WO2019062369A1 - 应用管理方法、装置、存储介质及电子设备 - Google Patents
应用管理方法、装置、存储介质及电子设备 Download PDFInfo
- Publication number
- WO2019062369A1 WO2019062369A1 PCT/CN2018/100684 CN2018100684W WO2019062369A1 WO 2019062369 A1 WO2019062369 A1 WO 2019062369A1 CN 2018100684 W CN2018100684 W CN 2018100684W WO 2019062369 A1 WO2019062369 A1 WO 2019062369A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- application
- feature
- coefficient
- neural network
- self
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0608—Saving storage space on storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/088—Non-supervised learning, e.g. competitive learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/445—Program loading or initiating
- G06F9/44594—Unloading
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/0604—Improving or facilitating administration, e.g. storage management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0673—Single storage device
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
- G06F9/5016—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5011—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
- G06F9/5022—Mechanisms to release resources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Definitions
- the present application relates to the field of electronic device technologies, and in particular, to an application management method, apparatus, storage medium, and electronic device.
- the embodiment of the present application provides an application management method, device, storage medium, and electronic device, which can reduce the memory occupation of the electronic device and improve the running fluency of the electronic device.
- An embodiment of the present application provides an application management method, including:
- the self-organizing neural network model including a feature coefficient matrix of the application
- the embodiment of the present application further provides an application management apparatus, including:
- An acquiring module configured to collect multiple feature information of the application according to an operation record of the application
- a learning module configured to learn the plurality of feature information by a self-organizing neural network to obtain a self-organizing neural network model of the application, where the self-organizing neural network model includes a feature coefficient matrix of the application;
- a calculating module configured to calculate, according to current feature information of the application, a first feature coefficient of the application when the application enters a background operation
- a determining module configured to determine a second feature coefficient from the matrix of feature coefficients according to the first feature coefficient
- a determining module configured to determine, according to the second feature coefficient, whether the application can be cleaned up.
- the embodiment of the present application further provides a storage medium, where the computer program stores a computer program, and when the computer program runs on the computer, the computer performs the following steps:
- the self-organizing neural network model including a feature coefficient matrix of the application
- the embodiment of the present application further provides an electronic device, including a processor and a memory, where the computer stores a computer program, and the processor executes the following steps by calling the computer program stored in the memory:
- the self-organizing neural network model including a feature coefficient matrix of the application
- FIG. 1 is a schematic diagram of a system of an application management apparatus according to an embodiment of the present application.
- FIG. 2 is a schematic diagram of an application scenario of an application management method provided by an embodiment of the present application.
- FIG. 3 is a schematic flowchart of an application management method provided by an embodiment of the present application.
- FIG. 4 is another schematic flowchart of an application management method provided by an embodiment of the present application.
- FIG. 5 is still another schematic flowchart of an application management method provided by an embodiment of the present application.
- FIG. 6 is a schematic structural diagram of a self-organizing neural network in an embodiment of the present application.
- FIG. 7 is a schematic structural diagram of an application management apparatus according to an embodiment of the present application.
- FIG. 8 is another schematic structural diagram of an application management apparatus according to an embodiment of the present application.
- FIG. 9 is still another schematic structural diagram of an application management apparatus according to an embodiment of the present application.
- FIG. 10 is a schematic structural diagram of an electronic device according to an embodiment of the present application.
- FIG. 11 is another schematic structural diagram of an electronic device according to an embodiment of the present application.
- FIG. 1 is a schematic diagram of a system of an application management apparatus according to an embodiment of the present application.
- the application management device is mainly configured to: collect multiple feature information of the application according to an operation record of the application; learn the collected multiple features by using a self-organizing neural network model to obtain a self-organizing neural network model of the application
- the self-organizing neural network model includes a matrix of feature coefficients of the application; collecting current feature information of the application, and calculating a first feature coefficient of the application according to the collected current feature information;
- the feature coefficient determines a second feature coefficient from the matrix of feature coefficients; and then determines whether the application can be cleaned based on the second feature coefficient.
- FIG. 2 is a schematic diagram of an application scenario of an application management method according to an embodiment of the present application.
- the electronic device manages multiple applications running in the background. For example, the application runs in the background, application b, and application c.
- the electronic device determines whether the application a, the application b, and the application c can be cleaned, respectively. For example, if it is determined that the application a, the application c cannot be cleaned, and the application b can be cleaned up, the electronic device can clean up the application b and keep the application a and the application c continue to run in the background. Thus, the electronic device can close the application b to release the memory occupied by the application b.
- An embodiment of the present application provides an application management method, including:
- the self-organizing neural network model including a feature coefficient matrix of the application
- the feature coefficient matrix includes a plurality of feature coefficients
- the step of determining the second feature coefficient from the feature coefficient matrix according to the first feature coefficient comprises:
- the feature coefficient corresponding to the difference having the smallest absolute value is determined as the second feature coefficient.
- the step of determining, according to the second feature coefficient, whether the application can be cleaned includes:
- the step of acquiring a preset duration parameter comprises:
- the step of determining, according to the comparison result, whether the application can be cleaned includes:
- the determination result is that the application cannot be cleaned
- the determination result is that the application can be cleaned up.
- the step of learning the plurality of feature information by the ad hoc neural network to obtain the self-organizing neural network model of the application comprises:
- the self-organizing neural network model of the application is obtained according to the obtained plurality of characteristic coefficients.
- the calculation is performed according to the following formula:
- W(t) W(t-1)+ ⁇ (t-1)L(t-1)(I(t-1)-W(t-1))
- W(t) is the connection weight of the node at time t
- W(t) is also the calculated characteristic coefficient
- t represents the time when the calculation is performed
- W(t-1) is the connection of the node at time t-1.
- Weight, ⁇ (t-1) is the rate of change of the radius of the field at time t-1
- L(t-1) is the learning rate at time t-1
- I(t-1) is input to self-organized at time t-1 Characteristic information of input nodes in a neural network.
- FIG. 3 is a schematic flowchart diagram of an application management method according to an embodiment of the present application.
- the application management method can be applied to an electronic device.
- the electronic device can be a device such as a smartphone or a tablet.
- the application management method can include the following steps:
- S110 Collect multiple feature information of the application according to an operation record of the application.
- the application may be any application installed in the electronic device, such as a communication application, a multimedia application, a game application, a news application, or a shopping application, and the like.
- the plurality of feature information of the application may include feature information inherent to the application itself, such as an application type of the application; and the plurality of feature information of the application may also include features generated by the application during operation Information, such as the running time of the application in the background, the number of times to enter the background in a day, the last running time in the foreground, or the way to enter the background (for example, being switched by the start key (ie, HOME key), being switched by the return key, Or switching by other APPs, etc.; the plurality of feature information of the application may further include feature information generated during the operation of the electronic device, such as the timeout period of the electronic device (ie, the screen is off), the duration of the bright screen, and the remaining power. , network status, or state of charge, and more.
- the electronic device can record its own operating status. For example, the electronic device can record, by the log file, the time each time the electronic device is turned off, the time of the bright screen, the power information, the network status, the charging status, and the like. After the above application starts running, the electronic device can also record the running status of the application. For example, the log file in the electronic device can record the time when the application starts running, the time when the background enters the running, the switching mode into the background, and the like.
- the plurality of feature information of the application may be obtained according to the running record of the application.
- the application management request may be triggered by an instruction of the user, or may be an operation triggered by the electronic device.
- the timing can be set in the electronic device, and the electronic device triggers the application management operation by itself when the time reaches the set timing.
- the plurality of feature information is learned by a self-organizing neural network to obtain a self-organizing neural network model of the application.
- FIG. 6 is a schematic structural diagram of a self-organizing neural network in an embodiment of the present application.
- Self-organizing neural networks self-organize and adaptively change network parameters and structures by automatically finding the intrinsic laws and essential properties in the sample.
- the self-organizing neural network has an input layer and a calculation layer.
- the input layer and the calculation layer respectively include a plurality of nodes.
- the electronic device collects the plurality of feature information of the application
- the plurality of feature information of the application is input to the input layer of the self-organizing neural network.
- the self-organizing neural network learns the plurality of feature information to obtain a self-organizing neural network model of the application.
- the self-organizing neural network model includes a matrix of feature coefficients of the application, and the matrix of feature coefficients includes a plurality of feature coefficients.
- step S120 learning the plurality of feature information by using a self-organizing neural network to obtain the self-organizing neural network model of the application, comprising the following steps:
- the node with the best matching with the node (input node) in the computing layer is determined according to the following formula 1 (computing node) ):
- DistFromInput represents the distance between the node (computing node) with the best matching of the node (input node) in the input layer to the input node, and I represents the node input into the input layer.
- Characteristic information of (input node) W represents the connection weight of all nodes in the domain radius, and n represents the number of feature information input to the node (input node).
- connection weight W of all nodes in the radius of the domain is calculated according to the following formula 2:
- W(t) is the connection weight calculated at time t, and is also the calculated characteristic coefficient
- W(t-1) is the connection weight of the node at time (t-1)
- ⁇ (t-1) is the rate of change of the radius of the field at (t-1)
- L(t-1) is the learning rate at time (t-1)
- I(t-1) is the time of (t-1) Characteristic information input to an input node in a self-organizing neural network.
- ⁇ (t-1) represents the radius of the field at time (t-1), and ⁇ (t-1) is calculated according to the following formula 5:
- ⁇ 0 is the initial domain radius and ⁇ 0 is a fixed constant, for example ⁇ 0 can take values 5, 10, and so on.
- ⁇ (t-1) is the iteration coefficient at time (t-1), and ⁇ (t-1) is calculated according to the following formula 6:
- ⁇ 0 is the initial iteration coefficient and ⁇ 0 is a fixed constant, for example ⁇ 0 may take a value of 1000.
- the electronic device After collecting the plurality of feature information of the application, the electronic device performs an iterative operation according to Equations 1 to 6, to obtain a plurality of connection weights W(t) of the application.
- the connection weight W(t) is also the characteristic coefficient.
- the calculated plurality of connection weights W(t) constitute a matrix of feature coefficients.
- the feature coefficient matrix includes a plurality of feature coefficients.
- the characteristic coefficient matrix and the correspondence between the plurality of feature coefficients and the time in the feature coefficient matrix constitute a self-organizing neural network model of the application.
- the electronic device when constructing the self-organizing neural network model of the application, the electronic device can perform a large number of operations according to the above formula.
- the electronic device can perform a judgment operation on whether the application can be cleaned up.
- the electronic device collects current feature information of the application.
- the current feature information may include feature information inherent to the application itself, such as an application type of the application; the current feature information may also include feature information generated by the application during operation, for example, the application is in the background.
- the running time the number of times to enter the background in a day, the last running time in the foreground, or the way to enter the background (such as being switched by the start key (ie HOME key), being switched by the return key, or being switched by other APP), etc.
- the current feature information may also include feature information generated during the operation of the electronic device, such as the timeout period of the electronic device (ie, the screen is off), the duration of the bright screen, the remaining power, the network status, or the charging status, and the like.
- the electronic device calculates current feature information of the application according to Equations 1 to 6 above to obtain a first feature coefficient of the application.
- the second feature coefficient of the application may be determined according to the first feature coefficient and the applied self-organizing neural network model.
- the self-organizing neural network model of the application includes a matrix of feature coefficients of the application.
- the feature coefficient matrix includes a plurality of feature coefficients.
- the electronic device may respectively compare the first feature coefficient with a plurality of feature coefficients in the feature coefficient matrix to determine a second feature coefficient from a plurality of feature coefficients in the feature coefficient matrix.
- first and second are used only to distinguish similar objects, and are not necessarily used to describe a specific order or order. It should be understood that the objects so described are interchangeable where appropriate.
- step S140 determining a second feature coefficient from the feature coefficient matrix according to the first feature coefficient, comprising the following steps:
- the electronic device calculates the first feature coefficient of the application, respectively comparing the first feature coefficient with the plurality of feature coefficients in the feature coefficient matrix, and calculating the first feature coefficient and the The difference between the plurality of feature coefficients and the absolute value of the difference is calculated.
- the smallest one is determined from the absolute values of the plurality of differences. Subsequently, the feature coefficient corresponding to the difference whose absolute value is the smallest is determined as the second feature coefficient.
- the second feature coefficient may be used to determine whether the application can be cleaned. If the application cannot be cleaned, then the application is kept running in the background, allowing the application's processes to reside in the system. If the application can be cleaned, the process of the application may be ended to close the application and release the memory occupied by the application.
- step S150 determining whether the application can be cleaned according to the second feature coefficient comprises the following steps:
- the preset duration parameter may be a duration value preset in the electronic device.
- the preset duration parameter can be 10 minutes.
- the electronic device calculates the second characteristic coefficient of the application
- the preset duration parameter is retrieved from the electronic device.
- the second characteristic coefficient is compared with the preset duration parameter to obtain a comparison result.
- the comparison result is a size relationship between the second feature coefficient and the preset duration parameter.
- the comparison result includes that the second feature coefficient is less than or equal to the preset duration parameter, or the second feature coefficient is greater than the preset duration parameter.
- the electronic device determines whether the application can be cleaned according to the comparison result.
- step S151 acquiring a preset duration parameter, including the following steps:
- S1511 Determine, according to the second feature coefficient, a time corresponding to the second feature coefficient
- S1512 Acquire a preset duration parameter according to the time period in which the time is located and the preset correspondence relationship, where the preset correspondence relationship is a correspondence between the time segment and the preset duration parameter.
- the second characteristic coefficient is a characteristic coefficient determined from the applied self-organizing neural network model.
- the second characteristic coefficient is a function of time t.
- the electronic device may determine a time corresponding to the second feature coefficient according to the second characteristic coefficient.
- the correspondence between the time period and the preset duration parameter can be preset in the electronic device. For example, 24 hours a day can be divided into 4 time periods: 0:00 to 6:00, 6:00 to 12:00, 12:00 to 18:00, and 18:00 to 24:00.
- the corresponding preset duration parameter can be set for each time period. For example, the preset duration parameter corresponding to 0:00 to 6:00 is 8 minutes, the preset duration parameter corresponding to 6:00 to 12:00 is 10 minutes, and the preset duration parameter corresponding to 12:00 to 18:00 is 15 minutes, the preset duration parameter corresponding to 18:00 ⁇ 24:00 is 20 minutes.
- the electronic device After determining the time corresponding to the second characteristic coefficient, the electronic device further determines the time period in which the time is. Then, the preset duration parameter is obtained according to the correspondence between the time period and the time period and the preset duration parameter.
- step S153 determining, according to the comparison result, whether the application can be cleaned includes the following steps:
- the electronic device After the electronic device obtains a comparison result between the second characteristic coefficient and the preset duration parameter, it is determined whether the application can be cleaned according to the comparison result.
- the second feature coefficient when the second feature coefficient is less than or equal to the preset duration parameter, it indicates that the application may be switched to the foreground operation again during the time period of the preset duration parameter, and the determination result is The application cannot be cleaned up.
- the second feature coefficient is less than the preset duration parameter of 10 minutes, indicating that the application may be switched to the foreground operation by the user again within the next 10 minutes, at which time the application may not be cleaned up.
- the second feature coefficient is greater than the preset duration parameter, it indicates that the application is not switched to the foreground operation again during the time period of the preset duration parameter, that is, the user does not use the application again. At this time, the judgment result is that the application can be cleaned up.
- the second feature coefficient is greater than the preset duration parameter by 15 minutes, indicating that the application will not be run again by the user for the next 15 minutes, at which time the application can be cleaned up. The electronic device can then close the application.
- the application management method collects multiple feature information of the application according to the running record of the application; the plurality of feature information is learned by the self-organizing neural network to obtain the application. a self-organizing neural network model; when the application enters the background operation, calculating a first feature coefficient of the application according to current feature information of the application; and determining a second feature from the feature coefficient matrix according to the first feature coefficient a coefficient; determining whether the application can be cleaned according to the second characteristic coefficient.
- the solution determines whether the application can be cleaned according to the second characteristic coefficient of the application, so that the application can be accurately determined according to the actual running condition of the application, when the application can be cleaned,
- the application is closed in time to release the memory of the electronic device, thereby reducing the memory usage of the electronic device and improving the smooth running of the electronic device.
- the embodiment of the present application further provides an application management device, which may be integrated into an electronic device, and the electronic device may be a device such as a smart phone or a tablet computer.
- An embodiment of the present application provides an application management apparatus, including:
- An acquiring module configured to collect multiple feature information of the application according to an operation record of the application
- a learning module configured to learn the plurality of feature information by a self-organizing neural network to obtain a self-organizing neural network model of the application, where the self-organizing neural network model includes a feature coefficient matrix of the application;
- a calculating module configured to calculate, according to current feature information of the application, a first feature coefficient of the application when the application enters a background operation
- a determining module configured to determine a second feature coefficient from the matrix of feature coefficients according to the first feature coefficient
- a determining module configured to determine, according to the second feature coefficient, whether the application can be cleaned up.
- the application management apparatus 200 may include: an acquisition module 201, a learning module 202, a calculation module 203, a determination module 204, and a determination module 205.
- the collecting module 201 is configured to collect multiple feature information of the application according to an operation record of the application.
- the application may be any application installed in the electronic device, such as a communication application, a multimedia application, a game application, a news application, or a shopping application.
- the plurality of feature information of the application may include feature information inherent to the application itself, such as an application type of the application; and the plurality of feature information of the application may also include features generated by the application during operation Information, such as the running time of the application in the background, the number of times to enter the background in a day, the last running time in the foreground, or the way to enter the background (for example, being switched by the start key (ie, HOME key), being switched by the return key, Or switching by other APPs, etc.; the plurality of feature information of the application may further include feature information generated during the operation of the electronic device, such as the timeout period of the electronic device (ie, the screen is off), the duration of the bright screen, and the remaining power. , network status, or state of charge, and more.
- the application management device 200 can record the operating status of the electronic device during the operation of the electronic device.
- the log file can be used to record the time each time the electronic device is turned off, the time of the bright screen, the power information, the network status, the charging status, and the like.
- the application management device 200 can also record the running status of the application.
- the log file in the electronic device can record the time when the application starts running, the time when the background enters the running, the switching mode into the background, and the like.
- the collection module 201 may acquire a plurality of feature information of the application according to the running record of the application.
- the application management request may be triggered by an instruction of the user, or may be an operation triggered by the electronic device.
- the timing can be set in the electronic device, and the electronic device triggers the application management operation by itself when the time reaches the set timing.
- the learning module 202 is configured to learn the plurality of feature information by using a self-organizing neural network to obtain the self-organizing neural network model of the application.
- the self-organizing neural network automatically changes the network parameters and structure by automatically searching for the inherent laws and essential attributes in the sample.
- Self-organizing neural networks have an input layer and a computing layer.
- the input layer and the calculation layer respectively include a plurality of nodes.
- the learning module 202 inputs the plurality of feature information of the application to the input layer of the self-organizing neural network. Subsequently, the self-organizing neural network learns the plurality of feature information to obtain a self-organizing neural network model of the application.
- the self-organizing neural network model includes a matrix of feature coefficients of the application, and the matrix of feature coefficients includes a plurality of feature coefficients.
- the learning module 202 is configured to perform the following steps:
- the self-organizing neural network model of the application is obtained according to the obtained plurality of characteristic coefficients.
- the learning module 202 determines the node with the best matching degree with the node (input node) in the computing layer according to the following formula 7. (calculate node):
- DistFromInput represents the distance between the node (computing node) with the best matching of the node (input node) in the input layer to the input node, and I represents the node input into the input layer.
- Characteristic information of (input node) W represents the connection weight of all nodes in the domain radius, and n represents the number of feature information input to the node (input node).
- connection weight W of all nodes in the radius of the domain is calculated according to the following formula:
- W(t) is the connection weight calculated at time t, and is also the calculated characteristic coefficient
- W(t-1) is the connection weight of the node at time (t-1)
- ⁇ (t-1) is the rate of change of the radius of the field at (t-1)
- L(t-1) is the learning rate at time (t-1)
- I(t-1) is the time of (t-1) Characteristic information input to an input node in a self-organizing neural network.
- ⁇ (t-1) represents the radius of the field at time (t-1), and ⁇ (t-1) is calculated according to the following formula XI:
- ⁇ 0 is the initial domain radius and ⁇ 0 is a fixed constant, for example ⁇ 0 can take values 5, 10, and so on.
- ⁇ (t-1) is the iteration coefficient at time (t-1), and ⁇ (t-1) is calculated according to the following formula:
- ⁇ 0 is the initial iteration coefficient and ⁇ 0 is a fixed constant, for example ⁇ 0 may take a value of 1000.
- the learning module 202 After the collection module 201 collects the plurality of feature information of the application, the learning module 202 performs an iterative operation according to Equations 7 to 12 to obtain a plurality of connection weights W(t) of the application.
- the connection weight W(t) is also the characteristic coefficient.
- the calculated plurality of connection weights W(t) constitute a matrix of feature coefficients.
- the feature coefficient matrix includes a plurality of feature coefficients.
- the characteristic coefficient matrix and the correspondence between the plurality of feature coefficients and the time in the feature coefficient matrix constitute a self-organizing neural network model of the application.
- the learning module 202 can perform a large number of operations according to the above formula.
- the calculating module 203 is configured to calculate a first feature coefficient of the application according to current feature information of the application when the application enters the background operation.
- the application management device 200 can perform a determination operation on whether the application can be cleaned up.
- the calculation module 203 collects current feature information of the application.
- the current feature information may include feature information inherent to the application itself, such as an application type of the application; the current feature information may also include feature information generated by the application during operation, for example, the application is in the background.
- the running time, the number of times to enter the background in a day, the last running time in the foreground, or the way to enter the background such as being switched by the start key (ie HOME key), being switched by the return key, or being switched by other APP), etc.
- the current feature information may also include feature information generated during the operation of the electronic device, such as the timeout period of the electronic device (ie, the screen is off), the duration of the bright screen, the remaining power, the network status, or the charging status, and the like.
- the calculation module 203 calculates the current feature information of the application according to the above formulas 7 to 12 to obtain the first feature coefficient of the application.
- the determining module 204 is configured to determine, according to the first feature coefficient, a second feature coefficient from the matrix of feature coefficients.
- the determining module 204 may determine the second feature coefficient of the application according to the first feature coefficient and the applied self-organizing neural network model.
- the self-organizing neural network model of the application includes a matrix of feature coefficients of the application.
- the feature coefficient matrix includes a plurality of feature coefficients.
- the determining module 204 may respectively compare the first feature coefficient with a plurality of feature coefficients in the feature coefficient matrix to determine a second feature coefficient from a plurality of feature coefficients in the feature coefficient matrix.
- the determination module 204 includes a calculation sub-module 2041 and a determination sub-module 2042.
- a calculation submodule 2041 configured to separately calculate an absolute value of a difference between the first characteristic coefficient and the plurality of feature coefficients
- the determining sub-module 2042 is configured to determine a feature coefficient corresponding to the difference with the smallest absolute value as the second feature coefficient.
- the calculation sub-module 2041 compares the first feature coefficient with the plurality of feature coefficients in the feature coefficient matrix, respectively, to calculate the first A difference between the characteristic coefficient and the plurality of characteristic coefficients, and calculating an absolute value of the difference.
- the determining sub-module 2042 determines the smallest one from the absolute values of the plurality of differences. Subsequently, the feature coefficient corresponding to the difference whose absolute value is the smallest is determined as the second feature coefficient.
- the determining module 205 is configured to determine, according to the second feature coefficient, whether the application can be cleaned.
- the determining module 205 may perform a determination according to the second feature coefficient to determine whether the application can be cleaned. If the application cannot be cleaned, then the application is kept running in the background, allowing the application's processes to reside in the system. If the application can be cleaned, the process of the application may be ended to close the application and release the memory occupied by the application.
- the determining module 205 includes: an obtaining submodule 2051, a comparing submodule 2052, and a judging submodule 2053.
- the obtaining submodule 2051 is configured to obtain a preset duration parameter.
- a comparison submodule 2052 configured to compare the second feature coefficient with the preset duration parameter to obtain a comparison result
- the determining sub-module 2053 is configured to determine, according to the comparison result, whether the application can be cleaned up.
- the preset duration parameter may be a duration value preset in the electronic device.
- the preset duration parameter can be 10 minutes.
- the obtaining submodule 2051 retrieves the preset duration parameter from the electronic device.
- the comparison sub-module 2052 compares the second feature coefficient with the preset duration parameter to obtain a comparison result.
- the comparison result is a size relationship between the second feature coefficient and the preset duration parameter.
- the comparison result includes that the second feature coefficient is less than or equal to the preset duration parameter, or the second feature coefficient is greater than the preset duration parameter.
- the judging sub-module 2053 judges whether the application can be cleaned according to the comparison result.
- the acquisition sub-module 2051 is configured to perform the following steps:
- the second characteristic coefficient is a characteristic coefficient determined from the applied self-organizing neural network model.
- the second characteristic coefficient is a function of time t.
- the obtaining submodule 2051 may determine a time corresponding to the second feature coefficient according to the second feature coefficient.
- the correspondence between the time period and the preset duration parameter can be preset in the electronic device. For example, 24 hours a day can be divided into 4 time periods: 0:00 to 6:00, 6:00 to 12:00, 12:00 to 18:00, and 18:00 to 24:00.
- the corresponding preset duration parameter can be set for each time period. For example, the preset duration parameter corresponding to 0:00 to 6:00 is 8 minutes, the preset duration parameter corresponding to 6:00 to 12:00 is 10 minutes, and the preset duration parameter corresponding to 12:00 to 18:00 is 15 minutes, the preset duration parameter corresponding to 18:00 ⁇ 24:00 is 20 minutes.
- the obtaining sub-module 2051 determines the time corresponding to the second feature coefficient
- the time period in which the time is located is further determined.
- the preset duration parameter is obtained according to the correspondence between the time period and the time period and the preset duration parameter.
- the determining sub-module 2053 is configured to perform the following steps:
- the determination result is that the application cannot be cleaned
- the determination result is that the application can be cleaned up.
- the determining sub-module 2053 determines whether the application can be cleaned according to the comparison result.
- the second feature coefficient when the second feature coefficient is less than or equal to the preset duration parameter, it indicates that the application may be switched to the foreground operation again during the time period of the preset duration parameter, and the determination result is The application cannot be cleaned up.
- the second feature coefficient is less than the preset duration parameter of 10 minutes, indicating that the application may be switched to the foreground operation by the user again within the next 10 minutes, at which time the application may not be cleaned up.
- the second feature coefficient is greater than the preset duration parameter, it indicates that the application is not switched to the foreground operation again during the time period of the preset duration parameter, that is, the user does not use the application again. At this time, the judgment result is that the application can be cleaned up.
- the second feature coefficient is greater than the preset duration parameter by 15 minutes, indicating that the application will not be run again by the user for the next 15 minutes, at which time the application can be cleaned up. Subsequently, the application management device 200 can close the application.
- each of the above modules may be implemented as a separate entity, or may be implemented in any combination as one or several entities.
- the application management device 200 collects multiple feature information of the application according to the running record of the application by the collecting module 201.
- the learning module 202 performs the plurality of feature information through the self-organizing neural network. Learning to obtain a self-organizing neural network model of the application; the computing module 203, when the application enters the background operation, calculating a first feature coefficient of the application according to current feature information of the application; the determining module 204 is configured according to the The first feature coefficient determines a second feature coefficient from the feature coefficient matrix; the determining module 205 determines whether the application can be cleaned according to the second feature coefficient.
- the solution determines whether the application can be cleaned according to the second characteristic coefficient of the application, so that the application can be accurately determined according to the actual running condition of the application, when the application can be cleaned,
- the application is closed in time to release the memory of the electronic device, thereby reducing the memory usage of the electronic device and improving the smooth running of the electronic device.
- An embodiment of the present application further provides an electronic device.
- the electronic device can be a device such as a smartphone or a tablet.
- the electronic device 300 includes a processor 301 and a memory 302.
- the processor 301 is electrically connected to the memory 302.
- the processor 301 is a control center of the electronic device 300, and connects various parts of the entire electronic device using various interfaces and lines, and executes the electronic by running or calling a computer program stored in the memory 302 and calling data stored in the memory 302.
- the various functions and processing data of the device enable overall monitoring of the electronic device.
- the processor 301 in the electronic device 300 loads the instructions corresponding to the processes of one or more computer programs into the memory 302 according to the following steps, and is executed by the processor 301 to be stored in the memory 302.
- the self-organizing neural network model including a feature coefficient matrix of the application
- the feature coefficient matrix includes a plurality of feature coefficients, and when determining the second feature coefficient from the feature coefficient matrix according to the first feature coefficient, the processor 301 performs the following steps:
- the feature coefficient corresponding to the difference having the smallest absolute value is determined as the second feature coefficient.
- the processor 301 when determining whether the application can be cleaned according to the second characteristic coefficient, performs the following steps:
- the processor 301 when the preset duration parameter is obtained, the processor 301 performs the following steps:
- the processor 301 when determining whether the application can be cleaned according to the comparison result, performs the following steps:
- the determination result is that the application cannot be cleaned
- the determination result is that the application can be cleaned up.
- the processor 301 when the plurality of feature information is learned by a self-organizing neural network to obtain the self-organizing neural network model of the application, the processor 301 performs the following steps:
- the self-organizing neural network model of the application is obtained according to the obtained plurality of characteristic coefficients.
- the processor 301 calculates according to the following formula when calculating a plurality of feature coefficients of the application according to the collected plurality of feature information and the connection weight of each node in the self-organizing neural network:
- W(t) W(t-1)+ ⁇ (t-1)L(t-1)(I(t-1)-W(t-1))
- W(t) is the connection weight of the node at time t
- W(t) is also the characteristic coefficient calculated at time t
- t is the time when the calculation is performed
- W(t-1) is the node at time t-1.
- the connection weight, ⁇ (t-1) is the rate of change of the radius of the field at (t-1)
- L(t-1) is the learning rate at time (t-1)
- I(t-1) is (t -1) Time information input to the input node in the ad hoc neural network.
- Memory 302 can be used to store computer programs and data.
- the computer program stored in the memory 302 contains instructions executable in the processor.
- Computer programs can be combined into various functional modules.
- the processor 301 executes various functional applications and data processing by calling a computer program stored in the memory 302.
- the electronic device 300 further includes: a radio frequency circuit 303, a display screen 304, a control circuit 305, an input unit 306, an audio circuit 307, a sensor 308, and a power source 309.
- the processor 301 is electrically connected to the radio frequency circuit 303, the display screen 304, the control circuit 305, the input unit 306, the audio circuit 307, the sensor 308, and the power source 309, respectively.
- the radio frequency circuit 303 is configured to transceive radio frequency signals to communicate with network devices or other electronic devices through wireless communication.
- the display screen 304 can be used to display information entered by the user or information provided to the user as well as various graphical user interfaces of the electronic device, which can be composed of images, text, icons, video, and any combination thereof.
- the control circuit 305 is electrically connected to the display screen 304 for controlling the display screen 304 to display information.
- the input unit 306 can be configured to receive input digits, character information, or user characteristic information (eg, fingerprints), and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function controls.
- the input unit 306 can include a fingerprint identification module.
- the audio circuit 307 can provide an audio interface between the user and the electronic device through a speaker and a microphone.
- Sensor 308 is used to collect external environmental information.
- Sensor 308 can include one or more of ambient brightness sensors, acceleration sensors, gyroscopes, and the like.
- Power source 309 is used to power various components of electronic device 300.
- the power source 309 can be logically coupled to the processor 301 through a power management system to enable functions such as managing charging, discharging, and power management through the power management system.
- the electronic device 300 may further include a camera, a Bluetooth module, and the like, and details are not described herein again.
- an embodiment of the present application provides an electronic device, which collects multiple feature information of the application according to an operation record of the application; and learns the multiple feature information by using a self-organizing neural network to obtain The self-organizing neural network model of the application; when the application enters the background operation, calculating a first feature coefficient of the application according to current feature information of the application; determining from the feature coefficient matrix according to the first feature coefficient And generating a second characteristic coefficient; determining, according to the second characteristic coefficient, whether the application can be cleaned.
- the solution determines whether the application can be cleaned according to the second characteristic coefficient of the application, so that the application can be accurately determined according to the actual running condition of the application, when the application can be cleaned,
- the application is closed in time to release the memory of the electronic device, thereby reducing the memory usage of the electronic device and improving the smooth running of the electronic device.
- the embodiment of the present application further provides a storage medium, where the computer program stores a computer program, and when the computer program runs on the computer, the computer executes the application management method described in any of the foregoing embodiments.
- the embodiment of the present application provides a storage medium, where the computer program stores a computer program, and when the computer program runs on a computer, the computer performs the following steps:
- the self-organizing neural network model including a feature coefficient matrix of the application
- the feature coefficient matrix includes a plurality of feature coefficients
- the step of determining the second feature coefficient from the feature coefficient matrix according to the first feature coefficient comprises:
- the feature coefficient corresponding to the difference having the smallest absolute value is determined as the second feature coefficient.
- the step of determining, according to the second feature coefficient, whether the application can be cleaned includes:
- the step of acquiring a preset duration parameter comprises:
- the step of determining, according to the comparison result, whether the application can be cleaned includes:
- the determination result is that the application cannot be cleaned
- the determination result is that the application can be cleaned up.
- the storage medium may include, but is not limited to, a read only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Human Computer Interaction (AREA)
- Biophysics (AREA)
- Molecular Biology (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Telephone Function (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
一种应用管理方法、装置、存储介质及电子设备,该方法包括:采集应用的多个特征信息;对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型;计算所述应用的第一特征系数;根据所述第一特征系数从特征系数矩阵中确定出第二特征系数;根据所述第二特征系数判断所述应用是否可被清理。
Description
本申请要求于2017年09月30日提交中国专利局、申请号为201710940338.9、发明名称为“应用管理方法、装置、存储介质及电子设备”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请涉及电子设备技术领域,特别涉及一种应用管理方法、装置、存储介质及电子设备。
随着电子技术的发展,诸如智能手机等电子设备的功能越来越丰富。用户通常在电子设备上安装有多个应用。当电子设备中打开多个应用时,若电子设备退回主屏幕或者停留在某一应用的应用界面,则电子设备中打开的多个应用依然会在电子设备的后台运行。
发明内容
本申请实施例提供一种应用管理方法、装置、存储介质及电子设备,可以减少电子设备的内存占用,提高电子设备的运行流畅度。
本申请实施例提供一种应用管理方法,包括:
根据应用的运行记录采集所述应用的多个特征信息;
通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;
当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;
根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;
根据所述第二特征系数判断所述应用是否可被清理。
本申请实施例还提供一种应用管理装置,包括:
采集模块,用于根据应用的运行记录采集所述应用的多个特征信息;
学习模块,用于通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;
计算模块,用于当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;
确定模块,用于根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;
判断模块,用于根据所述第二特征系数判断所述应用是否可被清理。
本申请实施例还提供一种存储介质,所述存储介质中存储有计算机程序,当所述计算机程序在计算机上运行时,使得所述计算机执行以下步骤:
根据应用的运行记录采集所述应用的多个特征信息;
通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;
当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;
根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;
根据所述第二特征系数判断所述应用是否可被清理。
本申请实施例还提供一种电子设备,包括处理器和存储器,所述存储器中存储有计算机程序,所述处理器通过调用所述存储器中存储的所述计算机程序,用于执行以下步骤:
根据应用的运行记录采集所述应用的多个特征信息;
通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;
当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;
根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;
根据所述第二特征系数判断所述应用是否可被清理。
为了更清楚地说明本申请实施例中的技术方案,下面将对实施例描述中所需要使用的附图作简单地介绍。显而易见地,下面描述中的附图仅仅是本申请的一些实施例,对于本领域技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的应用管理装置的***示意图。
图2是本申请实施例提供的应用管理方法的应用场景示意图。
图3是本申请实施例提供的应用管理方法的流程示意图。
图4是本申请实施例提供的应用管理方法的另一流程示意图。
图5是本申请实施例提供的应用管理方法的又一流程示意图。
图6是本申请实施例中的自组织神经网络的架构示意图。
图7是本申请实施例提供的应用管理装置的结构示意图。
图8是本申请实施例提供的应用管理装置的另一结构示意图。
图9是本申请实施例提供的应用管理装置的又一结构示意图。
图10是本申请实施例提供的电子设备的结构示意图。
图11是本申请实施例提供的电子设备的另一结构示意图。
下面将结合本申请实施例中的附图,对本申请实施例中的技术方案进行清楚、完整地描述。显然,所描述的实施例仅仅是本申请一部分实施例,而不是全部的实施例。基于本申请中的实施例,本领域技术人员在没有付出创造性劳动前提下所获得的所有其他实施例,都属于本申请的保护范围。
参考图1,图1为本申请实施例提供的应用管理装置的***示意图。该应用管理装置主要用于:根据应用的运行记录采集所述应用的多个特征信息;通过自组织神经网络模型对采集到的多个特征进行学习,以得到所述应用的自组织神经网络模型,其中所述自组织神经网络模型包括所述应用的特征系数矩阵;采集所述应用的当前特征信息,并根据采集到的当前特征信息计算所述应用的第一特征系数;根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;随后根据所述第二特征系数判断所述应用是否可被清理。
参考图2,图2为本申请实施例提供的应用管理方法的应用场景示意图。其中,电子设备对后台运行的多个应用进行管理。例如,后台运行有应用a、应用b以及应用c。电子设备分别判断应用a、应用b、应用c是否可被清理。例如,判断为应用a、应用c不可被清理,应用b可被清理,则电子设备可以清理掉应用b,保持应用a和应用c继续在后台运行。从而,电子设备可以关闭应用b,以释放应用b 所占用的内存。
本申请实施例提供一种应用管理方法,包括:
根据应用的运行记录采集所述应用的多个特征信息;
通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;
当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;
根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;
根据所述第二特征系数判断所述应用是否可被清理。
在一些实施例中,所述特征系数矩阵包括多个特征系数,所述根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数的步骤包括:
分别计算所述第一特征系数与所述多个特征系数的差值的绝对值;
将绝对值最小的差值对应的特征系数确定为第二特征系数。
在一些实施例中,所述根据所述第二特征系数判断所述应用是否可被清理的步骤包括:
获取预设时长参数;
将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果;
根据所述比较结果判断所述应用是否可被清理。
在一些实施例中,所述获取预设时长参数的步骤包括:
根据所述第二特征系数确定所述第二特征系数对应的时刻;
根据所述时刻所处的时间段以及预设对应关系获取预设时长参数,所述预设对应关系为时间段与预设时长参数之间的对应关系。
在一些实施例中,所述根据所述比较结果判断所述应用是否可被清理的步骤包括:
当所述第二特征系数小于或等于所述预设时长参数时,判断结果为所述应用不可被清理;
当所述第二特征系数大于所述预设时长参数时,判断结果为所述应用可被清理。
在一些实施例中,所述通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型的步骤包括:
根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数;
根据得到的多个特征系数获取所述应用的自组织神经网络模型。
在一些实施例中,根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数时,根据以下公式进行计算:
W(t)=W(t-1)+Θ(t-1)L(t-1)(I(t-1)-W(t-1))
其中,W(t)为t时刻节点的连接权值,并且W(t)也为所计算的特征系数,t表示进行计算时的时刻,W(t-1)为t-1时刻节点的连接权值,Θ(t-1)为t-1时刻的领域半径变化率,L(t-1)为t-1时刻的学习率,I(t-1)为t-1时刻输入到自组织神经网络中的输入节点的特征信息。
参考图3,图3为本申请实施例提供的应用管理方法的流程示意图。该应用管理方法可以应用于电子设备中。该电子设备可以是智能手机、平板电脑等设备。该应用管理方法可以包括以下步骤:
S110,根据应用的运行记录采集所述应用的多个特征信息。
其中,所述应用可以是安装在电子设备中的任意应用程序,例如通讯应用程序、多媒体应用程序、 游戏应用程序、资讯应用程序、或者购物应用程序等等。
所述应用的多个特征信息可以包括所述应用自身所固有的特征信息,例如所述应用的应用类型;所述应用的多个特征信息也可以包括所述应用在运行过程中所产生的特征信息,例如所述应用在后台的运行时长、一天中进入后台的次数、上一次在前台的运行时长、或者进入后台的方式(例如被起始键(即HOME键)切换、被返回键切换、或者被其他APP切换)等等;所述应用的多个特征信息还可以包括电子设备运行过程中所产生的特征信息,例如电子设备的熄屏(即灭屏)时长、亮屏时长、剩余电量、网络状态、或者充电状态等等。
在电子设备的运行过程中,电子设备可以对自身的运行状况进行记录。例如电子设备可以通过日志文件记录电子设备每次熄屏的时刻、亮屏的时刻、电量信息、网络状态、充电状态等等。当上述应用开始运行后,电子设备还可以对所述应用的运行状况进行记录。例如电子设备中的日志文件可以记录所述应用开始运行的时刻、进入后台运行的时刻、进入后台的切换方式等等。
当电子设备接收到应用管理请求时,可以根据所述应用的运行记录获取所述应用的多个特征信息。其中,所述应用管理请求可以是通过用户的指令所触发的,也可以是电子设备自行触发的操作。例如,电子设备中可以设置定时,每当时间到达设置的定时,电子设备即自行触发应用管理操作。
S120,通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型。
参考图6,图6为本申请实施例中的自组织神经网络的架构示意图。自组织神经网络是通过自动寻找样本中的内在规律和本质属性,自组织、自适应地改变网络参数与结构。其中,自组织神经网络具有输入层和计算层。输入层和计算层分别包括多个节点。
电子设备采集到应用的多个特征信息后,将所述应用的多个特征信息输入到自组织神经网络的输入层。随后,所述自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型。其中,所述自组织神经网络模型包括所述应用的特征系数矩阵,所述特征系数矩阵包括多个特征系数。
在一些实施例中,如图4所示,步骤S120、通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,包括以下步骤:
S121,根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数;
S122,根据得到的多个特征系数获取所述应用的自组织神经网络模型。
具体地,自组织神经网络的输入层中的某个节点(输入节点)获取到特征信息后,根据以下公式一确定计算层中与所述节点(输入节点)匹配性最好的节点(计算节点):
其中,DistFromInput表示计算层中与输入层中的所述节点(输入节点)匹配性最好的节点(计算节点)到所述输入节点之间的距离,I表示输入到输入层中的所述节点(输入节点)的特征信息,W表示领域半径内所有节点的连接权值,n表示输入到所述节点(输入节点)的特征信息的个数。
其中,领域半径内所有节点的连接权值W根据以下公式二计算:
公式二:W(t)=W(t-1)+Θ(t-1)L(t-1)(I(t-1)-W(t-1))
其中,t表示进行计算时的时刻,W(t)为t时刻所计算的连接权值,也为所计算的特征系数,W(t-1)为(t-1)时刻节点的连接权值,Θ(t-1)为(t-1)时刻的领域半径变化率,L(t-1)为(t-1)时刻的学习率,I(t-1)为(t-1)时刻输入到自组织神经网络中的输入节点的特征信息。分别根据以下公式三和公式四计算Θ(t-1)和L(t-1):
其中,e为自然常数(超越数),e为定值,e约等于2.71828。L
0为固定常数,例如L
0可以取值0.3、0.5等。σ(t-1)表示(t-1)时刻的领域半径,σ(t-1)根据以下公式五进行计算:
其中,σ
0为初始领域半径,σ
0为固定常数,例如σ
0可以取值5、10等等。λ(t-1)为(t-1)时刻的迭代系数,λ(t-1)根据以下公式六进行计算:
其中,λ
0为初始迭代系数,λ
0为固定常数,例如λ
0可以取值1000。
电子设备采集到上述应用的多个特征信息后,根据公式一至公式六进行迭代运算,得到所述应用的多个连接权值W(t)。连接权值W(t)也即为特征系数。最后,计算得到的多个连接权值W(t)构成特征系数矩阵。所述特征系数矩阵即包括多个特征系数。所述特征系数矩阵以及特征系数矩阵中的多个特征系数与时间之间的对应关系即构成所述应用的自组织神经网络模型。
需要说明的是,在构建应用的自组织神经网络模型时,电子设备可以根据以上公式进行大量次数的运算。
S130,当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数。
其中,当所述应用由前台运行进入后台运行时,电子设备即可对所述应用进行是否可被清理的判断操作。
具体地,当所述应用进入后台运行时,电子设备采集所述应用当前的特征信息。其中,当前特征信息可以包括所述应用自身所固有的特征信息,例如所述应用的应用类型;当前特征信息也可以包括所述应用在运行过程中所产生的特征信息,例如所述应用在后台的运行时长、一天中进入后台的次数、上一次在前台的运行时长、或者进入后台的方式(例如被起始键(即HOME键)切换、被返回键切换、或者被其他APP切换)等等;当前特征信息还可以包括电子设备运行过程中所产生的特征信息,例如电子设备的熄屏(即灭屏)时长、亮屏时长、剩余电量、网络状态、或者充电状态等等。
随后,电子设备根据上述公式一至公式六对所述应用的当前特征信息进行计算,以得到所述应用的第一特征系数。
S140,根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数。
其中,电子设备计算得到所述应用的第一特征系数后,即可根据所述第一特征系数以及所述应用的自组织神经网络模型确定出所述应用的第二特征系数。
具体地,所述应用的自组织神经网络模型包括所述应用的特征系数矩阵。所述特征系数矩阵中包括多个特征系数。电子设备可以分别将所述第一特征系数与所述特征系数矩阵中的多个特征系数进行比较,以从所述特征系数矩阵中的多个特征系数中确定出第二特征系数。
本申请实施例中,术语“第一”、“第二”只用于区别类似的对象,而不必用于描述特定的顺序或先后次序。应当理解,这样描述的对象在适当情况下可以互换。
在一些实施例中,如图4所示,步骤S140、根据所述第一特征系数从所述特征系数矩阵中确定出第 二特征系数,包括以下步骤:
S141,分别计算所述第一特征系数与所述多个特征系数的差值的绝对值;
S142,将绝对值最小的差值对应的特征系数确定为第二特征系数。
其中,电子设备计算得出所述应用的第一特征系数后,分别将所述第一特征系数与所述特征系数矩阵中的多个特征系数进行比较,计算所述第一特征系数与所述多个特征系数的差值,并计算所述差值的绝对值。
得到多个差值的绝对值后,从多个差值的绝对值中确定出最小的一个。随后,将绝对值最小的差值所对应的特征系数确定为第二特征系数。
S150,根据所述第二特征系数判断所述应用是否可被清理。
其中,电子设备计算得出所述应用的第二特征系数后,即可根据所述第二特征系数进行判断,以确定所述应用是否可被清理。若所述应用不可被清理,则保留所述应用继续在后台运行,允许所述应用的进程驻留在***中。若所述应用可以被清理,则可以结束所述应用的进程,以关闭所述应用,释放所述应用占用的内存。
在一些实施例中,如图4所示,步骤S150、根据所述第二特征系数判断所述应用是否可被清理,包括以下步骤:
S151,获取预设时长参数;
S152,将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果;
S153,根据所述比较结果判断所述应用是否可被清理。
其中,预设时长参数可以是预先设置在电子设备中的一个时长值。例如,预设时长参数可以为10分钟。电子设备计算得出所述应用的第二特征系数后,从所述电子设备中调取所述预设时长参数。随后,将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果。其中,比较结果为所述第二特征系数与所述预设时长参数之间的大小关系。比较结果包括所述第二特征系数小于或等于所述预设时长参数,或者所述第二特征系数大于所述预设时长参数。随后,电子设备根据所述比较结果判断所述应用是否可被清理。
在一些实施例中,如图5所示,步骤S151、获取预设时长参数,包括以下步骤:
S1511,根据所述第二特征系数确定所述第二特征系数对应的时刻;
S1512,根据所述时刻所处的时间段以及预设对应关系获取预设时长参数,所述预设对应关系为时间段与预设时长参数之间的对应关系。
其中,第二特征系数是从所述应用的自组织神经网络模型中确定出的特征系数。第二特征系数为关于时间t的函数。电子设备可以根据所述第二特征系数确定所述第二特征系数对应的时刻。
电子设备中可以预先设置时间段与预设时长参数之间的对应关系。例如,一天中的24个小时可以分为4个时间段:0:00~6:00,6:00~12:00,12:00~18:00,18:00~24:00。每个时间段可以设置对应的预设时长参数。例如,0:00~6:00对应的预设时长参数为8分钟,6:00~12:00对应的预设时长参数为10分钟,12:00~18:00对应的预设时长参数为15分钟,18:00~24:00对应的预设时长参数为20分钟。
电子设备确定出所述第二特征系数对应的时刻后,进一步确定所述时刻所处的时间段。随后,根据时间段以及时间段与预设时长参数之间的对应关系获取预设时长参数。
在一些实施例中,如图5所示,步骤S153、根据所述比较结果判断所述应用是否可被清理,包括以下步骤:
S1531,当所述第二特征系数小于或等于所述预设时长参数时,判断结果为所述应用不可被清理;
S1532,当所述第二特征系数大于所述预设时长参数时,判断结果为所述应用可被清理。
其中,电子设备得到所述第二特征系数与预设时长参数之间的比较结果后,根据所述比较结果判断所述应用是否可被清理。
具体地,当所述第二特征系数小于或等于所述预设时长参数时,表示所述应用在所述预设时长参数的时间段内可能被再次切换到前台运行,此时判断结果为所述应用不可被清理。例如,第二特征系数小于预设时长参数10分钟,表示所述应用在接下来的10分钟内可能被用户再次切换到前台运行,此时所述应用不可被清理。
当所述第二特征系数大于所述预设时长参数时,表示所述应用在所述预设时长参数的时间段内不会被再次切换到前台运行,也即用户不会再次使用所述应用,此时判断结果为所述应用可以被清理。例如,第二特征系数大于预设时长参数15分钟,表示所述应用在接下来的15分钟内不会被用户再次运行,此时所述应用可以被清理。随后,电子设备可以关闭所述应用。
具体实施时,本申请不受所描述的各个步骤的执行顺序的限制,在不产生冲突的情况下,某些步骤还可以采用其它顺序进行或者同时进行。
由上可知,本申请实施例提供的应用管理方法,根据应用的运行记录采集所述应用的多个特征信息;通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型;当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;根据所述第一特征系数从特征系数矩阵中确定出第二特征系数;根据所述第二特征系数判断所述应用是否可被清理。该方案根据应用的第二特征系数来判断所述应用是否可被清理,从而可以根据所述应用的实际运行情况准确判断出所述应用何时可以被清理,在所述应用可以被清理时,及时关闭所述应用以释放电子设备的内存,进而可以减少电子设备的内存占用,提高电子设备的运行流畅度。
本申请实施例还提供一种应用管理装置,该装置可以集成在电子设备中,该电子设备可以是智能手机、平板电脑等设备。
本申请实施例提供一种应用管理装置,包括:
采集模块,用于根据应用的运行记录采集所述应用的多个特征信息;
学习模块,用于通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;
计算模块,用于当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;
确定模块,用于根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;
判断模块,用于根据所述第二特征系数判断所述应用是否可被清理。
如图7所示,应用管理装置200可以包括:采集模块201、学习模块202、计算模块203、确定模块204以及判断模块205。
采集模块201,用于根据应用的运行记录采集所述应用的多个特征信息。
其中,所述应用可以是安装在电子设备中的任意应用程序,例如通讯应用程序、多媒体应用程序、游戏应用程序、资讯应用程序、或者购物应用程序等等。
所述应用的多个特征信息可以包括所述应用自身所固有的特征信息,例如所述应用的应用类型;所述应用的多个特征信息也可以包括所述应用在运行过程中所产生的特征信息,例如所述应用在后台的 运行时长、一天中进入后台的次数、上一次在前台的运行时长、或者进入后台的方式(例如被起始键(即HOME键)切换、被返回键切换、或者被其他APP切换)等等;所述应用的多个特征信息还可以包括电子设备运行过程中所产生的特征信息,例如电子设备的熄屏(即灭屏)时长、亮屏时长、剩余电量、网络状态、或者充电状态等等。
在电子设备的运行过程中,应用管理装置200可以对电子设备的运行状况进行记录。例如可以通过日志文件记录电子设备每次熄屏的时刻、亮屏的时刻、电量信息、网络状态、充电状态等等。当上述应用开始运行后,应用管理装置200还可以对所述应用的运行状况进行记录。例如电子设备中的日志文件可以记录所述应用开始运行的时刻、进入后台运行的时刻、进入后台的切换方式等等。
当应用管理装置200接收到应用管理请求时,采集模块201可以根据所述应用的运行记录获取所述应用的多个特征信息。其中,所述应用管理请求可以是通过用户的指令所触发的,也可以是电子设备自行触发的操作。例如,电子设备中可以设置定时,每当时间到达设置的定时,电子设备即自行触发应用管理操作。
学习模块202,用于通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型。
其中,自组织神经网络是通过自动寻找样本中的内在规律和本质属性,自组织、自适应地改变网络参数与结构。自组织神经网络具有输入层和计算层。输入层和计算层分别包括多个节点。
采集模块201采集到应用的多个特征信息后,学习模块202将所述应用的多个特征信息输入到自组织神经网络的输入层。随后,所述自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型。其中,所述自组织神经网络模型包括所述应用的特征系数矩阵,所述特征系数矩阵包括多个特征系数。
在一些实施例中,学习模块202用于执行以下步骤:
根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数;
根据得到的多个特征系数获取所述应用的自组织神经网络模型。
具体地,自组织神经网络的输入层中的某个节点(输入节点)获取到特征信息后,学习模块202根据以下公式七确定计算层中与所述节点(输入节点)匹配性最好的节点(计算节点):
其中,DistFromInput表示计算层中与输入层中的所述节点(输入节点)匹配性最好的节点(计算节点)到所述输入节点之间的距离,I表示输入到输入层中的所述节点(输入节点)的特征信息,W表示领域半径内所有节点的连接权值,n表示输入到所述节点(输入节点)的特征信息的个数。
其中,领域半径内所有节点的连接权值W根据以下公式八计算:
公式八:W(t)=W(t-1)+Θ(t-1)L(t-1)(I(t-1)-W(t-1))
其中,t表示进行计算时的时刻,W(t)为t时刻所计算的连接权值,也为所计算的特征系数,W(t-1)为(t-1)时刻节点的连接权值,Θ(t-1)为(t-1)时刻的领域半径变化率,L(t-1)为(t-1)时刻的学习率,I(t-1)为(t-1)时刻输入到自组织神经网络中的输入节点的特征信息。分别根据以下公式九和公式十计算Θ(t-1)和L(t-1):
其中,e为自然常数(超越数),e为定值,e约等于2.71828。L
0为固定常数,例如L
0可以取值0.3、0.5等。σ(t-1)表示(t-1)时刻的领域半径,σ(t-1)根据以下公式十一进行计算:
其中,σ
0为初始领域半径,σ
0为固定常数,例如σ
0可以取值5、10等等。λ(t-1)为(t-1)时刻的迭代系数,λ(t-1)根据以下公式十二进行计算:
其中,λ
0为初始迭代系数,λ
0为固定常数,例如λ
0可以取值1000。
采集模块201采集到上述应用的多个特征信息后,学习模块202根据公式七至公式十二进行迭代运算,得到所述应用的多个连接权值W(t)。连接权值W(t)也即为特征系数。最后,计算得到的多个连接权值W(t)构成特征系数矩阵。所述特征系数矩阵即包括多个特征系数。所述特征系数矩阵以及特征系数矩阵中的多个特征系数与时间之间的对应关系即构成所述应用的自组织神经网络模型。
需要说明的是,在构建应用的自组织神经网络模型时,学习模块202可以根据以上公式进行大量次数的运算。
计算模块203,用于当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数。
其中,当所述应用由前台运行进入后台运行时,应用管理装置200即可对所述应用进行是否可被清理的判断操作。
具体地,当所述应用进入后台运行时,计算模块203采集所述应用当前的特征信息。其中,当前特征信息可以包括所述应用自身所固有的特征信息,例如所述应用的应用类型;当前特征信息也可以包括所述应用在运行过程中所产生的特征信息,例如所述应用在后台的运行时长、一天中进入后台的次数、上一次在前台的运行时长、或者进入后台的方式(例如被起始键(即HOME键)切换、被返回键切换、或者被其他APP切换)等等;当前特征信息还可以包括电子设备运行过程中所产生的特征信息,例如电子设备的熄屏(即灭屏)时长、亮屏时长、剩余电量、网络状态、或者充电状态等等。
随后,计算模块203根据上述公式七至公式十二对所述应用的当前特征信息进行计算,以得到所述应用的第一特征系数。
确定模块204,用于根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数。
其中,计算模块203计算得到所述应用的第一特征系数后,确定模块204即可根据所述第一特征系数以及所述应用的自组织神经网络模型确定出所述应用的第二特征系数。
具体地,所述应用的自组织神经网络模型包括所述应用的特征系数矩阵。所述特征系数矩阵中包括多个特征系数。确定模块204可以分别将所述第一特征系数与所述特征系数矩阵中的多个特征系数进行比较,以从所述特征系数矩阵中的多个特征系数中确定出第二特征系数。
在一些实施例中,如图8所示,确定模块204包括:计算子模块2041、确定子模块2042。
计算子模块2041,用于分别计算所述第一特征系数与所述多个特征系数的差值的绝对值;
确定子模块2042,用于将绝对值最小的差值对应的特征系数确定为第二特征系数。
其中,计算模块203计算得出所述应用的第一特征系数后,计算子模块2041分别将所述第一特征系 数与所述特征系数矩阵中的多个特征系数进行比较,计算所述第一特征系数与所述多个特征系数的差值,并计算所述差值的绝对值。
得到多个差值的绝对值后,确定子模块2042从多个差值的绝对值中确定出最小的一个。随后,将绝对值最小的差值所对应的特征系数确定为第二特征系数。
判断模块205,用于根据所述第二特征系数判断所述应用是否可被清理。
其中,确定模块204得出所述应用的第二特征系数后,判断模块205即可根据所述第二特征系数进行判断,以确定所述应用是否可被清理。若所述应用不可被清理,则保留所述应用继续在后台运行,允许所述应用的进程驻留在***中。若所述应用可以被清理,则可以结束所述应用的进程,以关闭所述应用,释放所述应用占用的内存。
在一些实施例中,如图9所示,判断模块205包括:获取子模块2051、比较子模块2052、判断子模块2053。
获取子模块2051,用于获取预设时长参数;
比较子模块2052,用于将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果;
判断子模块2053,用于根据所述比较结果判断所述应用是否可被清理。
其中,预设时长参数可以是预先设置在电子设备中的一个时长值。例如,预设时长参数可以为10分钟。确定模块204得出所述应用的第二特征系数后,获取子模块2051从所述电子设备中调取所述预设时长参数。随后,比较子模块2052将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果。其中,比较结果为所述第二特征系数与所述预设时长参数之间的大小关系。比较结果包括所述第二特征系数小于或等于所述预设时长参数,或者所述第二特征系数大于所述预设时长参数。随后,判断子模块2053根据所述比较结果判断所述应用是否可被清理。
在一些实施例中,获取子模块2051用于执行以下步骤:
根据所述第二特征系数确定所述第二特征系数对应的时刻;
根据所述时刻所处的时间段以及预设对应关系获取预设时长参数,所述预设对应关系为时间段与预设时长参数之间的对应关系。
其中,第二特征系数是从所述应用的自组织神经网络模型中确定出的特征系数。第二特征系数为关于时间t的函数。获取子模块2051可以根据所述第二特征系数确定所述第二特征系数对应的时刻。
电子设备中可以预先设置时间段与预设时长参数之间的对应关系。例如,一天中的24个小时可以分为4个时间段:0:00~6:00,6:00~12:00,12:00~18:00,18:00~24:00。每个时间段可以设置对应的预设时长参数。例如,0:00~6:00对应的预设时长参数为8分钟,6:00~12:00对应的预设时长参数为10分钟,12:00~18:00对应的预设时长参数为15分钟,18:00~24:00对应的预设时长参数为20分钟。
获取子模块2051确定出所述第二特征系数对应的时刻后,进一步确定所述时刻所处的时间段。随后,根据时间段以及时间段与预设时长参数之间的对应关系获取预设时长参数。
在一些实施例中,判断子模块2053用于执行以下步骤:
当所述第二特征系数小于或等于所述预设时长参数时,判断结果为所述应用不可被清理;
当所述第二特征系数大于所述预设时长参数时,判断结果为所述应用可被清理。
其中,比较子模块2052得到所述第二特征系数与预设时长参数之间的比较结果后,判断子模块2053根据所述比较结果判断所述应用是否可被清理。
具体地,当所述第二特征系数小于或等于所述预设时长参数时,表示所述应用在所述预设时长参 数的时间段内可能被再次切换到前台运行,此时判断结果为所述应用不可被清理。例如,第二特征系数小于预设时长参数10分钟,表示所述应用在接下来的10分钟内可能被用户再次切换到前台运行,此时所述应用不可被清理。
当所述第二特征系数大于所述预设时长参数时,表示所述应用在所述预设时长参数的时间段内不会被再次切换到前台运行,也即用户不会再次使用所述应用,此时判断结果为所述应用可以被清理。例如,第二特征系数大于预设时长参数15分钟,表示所述应用在接下来的15分钟内不会被用户再次运行,此时所述应用可以被清理。随后,应用管理装置200可以关闭所述应用。
具体实施时,以上各个模块可以作为独立的实体来实现,也可以进行任意组合,作为同一或若干个实体来实现。
由上可知,本申请实施例提供的应用管理装置200,通过采集模块201根据应用的运行记录采集所述应用的多个特征信息;学习模块202通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型;计算模块203当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;确定模块204根据所述第一特征系数从特征系数矩阵中确定出第二特征系数;判断模块205根据所述第二特征系数判断所述应用是否可被清理。该方案根据应用的第二特征系数来判断所述应用是否可被清理,从而可以根据所述应用的实际运行情况准确判断出所述应用何时可以被清理,在所述应用可以被清理时,及时关闭所述应用以释放电子设备的内存,进而可以减少电子设备的内存占用,提高电子设备的运行流畅度。
本申请实施例还提供一种电子设备。该电子设备可以是智能手机、平板电脑等设备。如图10所示,电子设备300包括处理器301和存储器302。其中,处理器301与存储器302电性连接。
处理器301是电子设备300的控制中心,利用各种接口和线路连接整个电子设备的各个部分,通过运行或调用存储在存储器302内的计算机程序,以及调用存储在存储器302内的数据,执行电子设备的各种功能和处理数据,从而对电子设备进行整体监控。
在本实施例中,电子设备300中的处理器301会按照如下的步骤,将一个或一个以上的计算机程序的进程对应的指令加载到存储器302中,并由处理器301来运行存储在存储器302中的计算机程序,从而实现各种功能:
根据应用的运行记录采集所述应用的多个特征信息;
通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;
当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;
根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;
根据所述第二特征系数判断所述应用是否可被清理。
在一些实施例中,所述特征系数矩阵包括多个特征系数,根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数时,处理器301执行以下步骤:
分别计算所述第一特征系数与所述多个特征系数的差值的绝对值;
将绝对值最小的差值对应的特征系数确定为第二特征系数。
在一些实施例中,根据所述第二特征系数判断所述应用是否可被清理时,处理器301执行以下步骤:
获取预设时长参数;
将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果;
根据所述比较结果判断所述应用是否可被清理。
在一些实施例中,获取预设时长参数时,处理器301执行以下步骤:
根据所述第二特征系数确定所述第二特征系数对应的时刻;
根据所述时刻所处的时间段以及预设对应关系获取预设时长参数,所述预设对应关系为时间段与预设时长参数之间的对应关系。
在一些实施例中,根据所述比较结果判断所述应用是否可被清理时,处理器301执行以下步骤:
当所述第二特征系数小于或等于所述预设时长参数时,判断结果为所述应用不可被清理;
当所述第二特征系数大于所述预设时长参数时,判断结果为所述应用可被清理。
在一些实施例中,通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型时,处理器301执行以下步骤:
根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数;
根据得到的多个特征系数获取所述应用的自组织神经网络模型。
在一些实施例中,根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数时,处理器301根据以下公式进行计算:
W(t)=W(t-1)+Θ(t-1)L(t-1)(I(t-1)-W(t-1))
其中,W(t)为t时刻节点的连接权值,并且W(t)也为t时刻所计算的特征系数,t表示进行计算时的时刻,W(t-1)为t-1时刻节点的连接权值,Θ(t-1)为(t-1)时刻的领域半径变化率,L(t-1)为(t-1)时刻的学习率,I(t-1)为(t-1)时刻输入到自组织神经网络中的输入节点的特征信息。
存储器302可用于存储计算机程序和数据。存储器302存储的计算机程序中包含有可在处理器中执行的指令。计算机程序可以组成各种功能模块。处理器301通过调用存储在存储器302的计算机程序,从而执行各种功能应用以及数据处理。
在一些实施例中,如图11所示,电子设备300还包括:射频电路303、显示屏304、控制电路305、输入单元306、音频电路307、传感器308以及电源309。其中,处理器301分别与射频电路303、显示屏304、控制电路305、输入单元306、音频电路307、传感器308以及电源309电性连接。
射频电路303用于收发射频信号,以通过无线通信与网络设备或其他电子设备进行通信。
显示屏304可用于显示由用户输入的信息或提供给用户的信息以及电子设备的各种图形用户接口,这些图形用户接口可以由图像、文本、图标、视频和其任意组合来构成。
控制电路305与显示屏304电性连接,用于控制显示屏304显示信息。
输入单元306可用于接收输入的数字、字符信息或用户特征信息(例如指纹),以及产生与用户设置以及功能控制有关的键盘、鼠标、操作杆、光学或者轨迹球信号输入。其中,输入单元306可以包括指纹识别模组。
音频电路307可通过扬声器、传声器提供用户与电子设备之间的音频接口。
传感器308用于采集外部环境信息。传感器308可以包括环境亮度传感器、加速度传感器、陀螺仪等传感器中的一种或多种。
电源309用于给电子设备300的各个部件供电。在一些实施例中,电源309可以通过电源管理***与处理器301逻辑相连,从而通过电源管理***实现管理充电、放电、以及功耗管理等功能。
尽管图11中未示出,电子设备300还可以包括摄像头、蓝牙模块等,在此不再赘述。
由上可知,本申请实施例提供了一种电子设备,该电子设备根据应用的运行记录采集所述应用的多个特征信息;通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型;当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;根据所述第一特征系数从特征系数矩阵中确定出第二特征系数;根据所述第二特征系数判断所述应用是否可被清理。该方案根据应用的第二特征系数来判断所述应用是否可被清理,从而可以根据所述应用的实际运行情况准确判断出所述应用何时可以被清理,在所述应用可以被清理时,及时关闭所述应用以释放电子设备的内存,进而可以减少电子设备的内存占用,提高电子设备的运行流畅度。
本申请实施例还提供一种存储介质,该存储介质中存储有计算机程序,当该计算机程序在计算机上运行时,该计算机执行上述任一实施例所述的应用管理方法。
本申请实施例提供一种存储介质,所述存储介质中存储有计算机程序,当所述计算机程序在计算机上运行时,使得所述计算机执行以下步骤:
根据应用的运行记录采集所述应用的多个特征信息;
通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;
当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;
根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;
根据所述第二特征系数判断所述应用是否可被清理。
在一些实施例中,所述特征系数矩阵包括多个特征系数,所述根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数的步骤包括:
分别计算所述第一特征系数与所述多个特征系数的差值的绝对值;
将绝对值最小的差值对应的特征系数确定为第二特征系数。
在一些实施例中,所述根据所述第二特征系数判断所述应用是否可被清理的步骤包括:
获取预设时长参数;
将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果;
根据所述比较结果判断所述应用是否可被清理。
在一些实施例中,所述获取预设时长参数的步骤包括:
根据所述第二特征系数确定所述第二特征系数对应的时刻;
根据所述时刻所处的时间段以及预设对应关系获取预设时长参数,所述预设对应关系为时间段与预设时长参数之间的对应关系。
在一些实施例中,所述根据所述比较结果判断所述应用是否可被清理的步骤包括:
当所述第二特征系数小于或等于所述预设时长参数时,判断结果为所述应用不可被清理;
当所述第二特征系数大于所述预设时长参数时,判断结果为所述应用可被清理。
需要说明的是,本领域普通技术人员可以理解上述实施例的各种方法中的全部或部分步骤是可以通过程序来指令相关的硬件来完成,该程序可以存储于计算机可读存储介质中,该存储介质可以包括但不限于:只读存储器(ROM,Read Only Memory)、随机存取存储器(RAM,Random Access Memory)、磁盘或光盘等。
以上对本申请实施例所提供的应用管理方法、装置、存储介质及电子设备进行了详细介绍,本文中应用了具体个例对本申请的原理及实施方式进行了阐述,以上实施例的说明只是用于帮助理解本申请的 方法及其核心思想;同时,对于本领域的技术人员,依据本申请的思想,在具体实施方式及应用范围上均会有改变之处,综上所述,本说明书内容不应理解为对本申请的限制。
Claims (20)
- 一种应用管理方法,其中,包括:根据应用的运行记录采集所述应用的多个特征信息;通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;根据所述第二特征系数判断所述应用是否可被清理。
- 根据权利要求1所述的应用管理方法,其中,所述特征系数矩阵包括多个特征系数,所述根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数的步骤包括:分别计算所述第一特征系数与所述多个特征系数的差值的绝对值;将绝对值最小的差值对应的特征系数确定为第二特征系数。
- 根据权利要求1所述的应用管理方法,其中,所述根据所述第二特征系数判断所述应用是否可被清理的步骤包括:获取预设时长参数;将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果;根据所述比较结果判断所述应用是否可被清理。
- 根据权利要求3所述的应用管理方法,其中,所述获取预设时长参数的步骤包括:根据所述第二特征系数确定所述第二特征系数对应的时刻;根据所述时刻所处的时间段以及预设对应关系获取预设时长参数,所述预设对应关系为时间段与预设时长参数之间的对应关系。
- 根据权利要求3所述的应用管理方法,其中,所述根据所述比较结果判断所述应用是否可被清理的步骤包括:当所述第二特征系数小于或等于所述预设时长参数时,判断结果为所述应用不可被清理;当所述第二特征系数大于所述预设时长参数时,判断结果为所述应用可被清理。
- 根据权利要求1所述的应用管理方法,其中,所述通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型的步骤包括:根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数;根据得到的多个特征系数获取所述应用的自组织神经网络模型。
- 根据权利要求6所述的应用管理方法,其中,根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数时,根据以下公式进行计算:W(t)=W(t-1)+Θ(t-1)L(t-1)(I(t-1)-W(t-1))其中,W(t)为t时刻节点的连接权值,并且W(t)也为所计算的特征系数,t表示进行计算时的时刻,W(t-1)为t-1时刻节点的连接权值,Θ(t-1)为t-1时刻的领域半径变化率,L(t-1)为t-1时刻的学习率,I(t-1)为t-1时刻输入到自组织神经网络中的输入节点的特征信息。
- 一种应用管理装置,其中,包括:采集模块,用于根据应用的运行记录采集所述应用的多个特征信息;学习模块,用于通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;计算模块,用于当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;确定模块,用于根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;判断模块,用于根据所述第二特征系数判断所述应用是否可被清理。
- 一种存储介质,其中,所述存储介质中存储有计算机程序,当所述计算机程序在计算机上运行时,使得所述计算机执行以下步骤:根据应用的运行记录采集所述应用的多个特征信息;通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;根据所述第二特征系数判断所述应用是否可被清理。
- 根据权利要求9所述的存储介质,其中,所述特征系数矩阵包括多个特征系数,所述根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数的步骤包括:分别计算所述第一特征系数与所述多个特征系数的差值的绝对值;将绝对值最小的差值对应的特征系数确定为第二特征系数。
- 根据权利要求9所述的存储介质,其中,所述根据所述第二特征系数判断所述应用是否可被清理的步骤包括:获取预设时长参数;将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果;根据所述比较结果判断所述应用是否可被清理。
- 根据权利要求11所述的存储介质,其中,所述获取预设时长参数的步骤包括:根据所述第二特征系数确定所述第二特征系数对应的时刻;根据所述时刻所处的时间段以及预设对应关系获取预设时长参数,所述预设对应关系为时间段与预设时长参数之间的对应关系。
- 根据权利要求11所述的存储介质,其中,所述根据所述比较结果判断所述应用是否可被清理的步骤包括:当所述第二特征系数小于或等于所述预设时长参数时,判断结果为所述应用不可被清理;当所述第二特征系数大于所述预设时长参数时,判断结果为所述应用可被清理。
- 一种电子设备,其中,包括处理器和存储器,所述存储器中存储有计算机程序,所述处理器通过调用所述存储器中存储的所述计算机程序,用于执行以下步骤:根据应用的运行记录采集所述应用的多个特征信息;通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型,所述自组织神经网络模型包括所述应用的特征系数矩阵;当所述应用进入后台运行时,根据所述应用的当前特征信息计算所述应用的第一特征系数;根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数;根据所述第二特征系数判断所述应用是否可被清理。
- 根据权利要求14所述的电子设备,其中,所述特征系数矩阵包括多个特征系数,根据所述第一特征系数从所述特征系数矩阵中确定出第二特征系数时,所述处理器用于执行以下步骤:分别计算所述第一特征系数与所述多个特征系数的差值的绝对值;将绝对值最小的差值对应的特征系数确定为第二特征系数。
- 根据权利要求14所述的电子设备,其中,根据所述第二特征系数判断所述应用是否可被清理时,所述处理器用于执行以下步骤:获取预设时长参数;将所述第二特征系数与所述预设时长参数进行比较,以得到比较结果;根据所述比较结果判断所述应用是否可被清理。
- 根据权利要求16所述的电子设备,其中,获取预设时长参数时,所述处理器用于执行以下步骤:根据所述第二特征系数确定所述第二特征系数对应的时刻;根据所述时刻所处的时间段以及预设对应关系获取预设时长参数,所述预设对应关系为时间段与预 设时长参数之间的对应关系。
- 根据权利要求16所述的电子设备,其中,根据所述比较结果判断所述应用是否可被清理时,所述处理器用于执行以下步骤:当所述第二特征系数小于或等于所述预设时长参数时,判断结果为所述应用不可被清理;当所述第二特征系数大于所述预设时长参数时,判断结果为所述应用可被清理。
- 根据权利要求14所述的电子设备,其中,通过自组织神经网络对所述多个特征信息进行学习,以得到所述应用的自组织神经网络模型时,所述处理器用于执行以下步骤:根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数;根据得到的多个特征系数获取所述应用的自组织神经网络模型。
- 根据权利要求19所述的电子设备,其中,根据采集到的多个特征信息以及自组织神经网络中每个节点的连接权值计算出所述应用的多个特征系数时,所述处理器根据以下公式进行计算:W(t)=W(t-1)+Θ(t-1)L(t-1)(I(t-1)-W(t-1))其中,W(t)为t时刻节点的连接权值,并且W(t)也为所计算的特征系数,t表示进行计算时的时刻,W(t-1)为t-1时刻节点的连接权值,Θ(t-1)为t-1时刻的领域半径变化率,L(t-1)为t-1时刻的学习率,I(t-1)为t-1时刻输入到自组织神经网络中的输入节点的特征信息。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP18861678.3A EP3671441B1 (en) | 2017-09-30 | 2018-08-15 | Application management method and apparatus, storage medium, and electronic device |
US16/824,512 US11249645B2 (en) | 2017-09-30 | 2020-03-19 | Application management method, storage medium, and electronic apparatus |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710940338.9 | 2017-09-30 | ||
CN201710940338.9A CN107729081B (zh) | 2017-09-30 | 2017-09-30 | 应用管理方法、装置、存储介质及电子设备 |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US16/824,512 Continuation US11249645B2 (en) | 2017-09-30 | 2020-03-19 | Application management method, storage medium, and electronic apparatus |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019062369A1 true WO2019062369A1 (zh) | 2019-04-04 |
Family
ID=61210234
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2018/100684 WO2019062369A1 (zh) | 2017-09-30 | 2018-08-15 | 应用管理方法、装置、存储介质及电子设备 |
Country Status (4)
Country | Link |
---|---|
US (1) | US11249645B2 (zh) |
EP (1) | EP3671441B1 (zh) |
CN (1) | CN107729081B (zh) |
WO (1) | WO2019062369A1 (zh) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107729081B (zh) | 2017-09-30 | 2020-07-07 | Oppo广东移动通信有限公司 | 应用管理方法、装置、存储介质及电子设备 |
CN107870791B (zh) * | 2017-10-31 | 2019-07-16 | Oppo广东移动通信有限公司 | 应用管理方法、装置、存储介质及电子设备 |
KR102438784B1 (ko) * | 2018-01-05 | 2022-09-02 | 삼성전자주식회사 | 데이터를 난독화 및 복호화 하는 전자 장치 및 그의 제어 방법 |
CN110543333B (zh) * | 2018-05-29 | 2024-05-17 | 视联动力信息技术股份有限公司 | 针对处理器的休眠处理方法、装置、移动终端和存储介质 |
CN110866589B (zh) * | 2018-08-10 | 2023-06-30 | 阿里巴巴(中国)有限公司 | 深度神经网络模型的运行方法、装置及框架 |
CN111050385A (zh) * | 2018-10-15 | 2020-04-21 | 中兴通讯股份有限公司 | 一种应用清理方法、装置、设备及可读存储介质 |
CN109731336B (zh) | 2018-12-27 | 2022-09-09 | 三星电子(中国)研发中心 | 一种游戏应用的控制方法和装置 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105912370A (zh) * | 2016-05-03 | 2016-08-31 | 广东欧珀移动通信有限公司 | 移动终端的后台应用程序的控制方法、装置及移动终端 |
CN106855826A (zh) * | 2015-12-08 | 2017-06-16 | 阿里巴巴集团控股有限公司 | 一种后台应用程序的控制方法及装置 |
WO2017131469A1 (ko) * | 2016-01-28 | 2017-08-03 | 삼성전자 주식회사 | 어플리케이션을 제어하는 전자 장치와 이의 동작 방법 |
CN107145215A (zh) * | 2017-05-06 | 2017-09-08 | 维沃移动通信有限公司 | 一种后台应用程序清理方法及移动终端 |
CN107729081A (zh) * | 2017-09-30 | 2018-02-23 | 广东欧珀移动通信有限公司 | 应用管理方法、装置、存储介质及电子设备 |
CN107832848A (zh) * | 2017-10-31 | 2018-03-23 | 广东欧珀移动通信有限公司 | 应用管理方法、装置、存储介质及电子设备 |
CN107870791A (zh) * | 2017-10-31 | 2018-04-03 | 广东欧珀移动通信有限公司 | 应用管理方法、装置、存储介质及电子设备 |
Family Cites Families (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP6236296B2 (ja) * | 2013-11-14 | 2017-11-22 | 株式会社デンソーアイティーラボラトリ | 学習装置、学習プログラム、及び学習方法 |
CN103646086B (zh) * | 2013-12-13 | 2017-01-25 | 北京奇虎科技有限公司 | 一种垃圾文件的清理方法和装置 |
CN105224408A (zh) | 2015-09-22 | 2016-01-06 | 小米科技有限责任公司 | 数据清理方法、装置及*** |
JP6679086B2 (ja) * | 2015-11-11 | 2020-04-15 | 国立研究開発法人情報通信研究機構 | 学習装置、予測装置、学習方法、予測方法、およびプログラム |
CN105468416A (zh) | 2015-11-30 | 2016-04-06 | 魅族科技(中国)有限公司 | 一种应用程序清理的方法及终端 |
CN107291549B (zh) * | 2016-03-31 | 2020-11-24 | 阿里巴巴集团控股有限公司 | 一种管理应用程序的方法及装置 |
CN106202166B (zh) * | 2016-06-24 | 2020-08-18 | 北京奇虎技术服务有限公司 | 文件的清理方法、装置及相应的客户端 |
CN106155695A (zh) * | 2016-07-28 | 2016-11-23 | 努比亚技术有限公司 | 后台应用程序的清除控制装置及方法 |
CN107133094B (zh) * | 2017-06-05 | 2021-11-02 | 努比亚技术有限公司 | 应用管理方法、移动终端及计算机可读存储介质 |
US11106976B2 (en) * | 2017-08-19 | 2021-08-31 | Wave Computing, Inc. | Neural network output layer for machine learning |
CN110830400B (zh) * | 2018-08-13 | 2022-08-02 | 上海澜至半导体有限公司 | 判决反馈均衡处理装置和方法 |
-
2017
- 2017-09-30 CN CN201710940338.9A patent/CN107729081B/zh not_active Expired - Fee Related
-
2018
- 2018-08-15 WO PCT/CN2018/100684 patent/WO2019062369A1/zh unknown
- 2018-08-15 EP EP18861678.3A patent/EP3671441B1/en active Active
-
2020
- 2020-03-19 US US16/824,512 patent/US11249645B2/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106855826A (zh) * | 2015-12-08 | 2017-06-16 | 阿里巴巴集团控股有限公司 | 一种后台应用程序的控制方法及装置 |
WO2017131469A1 (ko) * | 2016-01-28 | 2017-08-03 | 삼성전자 주식회사 | 어플리케이션을 제어하는 전자 장치와 이의 동작 방법 |
CN105912370A (zh) * | 2016-05-03 | 2016-08-31 | 广东欧珀移动通信有限公司 | 移动终端的后台应用程序的控制方法、装置及移动终端 |
CN107145215A (zh) * | 2017-05-06 | 2017-09-08 | 维沃移动通信有限公司 | 一种后台应用程序清理方法及移动终端 |
CN107729081A (zh) * | 2017-09-30 | 2018-02-23 | 广东欧珀移动通信有限公司 | 应用管理方法、装置、存储介质及电子设备 |
CN107832848A (zh) * | 2017-10-31 | 2018-03-23 | 广东欧珀移动通信有限公司 | 应用管理方法、装置、存储介质及电子设备 |
CN107870791A (zh) * | 2017-10-31 | 2018-04-03 | 广东欧珀移动通信有限公司 | 应用管理方法、装置、存储介质及电子设备 |
Also Published As
Publication number | Publication date |
---|---|
EP3671441A4 (en) | 2020-08-12 |
EP3671441A1 (en) | 2020-06-24 |
CN107729081A (zh) | 2018-02-23 |
US11249645B2 (en) | 2022-02-15 |
EP3671441B1 (en) | 2021-10-27 |
US20200218456A1 (en) | 2020-07-09 |
CN107729081B (zh) | 2020-07-07 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019062369A1 (zh) | 应用管理方法、装置、存储介质及电子设备 | |
TWI754855B (zh) | 人臉圖像識別方法、裝置、電子設備及儲存介質 | |
CN108076224B (zh) | 应用程序控制方法、装置及存储介质和移动终端 | |
WO2019062413A1 (zh) | 应用程序管控方法、装置、存储介质及电子设备 | |
WO2020207249A1 (zh) | 通知消息的推送方法、装置、存储介质及电子设备 | |
WO2019085749A1 (zh) | 应用程序管控方法、装置、介质及电子设备 | |
WO2019062405A1 (zh) | 应用程序的处理方法、装置、存储介质及电子设备 | |
CN110798718B (zh) | 一种视频推荐方法以及装置 | |
CN113284142B (zh) | 图像检测方法、装置、计算机可读存储介质及计算机设备 | |
CN112069414A (zh) | 推荐模型训练方法、装置、计算机设备及存储介质 | |
WO2019062317A1 (zh) | 应用程序管控方法及电子设备 | |
CN110995810B (zh) | 一种基于人工智能的对象识别方法和相关装置 | |
CN113190757A (zh) | 多媒体资源推荐方法、装置、电子设备及存储介质 | |
US11630851B2 (en) | Systems and methods for providing predictions to applications executing on a computing device | |
CN107943570B (zh) | 应用管理方法、装置、存储介质及电子设备 | |
CN111984803B (zh) | 多媒体资源处理方法、装置、计算机设备及存储介质 | |
WO2019062358A1 (zh) | 应用程序管控方法及终端设备 | |
WO2019085750A1 (zh) | 应用程序管控方法、装置、介质及电子设备 | |
CN111797288A (zh) | 数据筛选方法、装置、存储介质及电子设备 | |
WO2019062404A1 (zh) | 应用程序的处理方法、装置、存储介质及电子设备 | |
WO2024148870A1 (zh) | 一种调频方法、装置、电子设备及可读存储介质 | |
CN107832848B (zh) | 应用管理方法、装置、存储介质及电子设备 | |
WO2019000468A1 (zh) | 用户位置识别方法、装置、存储介质及电子设备 | |
CN113052198A (zh) | 一种数据处理方法、装置、设备及存储介质 | |
CN107870791B (zh) | 应用管理方法、装置、存储介质及电子设备 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 18861678 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2018861678 Country of ref document: EP Effective date: 20200319 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |