CN116700832A - Application management method, device, terminal and storage medium - Google Patents

Abstract

The embodiment of the present application discloses an application management method, device, terminal, and storage medium, which belong to the field of computer technology. The method includes: acquiring the terminal operating status and user interest information of the first terminal, and the terminal operating status includes all The system state of the terminal and the application state of the application program in the terminal, the user interest information is used to indicate the user usage habits of the application program; input the terminal operation state and the user interest information into the antivirus prediction model , obtaining the first killing probability respectively corresponding to the at least one application program; performing background killing management on the at least one application program according to the first killing probability respectively corresponding to the at least one application program. The method can automatically check and kill background applications according to user habits, application status, and system status, thereby improving the efficiency of checking and killing applications.

Description

Application management method, device, terminal and storage medium

technical field

The embodiments of the present application relate to the field of computer technology, and in particular to an application management method, device, terminal and storage medium.

Background technique

During the use of the terminal, there are problems such as slow system response, fast power consumption, and short battery life. These problems are closely related to the application background management mechanism.

In the related art, opening multiple background applications by the user will lead to an increase in memory usage of the terminal. At this time, the user can enter the background management page and close the background applications that are not in use temporarily, thereby reducing the memory usage rate.

In the method in the related art, the user needs to frequently enter the background management page and manually close unused background applications, and the background management efficiency is low and the operation is complicated.

Contents of the invention

Embodiments of the present application provide an application management method, device, terminal, and storage medium. Described technical scheme is as follows:

According to one aspect of the present application, an application management method is provided, the method is applied in a first terminal, and the method includes:

Acquiring the terminal running state and user interest information of the first terminal, the terminal running state includes the system state of the terminal and the application state of the application program in the terminal, and the user interest information is used to indicate that the application program user habits;

Inputting the terminal operation status and the user interest information into the antivirus prediction model to obtain the first antivirus probabilities respectively corresponding to the at least one application program;

Perform background checking and killing management on the at least one application program according to the first checking and killing probability respectively corresponding to the at least one application program.

According to another aspect of the present application, an application management device is provided, the device is used to implement a first terminal, and the device includes:

An acquisition module, configured to acquire the terminal running state and user interest information of the first terminal, the terminal running state including the system state of the terminal and the application state of the application program in the terminal, and the user interest information is used for Indicates the usage habits of users of said application;

A predicting module, configured to input the terminal operating status and the user interest information into a killing prediction model to obtain first killing probabilities respectively corresponding to the at least one application;

The management module is configured to perform background checking and killing management on the at least one application program according to the first checking and killing probability respectively corresponding to the at least one application program.

According to another aspect of the present application, a terminal is provided, the terminal includes a processor and a memory, at least one instruction is stored in the memory, and the instruction is loaded and executed by the processor to implement the Implement the provided application management methods.

According to another aspect of the present application, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, and the instruction is loaded and executed by a processor to implement the application management method provided by the implementation of the present application .

The beneficial effects brought by the technical solutions provided by the embodiments of the present application may include:

The terminal predicts the application programs that the user does not need to use according to the current operating status and user interest information, and then performs scanning and killing management on these application programs. The terminal can use the neural network model (antivirus prediction model) to automatically perform application antivirus, and the antivirus prediction results can fit the user's usage habits, ensuring efficient and accurate background antivirus.

Description of drawings

In order to more clearly introduce the technical solutions in the embodiments of the present application, the drawings that need to be used in the description of the embodiments of the present application will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present application , for those skilled in the art, other drawings can also be obtained based on these drawings without creative work.

Fig. 1 is a structural block diagram of a computer system provided by an exemplary embodiment of the present application;

Fig. 2 is a flowchart of an application management method provided by an exemplary embodiment of the present application;

Fig. 3 is a flowchart of another application management method provided by another exemplary embodiment of the present application;

Fig. 4 is a structural block diagram of an application management device provided by an exemplary embodiment of the present application;

Fig. 5 is a structural block diagram of a terminal provided by an exemplary embodiment of the present application.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application clearer, the implementation manners of the present application will be further described in detail below in conjunction with the accompanying drawings.

When the following description refers to the accompanying drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with aspects of the present application as recited in the appended claims.

In the description of the present application, it should be understood that the terms "first", "second" and so on are used for descriptive purposes only, and should not be understood as indicating or implying relative importance. In the description of this application, it should be noted that, unless otherwise clearly stipulated and limited, the terms "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral Ground connection; it can be mechanical connection or electrical connection; it can be direct connection or indirect connection through an intermediary. Those of ordinary skill in the art can understand the specific meanings of the above terms in this application in specific situations. In addition, in the description of the present application, unless otherwise specified, "plurality" means two or more. "And/or" describes the association relationship of the associated objects, indicating that there may be three types of relationships, for example, A and/or B may indicate: A exists alone, A and B exist simultaneously, and B exists independently. The character "/" generally indicates that the objects associated before and after are in an "or" relationship.

Fig. 1 shows a schematic diagram of a computer system provided by an exemplary embodiment of the present application, where the computer system includes a terminal and a server.

Exemplarily, the application management method shown in the embodiment of the present application can be applied in a terminal, and the terminal 120 has a display screen and an application management function. Terminals can include mobile phones, tablet computers, laptops, laptops, desktop computers, all-in-one computers, intelligent robot workstations, TVs, set-top boxes, smart glasses, smart watches, digital cameras, MP4 playback terminals, MP5 playback terminals, learning machines , point reader, electronic paper book, electronic dictionary, vehicle-mounted terminal, virtual reality (English: Virtual Reality, abbreviation: VR) playback terminal or augmented reality (English: Augmented Reality, abbreviation: AR) playback terminal, etc.

In an optional implementation manner, the application management method provided in this application may be applied to an operating system of a terminal.

Exemplarily, the application management algorithm 102 is stored on the terminal 120, and when the client needs to use the social application management function, the client can call the application management algorithm to complete the application management.

Terminal 120 includes a first memory and a first processor. An application management algorithm is stored in the first memory; the above application management algorithm is invoked and executed by the first processor to implement the application management method provided in this application. The first memory may include but not limited to the following: random access memory (Random Access Memory, RAM), read-only memory (ReadOnly Memory, ROM), programmable read-only memory (Programmable Read-Only Memory, PROM), programmable Erasable Programmable Read-Only Memory (EPROM), and Electric Erasable Programmable Read-Only Memory (EEPROM).

The first processor may be composed of one or more integrated circuit chips. Optionally, the first processor may be a general processor, such as a central processing unit (Central Processing Unit, CPU) or a network processor (Network Processor, NP). Optionally, the first processor may implement the application management method provided in this application by running a program or code.

The terminal 120 and the server 140 are connected to each other through a wired or wireless network.

The server 140 is used to provide background services for the operating system of the terminal 120 . For example, the server 140 may be the background server 140 of the above-mentioned application program. The server 140 may be one server 140, or a server 140 cluster composed of multiple servers 140, or a cloud computing service center.

Schematically, the server 140 includes a processor and a memory, and the memory further includes an access module and an information flow push module. The server 140 is used to provide background services for the operating system. Optionally, the server 140 undertakes the main calculation work, and the terminal 120 undertakes the secondary calculation work; or, the server 140 undertakes the secondary calculation work, and the terminal 120 undertakes the main calculation work; or, a distributed computing architecture is adopted between the server 140 and the terminal 120 Perform collaborative computing.

Server 140 includes a second memory and a second processor. An application management algorithm is stored in the second memory; the above application management algorithm is invoked by the second processor to implement the application management method provided in this application. Optionally, the second memory may include but not limited to the following: RAM, ROM, PROM, EPROM, EEPROM. Optionally, the second processor may be a general processor, such as CPU or NP.

Please refer to FIG. 2 , which is a flowchart of an application management method provided in an exemplary embodiment of the present application. The application management method may be applied to a first terminal, and the first terminal may be the terminal shown in FIG. 1 above.

In Figure 2, the application management method includes:

Step 210, acquire the terminal running state and user interest information of the first terminal, the terminal running state includes the terminal system state and the application state of the application program in the terminal, and the user interest information is used to indicate the user's usage habit of the application program.

Optionally, the method further includes: displaying an authorization pop-up window for obtaining the user interest information, and the authorization window is used to request the user to authorize the acquisition of the user interest information. In response to receiving an authorization operation in the authorization pop-up window, it is determined that the user authorizes acquisition of user interest information, and then step 210 is performed.

Optionally, the terminal operating status and user interest information may be acquired in real time. For example, if the first terminal periodically uses the method provided by the embodiment of the present application to perform background scanning and killing management, the first terminal will obtain the current terminal operating status and user interest information of the first terminal in real time each time the process is triggered.

The terminal running state includes the system state of the operating system and the application state of the application program. Wherein, the application programs may refer to all the application programs on the first terminal, or the application programs may include the top n application programs in terms of memory usage.

Optionally, the application state in this embodiment of the present application may also refer to a process state. Process status includes: foreground process, visible process, service process, background process and running status of empty process.

Wherein, the system status includes: at least one of: processor (Central Processing Unit, CPU) status, memory usage status, foreground status, background status, network status, application unresponsive status, application crash status, and application startup time.

• Processor status, which may include at least one of the following information: processor utilization, speed, number of processes, number of threads, number of handles, uptime.

·Memory usage status, which may include at least one of the following information: used memory, available memory, committed memory, cached memory, paged buffer pool size, and non-paged buffer pool size.

·Foreground state, including at least one of the following information: screen state (screen off or on), foreground application state (or called foreground process state). The state of the foreground application may include at least one of the following information: the foreground is the system desktop, the application program running in the foreground, the duration of the application program running in the foreground, and the memory usage of the foreground application program/foreground process.

Background status, including at least one of the following information: background application identification, number of background applications, background process identification, number of background processes, running time of background applications in the background, total running time of background applications, background application/ Memory usage of background processes.

·Network status, which may include at least one of the following information: sending rate, receiving rate, delay, network type.

· The application non-response situation may include the number of application non-response occurrences in the past preset time period.

·Application flashback conditions may include the number of application flashback occurrences in the past preset time period.

·Application startup time may include the last startup time of each application, or the average time of application startup in the past preset time period.

The application status includes the running status corresponding to at least one application program. The application status of an application may include at least one of the following information: application identification, whether it is running, running in the foreground/background, process information, elapsed running time, package name, application startup time, memory usage, whether it is set background protection.

The user interest information is the information obtained by quantifying the user's usage habits of using the application program. For example, the user interest information is based on the user's usage time of each application, or the user interest information is based on the user's habit of manually cleaning the background, or the user interest information is based on the background protection application set by the user. get the information.

For example, the user interest information may be the information obtained by normalizing the usage time of each application program according to the usage time of the user in the past day, and the user interest information includes the application identifier and the normalized result of the usage duration corresponding to the application identifier .

For another example, the user interest information may be obtained according to the situation that the user has manually cleaned up background applications in the past day. The first terminal counts the applications manually cleaned up by the user in the past day and the times of cleaning, and normalizes the times of cleaning to obtain a normalized result of the times of cleaning. The user interest information may include the application program identifier and the normalized result of cleaning times corresponding to the application program identifier.

Optionally, the method for obtaining the user interest information includes: obtaining the use durations corresponding to the n application programs within a preset period of time; performing normalization processing on the n use durations to obtain the interest weights corresponding to the n application programs respectively; Determine n interest weights as user interest information.

For example, if n is 10, then the first terminal obtains the duration of the user using 10 applications in the past 24 hours: 1 hour, 1 hour, 1 hour, 1 hour, 2 hours, 2 hours, 2 hours, 2 hours Hours, 3 hours, 3 hours. According to the average value normalization algorithm, first calculate the total duration as: 1+1+1+1+2+2+2+2+3+3=18 hours. After normalizing the ten durations, we get: 1/18, 1/18, 1/18, 1/18, 1/9, 1/9, 1/9, 1/9, 1/6, 1/6 . Then the interest weights of the ten applications are: 1/18, 1/18, 1/18, 1/18, 1/9, 1/9, 1/9, 1/9, 1/6, 1/ 6. Of course, other normalization algorithms may also be used, for example, methods such as linear normalization, nonlinear normalization, and standardized normalization may be used to obtain interest weights according to the usage time of each application program.

Optionally, the method for obtaining the terminal running state includes: obtaining the system state of the first terminal, the system state includes: at least one of processor state, memory usage state, foreground state, and background state, and the foreground state includes: screen state, At least one of the foreground application states; obtain the application states corresponding to the n application programs respectively, and the application states include: at least one of process information, elapsed running time, and package name.

In an optional embodiment, if the input data of the model is shown in Table 1, the weight of user interest may include a normalized result of application usage time. The terminal running status may include system status and application status. The system state includes: CPU load, available memory, free memory, and foreground state; the application state includes: process ID, running time, and package name of the application.

Table 1

Step 220, input terminal operation status and user interest information into the antivirus prediction model to obtain first antivirus probabilities corresponding to at least one application program respectively.

The killing prediction model is a neural network model. The killing prediction model may use at least one of a fully connected neural network, a convolutional neural network, and a recurrent neural network (Recurrent Neural Network, RNN). For example, the killing prediction model includes a feature extraction layer and a fully connected layer. The feature extraction layer is used to extract features from the input data, and the fully connected layer is used to predict the features output by the feature extraction layer to obtain output data. The feature extraction layer can be implemented with a convolutional layer.

Optionally, when the application state input in step 210 includes the application states of n application programs, the first killing probability output in step 220 includes n first killing probabilities respectively corresponding to the n application programs.

Optionally, the terminal running status and user interest information are input into the antivirus prediction model to obtain the first antivirus probability and the first system state respectively corresponding to at least one application program. The first system state is a system state that is expected to be achieved after killing the background application according to the first killing probability. The first system status may include the information described in the above system status.

In an optional embodiment, the model input data is (n+1)*M dimensional data, and (n+1) includes: 1 set of system states and n sets of application data corresponding to n application programs. The system state includes M data, and a set of application data includes (M-1) application states and user interest information corresponding to the application, where n and M are positive integers.

The system state may include M pieces of data in the following information: processor state, memory usage state, foreground state, background state, network state, application unresponsiveness, application flashback, and application startup time.

The application status can include (M-1) pieces of data in the following information: application identification, whether it is running, running in the foreground/background, process information, running time, package name, application startup time, memory usage, whether it is Background protection is set.

Correspondingly, the model output data includes (n+1)-dimensional data. (n+1) includes killing probabilities corresponding to the n application programs (the first killing probability), and an expected system state (the first system state).

The estimated system state is predicted by the antivirus prediction model, and the system state of the first terminal after the background application is detected and killed according to the first antivirus probability. The estimated system status may include one of the following information: processor status, memory usage status, foreground status, background status, network status, application unresponsive status, application crash status, and application startup time.

That is, in this embodiment, the model input data is (n+1)*M dimensional data, and the output data is (n+1) dimensional data.

For example, as shown in Table 2, the output data of the model includes the corresponding killing probability of each application program and the available memory of the system.

Table 2

available memory Probability of process 1 being killed Probability of process 2 being killed Probability of process 3 being killed …

Step 230: Perform background checking and killing management on at least one application program according to the first checking and killing probability respectively corresponding to the at least one application program.

Optionally, the first terminal checks and kills the processes of the application programs whose first checking and killing probability is higher than 90%. Alternatively, the first terminal checks and kills the processes of the application programs whose first checking and killing probability is higher than the probability threshold and for which background protection is not set. The background protection is set by the user, and the application program with the background protection will not be checked and killed for the background process.

Alternatively, when the memory usage rate of the first terminal is higher than the threshold, the first terminal checks and kills processes of applications whose first checking probability is higher than the probability threshold and for which no background protection is set.

Optionally, the first terminal periodically executes steps 210 to 230 to perform periodic background checking and killing management.

To sum up, in the application management method provided by this embodiment, the terminal predicts the application programs that the user does not need to use according to the current running status and user interest information, and then performs scanning and killing management on these application programs. The terminal can use the neural network model (antivirus prediction model) to automatically perform application antivirus, and the antivirus prediction results can fit the user's usage habits, ensuring efficient and accurate background antivirus.

Exemplarily, an embodiment of training a killing prediction model is given.

Please refer to FIG. 3 . FIG. 3 is a flowchart of an application management method provided by another exemplary embodiment of the present application. The application management method may be applied to a first terminal, and the first terminal may be the terminal shown in FIG. 1 above.

In Figure 3, the method includes:

Step 310, receiving the general killing prediction model sent by the server. The general killing prediction model is trained according to the general sample data collected from at least two terminals, and the sample user interest information in the general sample data is uniformly set as a default value .

Exemplarily, the killing prediction model stored in the first terminal may be obtained after secondary training based on the general killing prediction model provided by the server. During the use of the first terminal, the anti-virus prediction model can be continuously automatically trained and updated according to the data in the user's use process. Finally, the anti-virus detection model running on each terminal can fit the usage habits of different users. According to the user's usage habits, the application scanning and killing management is carried out.

The first terminal may send a model acquisition request to the server, so that the server issues a general detection and killing prediction model for the first terminal. The general killing prediction model is a common killing prediction model for all users obtained by the server using large-scale data training.

Exemplarily, the first terminal obtains the anti-virus prediction model after training a general anti-virus detection model according to the terminal operating status of the first terminal.

In an optional embodiment, the general antivirus prediction model can be trained by the following method: input the sample terminal operating status and sample user information in the general sample data into the general antivirus prediction model to obtain the antivirus probability and system state; according to the killing probability and the loss of the killing probability label, as well as the loss of the system state and the system state label, train the general killing prediction model until convergence. Among them, the general sample data includes sample terminal operating status, sample user interest information and first tag data collected from at least two terminals; the first tag data includes killing probability tags and system status tags; the general sample data is based on at least two The data generated by the application killing behavior of a terminal, the killing probability label is determined according to the application program that the application killing behavior kills, and the system status label includes the system status after the killing behavior is executed.

Exemplarily, the general sample data may be obtained through log data uploaded by each terminal. For example, the operating system of each terminal will periodically send log data to the server merchant, and the log data records the terminal running status, the terminal application killing behavior, and the terminal running status after the application killing behavior. Then the terminal running state can be used as a sample terminal running state, and the killing probability label can be obtained according to the killing behavior of the terminal application (the killing probability label of the killed application is 1, and the killing probability of the unkilled application is 1). The label is 0), and the running status of the terminal after the killing behavior is applied can be used as the system status label. In this way, a general sample data set can be obtained.

Alternatively, general sample data can also be obtained from log data actively fed back by users. For example, user public opinion feedback and log data collected by the logkit log tool. This kind of log data is manually triggered and recorded by the user, and its content will be more detailed, and often the user encounters a problem. When giving after-sales feedback, in order to cooperate with the engineer to analyze the problem, the captured log is more suitable for the user scenario and can cover to some boundaries and corner cases.

In an optional embodiment, the general scanning and killing prediction model may also be acquired by the server according to the capability information of the first terminal. The first terminal sends a model acquisition request to the server, and the model acquisition request carries capability information of the first terminal; receives a general antivirus prediction model sent by the server, and the general antivirus prediction model is obtained according to the capability information.

The capability information of the terminal may include: at least one of terminal model, processor model, memory size, and related parameters of the processor (number of cores, frequency, cache, architecture, manufacturing process, etc.).

Exemplarily, the general killing prediction model is trained by the following method: training the teacher model according to the general sample data; determining the model structure of the general killing prediction model according to the capability information; training the general killing prediction according to the general sample data and the teacher model model; where the strength of capability information is proportional to the structural complexity of the model structure.

The server may determine the capability score of the terminal according to the capability information of the terminal, and select the corresponding model structure of the universal killing prediction model according to the capability score. Alternatively, the server may divide different model structures for different terminal models and processor models, so that the model structure fits the terminal capabilities.

For example, the server stores a table corresponding to the capability information and the model structure, and the server determines the model structure corresponding to the capability information according to the table. For example, as shown in Table 3, the server may look up a table according to the terminal model and processor model of the terminal to determine its corresponding model structure.

table 3

Terminal model Processor model model structure pear 14 xx1 generation Model structure 1 HW13 xx2 generation Model structure 2 OV5 xx1 generation Model structure 1 pear 20pro xx3 generation Model Structure 3

For more capable terminal models and processor models, more complex model structures can be used to improve model capabilities and improve model prediction accuracy. For example, the feature extraction layer has more layers, or the convolution kernel has a larger dimension, or the fully connected layer has more layers. For terminal models and processor models with weaker capabilities, a relatively simple model structure can be adopted to reduce the amount of model calculations and reduce the memory usage of the model running on the first terminal.

Among them, the general sample data includes sample terminal operating status, sample user interest information and first tag data collected from at least two terminals; the first tag data includes killing probability tags and system status tags; the general sample data is based on at least two The data generated by the application killing behavior of a terminal, the killing probability label is determined according to the application program that the application killing behavior kills, and the system status label includes the system status after the killing behavior is executed. Then the method for training the teacher model can be: input the sample terminal operating status and sample user information in the general sample data into the teacher model to obtain the second killing probability and the second system state; according to the second killing probability and the killing probability The first loss for labels, the second system state and the second loss for system state labels, train the teacher model until convergence.

Exemplarily, the server trains the teacher model until convergence according to the sum of the first loss and the second loss. Wherein, the loss function used to calculate the first loss and the loss function used to calculate the second loss may be the same or different. For example, the loss function may use a mean square error loss function, a cross-entropy loss function, an L1 norm loss function, and the like.

The method for training the general detection and killing prediction model can be: input the sample terminal operating status and sample user information in the general sample data into the teacher model to obtain the third detection probability and the third system state; The running status and sample user information are input into the general killing prediction model to obtain the fourth killing probability and the fourth system state; according to the fourth killing probability and killing probability label, the third loss, the fourth system state and the system state label The fourth loss of the fourth killing probability and the fifth loss of the third killing probability, the sixth loss of the fourth system state and the third system state, and train the general killing prediction model until convergence.

Exemplarily, according to the sum of the third loss, the fourth loss, the fifth loss, and the sixth loss, the server trains the universal killing prediction model until convergence. Wherein, the loss functions for calculating the third loss, the fourth loss, the fifth loss, and the sixth loss may be the same or different. For example, the loss function may use a mean square error loss function, a cross-entropy loss function, an L1 norm loss function, and the like.

Exemplarily, since the general antivirus prediction model is trained using sample data provided by many terminals, the general antivirus prediction model does not need to consider the user interest of each user. Therefore, when training the general antivirus prediction model, the input data can be The user interest information in is set to a preset value, for example, the user interest information corresponding to each application is set to 1.

Step 320, training a general hunting prediction model according to the user sample data to obtain the hunting prediction model; the user sample data includes terminal operating status, user interest information and second tag data collected from the first terminal.

After the first terminal acquires the general anti-virus detection model from the server, it can use the general anti-virus detection model for background application management and at the same time update and train the general anti-virus detection model.

Exemplarily, the first terminal can train the model once based on the loss of the first system state and the expected system state output in the prediction result after each prediction using the general antivirus prediction model (or antivirus prediction model), to update the model in real time.

Exemplarily, since the label of detection and killing probability cannot be obtained when the model is updated on the user side, the first terminal can guide and train the detection and killing prediction model according to the preset expected system state.

Exemplarily, when the model is used on the user side, the user interest information can be determined according to the user's usage habits, the killing prediction model can predict the killing probability of the application according to the user interest information, and the killing prediction model can be trained to output The system state is close to the desired system state.

Optionally, the second label data includes the expected system status label, and the expected system status label is a preset value; then the first terminal can input the terminal operating status and user interest information into the general antivirus prediction model to obtain the first antivirus probability and the first system state; according to the seventh loss of the label of the first system state and the expected system state, train the general killing prediction model until convergence, and obtain the killing prediction model.

Wherein, the desired system status includes: at least one of processor status, memory usage status, foreground status, background status, network status, application unresponsive status, application crash status, and application startup time. Exemplarily, the information contained in the expected system state is consistent with the information type of the system state (first system state) output by the model.

The desired system state may be provided by the server to the first terminal. The server may determine the expected system state of the next day (the expected system state of each terminal is the same) according to the daily log data reported by each terminal, and send the expected system state to the first terminal.

Alternatively, the server may determine the expected system status of the first terminal on the next day (the expected system status of each terminal is different) according to the log data reported by the first terminal, and send the expected system status to the first terminal.

Alternatively, the server may determine the expected system state of the first terminal according to the capability information of the first terminal, and send the expected system state to the first terminal. For terminals with stronger capabilities, the server may set a better expected system state, and for terminals with weaker capabilities, the server may set a poorer expected system state.

To sum up, with the application management method provided in this embodiment, the terminal can obtain a general antivirus prediction model from the server, and continuously update the model according to the usage data during the use of the model, so that the model can fit the use of each terminal According to the usage of different terminals, the detection and killing probability prediction is carried out in a targeted manner. Improve the accuracy of killing background applications in the killing prediction model.

The following are device embodiments of the present application, which can be used to implement the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

Please refer to FIG. 4 , which is a structural block diagram of an application management device provided by an exemplary embodiment of the present application. The application management device can be implemented as all or a part of the first terminal through software, hardware or a combination of the two. The unit includes:

The acquiring module 401 is configured to acquire the terminal running state and user interest information of the first terminal, the terminal running state includes the system state of the terminal and the application state of the application program in the terminal, and the user interest information is used To indicate the user's usage habits of the application;

A predicting module 402, configured to input the terminal operating status and the user interest information into a killing prediction model to obtain first killing probabilities respectively corresponding to the at least one application;

The management module 403 is configured to perform background checking and killing management on the at least one application program according to the first checking and killing probability respectively corresponding to the at least one application program.

In an optional embodiment, the number of said application programs is n, and n is a positive integer;

The obtaining module 401 is configured to obtain the usage durations corresponding to the n application programs within a preset period of time;

The obtaining module 401 is configured to perform normalization processing on the n usage durations to obtain interest weights respectively corresponding to the n application programs;

The obtaining module 401 is configured to determine n pieces of the interest weights as the user interest information.

In an optional embodiment, the number of said application programs is n, and n is a positive integer;

The acquisition module 401 is configured to acquire the system state of the first terminal, the system state includes: at least one of the processor state, memory usage state, foreground state, and background state of the terminal, the foreground The state includes: at least one of screen state and foreground application state;

The acquiring module 401 is configured to acquire the application states respectively corresponding to the n application programs, and the application states include: at least one of process information, elapsed running time, and package name.

In an optional embodiment, the device also includes:

The receiving module 405 is configured to receive the general antivirus prediction model sent by the server, the general antivirus prediction model is trained according to the general sample data collected from at least two terminals, and the sample user interest in the general sample data is The information is uniformly set as the default value;

The training module 406 is configured to train the general antivirus prediction model according to user sample data to obtain the antivirus prediction model; the user sample data includes the terminal operating status and all information collected from the first terminal. The user interest information and the second tag data.

In an optional embodiment, the device also includes:

A sending module 404, configured to send a model acquisition request to the server, where the model acquisition request carries capability information of the first terminal;

The receiving module 405 is configured to receive the general hunting prediction model sent by the server, where the general hunting prediction model is obtained according to the capability information.

In an optional embodiment, the general killing prediction model is trained by the following method:

training a teacher model according to the general sample data;

determining the model structure of the general killing prediction model according to the capability information;

training the general killing prediction model according to the general sample data and the teacher model;

Wherein, the strength of the capability information is directly proportional to the structural complexity of the model structure.

In an optional embodiment, the general sample data includes sample terminal operating status, sample user interest information and first tag data collected from at least two terminals; the first tag data includes killing probability tags and a system status label; the general sample data is data generated according to the application killing behavior of the at least two terminals, and the killing probability label is determined according to the killing application program of the application killing behavior, so The system status label includes the system status after executing the killing behavior;

The training of the teacher model according to the general sample data includes:

Input the sample terminal operating status and the sample user information in the general sample data into the teacher model to obtain a second killing probability and a second system state;

According to the second killing probability and the first loss of the killing probability label, the second system state and the second loss of the system state label, train the teacher model until convergence.

In an optional embodiment, the general sample data includes sample terminal operating status, sample user interest information and first tag data collected from at least two terminals; the first tag data includes killing probability tags and a system status label; the general sample data is data generated according to the application killing behavior of the at least two terminals, and the killing probability label is determined according to the killing application program of the application killing behavior, so The system status label includes the system status after executing the killing behavior;

The training of the general killing prediction model according to the general sample data and the teacher model includes:

Input the sample terminal operating state and the sample user information in the general sample data into the teacher model to obtain the third killing probability and the third system state;

Inputting the sample terminal operating status and the sample user information in the general sample data into the general antivirus prediction model to obtain a fourth antivirus probability and a fourth system state;

According to the fourth detection probability and the third loss of the detection probability label, the fourth system state and the fourth loss of the system state label, the fourth detection probability and the third detection The fifth loss of killing probability and the sixth loss of the fourth system state and the third system state are used to train the general killing prediction model until convergence.

In an optional embodiment, the second label data includes an expected system state label, and the expected system state label is a preset value;

The training module 406 is configured to input the terminal operation state and the user interest information into the general antivirus prediction model to obtain the first antivirus probability and the first system state;

The training module 406 is configured to train the general hunting prediction model until convergence according to the seventh loss of the first system state and the expected system state label, so as to obtain the killing prediction model.

In an optional embodiment, the desired system state includes: at least one of a processor state, a memory usage state, a foreground state, a background state, an application not responding, and an application crashing.

It should be noted that: when the application management device provided in the above embodiment executes the application management method, the division of the above-mentioned functional modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional modules according to needs. , which divides the internal structure of the device into different functional modules to complete all or part of the functions described above. In addition, the application management device and the application management method embodiments provided in the above embodiments belong to the same idea, and the specific implementation process thereof is detailed in the method embodiments, and will not be repeated here.

The present application also provides a terminal, the terminal includes a processor and a memory, at least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to implement the application management method provided by the above method embodiments. It should be noted that the terminal may be the terminal provided in FIG. 5 .

Fig. 5 shows a structural block diagram of a terminal 500 provided by an exemplary embodiment of the present application. The terminal 500 may be: a smart phone, a tablet computer, an MP3 player (Moving Picture Experts Group Audio Layer III, moving picture experts compress standard audio layer 3), MP4 (Moving Picture Experts Group Audio Layer IV, moving picture experts compress standard audio layer 4) Player, laptop or desktop computer. The terminal 500 may also be called a user account device, a portable terminal, a laptop terminal, a desktop terminal, and other names.

Generally, the terminal 500 includes: a processor 501 and a memory 502 .

The processor 501 may include one or more processing cores, such as a 4-core processor, an 8-core processor, and the like. The processor 501 can adopt at least one hardware form in DSP (Digital Signal Processing, digital signal processing), FPGA (Field-Programmable Gate Array, field programmable gate array), PLA (Programmable Logic Array, programmable logic array) accomplish. The processor 501 may also include a main processor and a coprocessor, the main processor is a processor for processing data in the wake-up state, and is also called a CPU (Central Processing Unit, central processing unit); the coprocessor is used to Low-power processor for processing data in standby state. In some embodiments, the processor 501 may be integrated with a GPU (Graphics Processing Unit, application manager), and the GPU is used for rendering and drawing content that needs to be displayed on the display screen. In some embodiments, the processor 501 may further include an AI (Artificial Intelligence, artificial intelligence) processor, where the AI processor is configured to process computing operations related to machine learning.

Memory 502 may include one or more computer-readable storage media, which may be non-transitory. The memory 502 may also include high-speed random access memory and non-volatile memory, such as one or more magnetic disk storage devices and flash memory storage devices. In some embodiments, the non-transitory computer-readable storage medium in the memory 502 is used to store at least one instruction, and the at least one instruction is used to be executed by the processor 501 to implement the application management provided by the method embodiments in this application method.

In some embodiments, the terminal 500 may optionally further include: a peripheral device interface 503 and at least one peripheral device. The processor 501, the memory 502, and the peripheral device interface 503 may be connected through buses or signal lines. Each peripheral device can be connected to the peripheral device interface 503 through a bus, a signal line or a circuit board. Specifically, the peripheral device includes: at least one of a radio frequency circuit 504 , a display screen 505 , a camera assembly 506 , an audio circuit 507 , and a power supply 509 .

The peripheral device interface 503 may be used to connect at least one peripheral device related to I/O (Input/Output, input/output) to the processor 501 and the memory 502 . In some embodiments, the processor 501, memory 502 and peripheral device interface 503 are integrated on the same chip or circuit board; in some other embodiments, any one of the processor 501, memory 502 and peripheral device interface 503 or The two can be implemented on a separate chip or circuit board, which is not limited in this embodiment.

The radio frequency circuit 504 is configured to receive and transmit RF (Radio Frequency, radio frequency) signals, also called electromagnetic signals. The radio frequency circuit 504 communicates with the communication network and other communication devices through electromagnetic signals. The radio frequency circuit 504 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals into electrical signals. Exemplarily, the radio frequency circuit 504 includes: an antenna system, an RF transceiver, one or more amplifiers, a tuner, an oscillator, a digital signal processor, a codec chipset, a user account identity module card, and the like. The radio frequency circuit 504 can communicate with other terminals through at least one wireless communication protocol. The wireless communication protocol includes, but is not limited to: World Wide Web, Metropolitan Area Network, Intranet, various generations of mobile communication networks (2G, 3G, 4G and 5G), wireless local area network and/or WiFi (Wireless Fidelity, Wireless Fidelity) network. In some embodiments, the radio frequency circuit 504 may also include circuits related to NFC (Near Field Communication, short-range wireless communication), which is not limited in this application.

The display screen 505 is used to display a UI (User Interface, user interface). The UI can include graphics, text, icons, video, and any combination thereof. When the display screen 505 is a touch display screen, the display screen 505 also has the ability to collect touch signals on or above the surface of the display screen 505 . The touch signal can be input to the processor 501 as a control signal for processing. At this time, the display screen 505 can also be used to provide virtual buttons and/or virtual keyboards, also called soft buttons and/or soft keyboards. In some embodiments, there may be one display screen 505, which is set on the front panel of the terminal 500; in other embodiments, there may be at least two display screens 505, which are respectively arranged on different surfaces of the terminal 500 or in a folding design; In some other embodiments, the display screen 505 may be a flexible display screen, which is arranged on a curved surface or a folded surface of the terminal 500 . Even, the display screen 505 can also be set as a non-rectangular irregular figure, that is, a special-shaped screen. The display screen 505 may be made of LCD (Liquid Crystal Display, liquid crystal display), OLED (Organic Light-Emitting Diode, organic light-emitting diode) and other materials.

The camera assembly 506 is used to capture images or videos. Exemplarily, the camera component 506 includes a front camera and a rear camera. Usually, the front camera is set on the front panel of the terminal, and the rear camera is set on the back of the terminal. In some embodiments, there are at least two rear cameras, which are any one of the main camera, depth-of-field camera, wide-angle camera, and telephoto camera, so as to realize the fusion of the main camera and the depth-of-field camera to realize the background blur function. Combined with the wide-angle camera to realize panoramic shooting and VR (Virtual Reality, virtual reality) shooting functions or other fusion shooting functions. In some embodiments, camera assembly 506 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. Dual color temperature flash refers to the combination of warm light flash and cold light flash, which can be used for light compensation under different color temperatures.

Audio circuitry 507 may include a microphone and speakers. The microphone is used to collect sound waves of the user account and the environment, and convert the sound waves into electrical signals and input them to the processor 501 for processing, or input them to the radio frequency circuit 504 to realize voice communication. For the purpose of stereo sound collection or noise reduction, there may be multiple microphones, which are respectively arranged at different parts of the terminal 500 . The microphone can also be an array microphone or an omnidirectional collection microphone. The speaker is used to convert the electrical signal from the processor 501 or the radio frequency circuit 504 into sound waves. The loudspeaker can be a conventional membrane loudspeaker or a piezoelectric ceramic loudspeaker. When the speaker is a piezoelectric ceramic speaker, it is possible not only to convert electrical signals into sound waves audible to humans, but also to convert electrical signals into sound waves inaudible to humans for purposes such as distance measurement. In some embodiments, the audio circuit 507 may also include a headphone jack.

The power supply 509 is used to supply power to various components in the terminal 500 . Power source 509 may be AC, DC, disposable or rechargeable batteries. When the power source 509 includes a rechargeable battery, the rechargeable battery may be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery charged through a wired line, and a wireless rechargeable battery is a battery charged through a wireless coil. The rechargeable battery can also be used to support fast charging technology.

In some embodiments, the terminal 500 further includes one or more sensors 510 . The one or more sensors 510 include, but are not limited to: an acceleration sensor 511 , a gyro sensor 512 , a pressure sensor 513 , an optical sensor 515 and a proximity sensor 516 .

The acceleration sensor 511 can detect the acceleration on the three coordinate axes of the coordinate system established by the terminal 500 . For example, the acceleration sensor 511 can be used to detect the components of the acceleration of gravity on the three coordinate axes. The processor 501 may control the display screen 505 to display the user interface in a landscape view or a portrait view according to the gravitational acceleration signal collected by the acceleration sensor 511 . The acceleration sensor 511 can also be used for collecting motion data of games or user accounts.

The gyro sensor 512 can detect the body direction and rotation angle of the terminal 500 , and the gyro sensor 512 can cooperate with the acceleration sensor 511 to collect 3D actions of the user account on the terminal 500 . According to the data collected by the gyroscope sensor 512, the processor 501 can realize the following functions: motion sensing (such as changing the UI according to the tilt operation of the user account), image stabilization during shooting, game control and inertial navigation.

The pressure sensor 513 may be disposed on a side frame of the terminal 500 and/or a lower layer of the display screen 505 . When the pressure sensor 513 is set on the side frame of the terminal 500 , it can detect the grip signal of the user account on the terminal 500 , and the processor 501 can perform left and right hand recognition or shortcut operations according to the grip signal collected by the pressure sensor 513 . When the pressure sensor 513 is disposed on the lower layer of the display screen 505, the processor 501 operates the pressure on the display screen 505 according to the user account to control the operable controls on the UI interface. The operable controls include at least one of button controls, scroll bar controls, icon controls, and menu controls.

The optical sensor 515 is used to collect ambient light intensity. In one embodiment, the processor 501 may control the display brightness of the display screen 505 according to the ambient light intensity collected by the optical sensor 515 . Specifically, when the ambient light intensity is high, the display brightness of the display screen 505 is increased; when the ambient light intensity is low, the display brightness of the display screen 505 is decreased. In another embodiment, the processor 501 may also dynamically adjust shooting parameters of the camera assembly 506 according to the ambient light intensity collected by the optical sensor 515 .

The proximity sensor 516 , also called a distance sensor, is usually arranged on the front panel of the terminal 500 . The proximity sensor 516 is used to collect the distance between the user account and the front of the terminal 500 . In one embodiment, when the proximity sensor 516 detects that the distance between the user account and the front of the terminal 500 gradually decreases, the processor 501 controls the display screen 505 to switch from the bright screen state to the off screen state; when the proximity sensor 516 detects When the distance between the user account and the front of the terminal 500 gradually increases, the processor 501 controls the display screen 505 to switch from the off-screen state to the on-screen state.

Those skilled in the art can understand that the structure shown in FIG. 5 does not constitute a limitation on the terminal 500, and may include more or less components than shown in the figure, or combine certain components, or adopt different component arrangements.

The memory further includes one or more programs, the one or more programs are stored in the memory, and the one or more programs include the application management method provided by the embodiment of the present application.

The present application also provides a computer device, which includes: a processor and a memory, at least one instruction, at least one section of program, code set or instruction set are stored in the storage medium, and the at least one instruction, at least one section of program, code set Or the instruction set is loaded and executed by the processor to implement the application management methods provided by the foregoing method embodiments.

The present application also provides a computer-readable storage medium, which stores at least one instruction, at least one section of program, code set or instruction set, and the at least one instruction, at least one section of program, code set or instruction set is loaded by a processor And execute to realize the application management method provided by the above method embodiments.

The present application also provides a computer program product or computer program, where the computer program product or computer program includes computer instructions, and the computer instructions are stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the application management method provided in the above optional implementation manner.

The serial numbers of the above embodiments of the present application are for description only, and do not represent the advantages and disadvantages of the embodiments.

Those of ordinary skill in the art can understand that all or part of the steps for implementing the above embodiments can be completed by hardware, and can also be completed by instructing related hardware through a program. The program can be stored in a computer-readable storage medium. The above-mentioned The storage medium mentioned may be a read-only memory, a magnetic disk or an optical disk, and the like.

The above descriptions are only exemplary embodiments of the present application that can be realized, and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the Within the protection scope of this application.

It should be noted that before and during the process of collecting relevant user data, this application can display a prompt interface, pop-up window or output voice prompt information. The prompt interface, pop-up window or voice prompt information It is used to remind the user that the relevant data is currently being collected, so that the application only starts to execute the relevant steps of obtaining the relevant data of the user after obtaining the confirmation operation issued by the user on the prompt interface or pop-up window, otherwise (that is, the user is not obtained When confirming the prompt interface or pop-up window), the relevant steps of obtaining user-related data are completed, that is, no user-related data is obtained. In other words, all user data collected in this application is collected with the consent and authorization of the user, and the collection, use and processing of relevant user data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

Claims (13)

1. An application management method, wherein the method is applied to a first terminal, and the method comprises the following steps:

acquiring a terminal running state of the first terminal and user interest information, wherein the terminal running state comprises a system state of the terminal and an application state of an application program in the terminal, and the user interest information is used for indicating a user using habit of the application program;

inputting the running state of the terminal and the user interest information into a searching and killing prediction model to obtain first searching and killing probabilities respectively corresponding to the at least one application program;

and carrying out background searching and killing management on the at least one application program according to the first searching and killing probability respectively corresponding to the at least one application program.

2. The method of claim 1, wherein the number of applications is n, n being a positive integer; the obtaining the user interest information comprises the following steps:

acquiring the use time lengths respectively corresponding to the n application programs in a preset period;

normalizing the n using time slots to obtain interest weights respectively corresponding to the n application programs;

and determining n interest weights as the user interest information.

3. The method of claim 1, wherein the number of applications is n, n being a positive integer; the obtaining the terminal operation state of the first terminal includes:

acquiring a system state of the first terminal, wherein the system state comprises: at least one of a processor state, a memory use state, a foreground state, and a background state, wherein the foreground state comprises: at least one of screen state and foreground application state;

acquiring application states corresponding to the n application programs respectively, wherein the application states comprise: at least one of process information, run time, package name.

4. A method according to any one of claims 1 to 3, wherein the method further comprises:

receiving a universal searching and killing prediction model sent by a server, wherein the universal searching and killing prediction model is obtained by training according to universal sample data acquired from at least two terminals, and sample user interest information in the universal sample data is uniformly set as a default value;

training the general killing prediction model according to user sample data to obtain the killing prediction model; the user sample data comprises the terminal running state, the user interest information and second tag data acquired from the first terminal.

5. The method of claim 4, wherein receiving the generic look-up predictive model sent by the server comprises:

sending a model acquisition request to the server, wherein the model acquisition request carries the capability information of the first terminal;

and receiving the general killing prediction model sent by the server, wherein the general killing prediction model is acquired according to the capability information.

6. The method of claim 5, wherein the generic look-up predictive model is trained using the following method:

training a teacher model according to the universal sample data;

determining a model structure of the general killing prediction model according to the capability information;

training the universal searching and killing prediction model according to the universal sample data and the teacher model;

wherein, the strength of the capability information is proportional to the structural complexity of the model structure.

7. The method of claim 6, wherein the generic sample data comprises sample terminal operational status, sample user interest information, and first tag data collected from at least two terminals; the first tag data comprises a searching and killing probability tag and a system state tag; the universal sample data are data generated according to application killing behaviors of the at least two terminals, the killing probability labels are determined according to application programs of the application killing behaviors, and the system state labels comprise system states after the application killing behaviors are executed;

The training of the teacher model based on the universal sample data includes:

inputting the sample terminal running state and the sample user information in the universal sample data into the teacher model to obtain a second searching and killing probability and a second system state;

and training the teacher model according to the second killing probability, the first loss of the killing probability label, the second system state and the second loss of the system state label until convergence.

8. The method of claim 6, wherein the generic sample data comprises sample terminal operational status, sample user interest information, and first tag data collected from at least two terminals; the first tag data comprises a searching and killing probability tag and a system state tag; the universal sample data are data generated according to application killing behaviors of the at least two terminals, the killing probability labels are determined according to application programs of the application killing behaviors, and the system state labels comprise system states after the application killing behaviors are executed;

the training the universal challenge predictive model according to the universal sample data and the teacher model includes:

Inputting the sample terminal running state and the sample user information in the universal sample data into the teacher model to obtain a third killing probability and a third system state;

inputting the sample terminal running state and the sample user information in the general sample data into the general searching and killing prediction model to obtain a fourth searching and killing probability and a fourth system state;

and training the general killing prediction model according to the third loss of the fourth killing probability and the killing probability label, the fourth loss of the fourth system state and the system state label, the fifth loss of the fourth killing probability and the third killing probability, and the sixth loss of the fourth system state and the third system state until convergence.

9. The method of claim 4, wherein the second tag data comprises a desired system state tag, the desired system state tag being a preset value;

the training of the universal killing prediction model according to the user sample data to obtain the killing prediction model comprises the following steps:

inputting the terminal running state and the user interest information into the general killing prediction model to obtain the first killing probability and a first system state;

And training the general killing prediction model until convergence according to the seventh loss of the first system state and the expected system state label to obtain the killing prediction model.

10. The method of claim 9, wherein the desired system state comprises: at least one of processor state, memory usage state, foreground state, background state, application unresponsive condition, application flashing condition.

11. An application management apparatus, wherein the apparatus is configured to implement a first terminal, the apparatus comprising:

the terminal operation state comprises a system state of the terminal and an application state of an application program in the terminal, and the user interest information is used for indicating the use habit of a user of the application program;

the prediction module is used for inputting the running state of the terminal and the user interest information into a searching and killing prediction model to obtain first searching and killing probabilities respectively corresponding to the at least one application program;

and the management module is used for carrying out background searching and killing management on the at least one application program according to the first searching and killing probability respectively corresponding to the at least one application program.

12. A terminal comprising a processor, and a memory coupled to the processor, and program instructions stored on the memory, the processor implementing the application management method of any of claims 1 to 10 when executing the program instructions.

13. A computer readable storage medium having stored therein program instructions, which when executed by a processor implement the application management method of any of claims 1 to 10.