the dividing feature to generate at least one sub-sample set;

    Removing the partitioning features of the samples in each of the subsample sets;

    Generating a child node of the current node, and removing the subsample set after dividing the feature as node information of the child node;

    Determining whether the child node meets a preset classification termination condition;

    And when the child node does not satisfy the preset classification termination condition, updating the target sample set to the sub-sample set after the dividing feature is removed, and returning an information gain rate of performing the acquiring feature for the target sample set classification;

    When the child node satisfies the preset classification termination condition, the child node is used as a leaf node, and the output of the leaf node is set according to the category of the subsample set after the division feature is removed.
20. The computer readable storage medium according to claim 19, wherein the processor further performs the following operations when performing the selecting a sample from the target sample set according to the information gain rate as a dividing feature:

    Determining a maximum information gain rate from the information gain rate;

    When the maximum information gain rate is greater than a preset threshold, selecting a sample corresponding to the maximum information gain rate as a partitioning feature; and

    When the maximum information gain rate is not greater than a preset threshold, the current node is taken as a leaf node, and the sample with the largest number of samples is selected as the output of the leaf node.
21. The computer readable storage medium according to claim 19, wherein when the processor performs the determining whether the child node satisfies a preset classification termination condition, the processor further performs the following operations:

    Determining whether the number of categories of samples in the subsample set after removing the partition feature corresponding to the child node is a preset number; and

    And determining, when the number of categories of the sample of the subsample set in the sub-node corresponding to the de-divided feature is the preset quantity, determining that the sub-node satisfies the preset classification termination condition.
22. The computer readable storage medium according to any one of claims 19 to 21, wherein when the processor performs the information gain rate of the acquisition feature for the target sample set classification, the following operations are also performed:

    Acquiring the information gain of the feature classification for the target sample set;

    Obtaining split information of the feature classification for the target sample set; and

    And an information gain rate classified by the information gain and the split information acquisition feature for the target sample set according to the information gain.
23. The computer readable storage medium according to claim 22, wherein the processor performs the following information gain ratio according to the information gain and the split information acquisition feature for a target sample set, and further performs the following operating:

    by

    

    Calculating the information gain rate of the feature classification for the target sample set;

    Where D is the sample set, g(D, A) is the information gain of feature A for the sample set D, HA(D) is the split information of feature A, gR(D, A) is the feature A, and sample D is classified. Information gain rate;

    The g(D, A) passes

    

    Calculated

    Where H(D) is the empirical entropy of the sample set D classification, H(D|A) is the conditional entropy of the feature A for the sample set D classification, n is the number of sample values of the feature A, and pi is the feature A. The probability that a sample of i values appears in the sample set, n and i are positive integers greater than zero.
24. The computer readable storage medium according to claim 17, wherein the processor further performs the following operations before performing the freezing of the target application:

    Detecting whether the target application belongs to a whitelist application, and when the target application belongs to a whitelist application, freezes the target application; and when the target application does not belong to the whitelist application, executing the opposite The target application is frozen.

Images