CN108762844A - Application program preloads method, apparatus, storage medium and terminal - Google Patents

Application program preloads method, apparatus, storage medium and terminal Download PDF

Info

Publication number
CN108762844A
CN108762844A CN201810532724.9A CN201810532724A CN108762844A CN 108762844 A CN108762844 A CN 108762844A CN 201810532724 A CN201810532724 A CN 201810532724A CN 108762844 A CN108762844 A CN 108762844A
Authority
CN
China
Prior art keywords
application
applications
application program
prediction
sample
Prior art date
Application number
CN201810532724.9A
Other languages
Chinese (zh)
Inventor
陈岩
Original Assignee
Oppo广东移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Oppo广东移动通信有限公司 filed Critical Oppo广东移动通信有限公司
Priority to CN201810532724.9A priority Critical patent/CN108762844A/en
Publication of CN108762844A publication Critical patent/CN108762844A/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/44Arrangements for executing specific programs
    • G06F9/445Program loading or initiating
    • G06F9/44505Configuring for program initiating, e.g. using registry, configuration files

Abstract

The embodiment of the present application discloses application program and preloads method, apparatus, storage medium and terminal.This method includes:It detects when being triggered using pre-loaded event, target candidate set of applications is determined using first kind prediction model;The destination application to be preloaded for including in the target candidate set of applications is determined using the corresponding second class prediction model of the target candidate set of applications;The destination application is preloaded.The application may be used the different prediction model of two classes and predicted stage by stage by using above-mentioned technical proposal, improve the accuracy of application program prediction.

Description

Application program preloads method, apparatus, storage medium and terminal

Technical field

The invention relates to application program loading technique field more particularly to application program preload method, apparatus, Storage medium and terminal.

Background technology

Currently, to have become people daily for the terminals such as smart mobile phone, tablet computer, laptop and intelligent appliance Essential electronic equipment in life.As terminal device is constantly intelligent, operation is all mounted in most terminal devices System so that terminal device can install rich and varied application program, meet the needs of user is different.

It is constantly promoted, can be installed in most terminal devices dozens or even hundreds of using journey with the configuration of terminal device Sequence, and become increasingly abundant with the function of application program, the resource of application program required load when running is also more and more.Work as user When selection starts an application program, terminal can load the resource needed for the application program launching, after to be loaded, Into the initial interface of application program, whole process will usually spend the several seconds even more than ten seconds time, lead to application program Starting efficiency is relatively low, and there is an urgent need for improve.

Invention content

A kind of application program of the embodiment of the present application offer preloads method, apparatus, storage medium and terminal, can optimize and answer With the preloading scheme of program.

In a first aspect, the embodiment of the present application provides a kind of application program pre-add support method, including:

It detects when being triggered using pre-loaded event, target candidate set of applications is determined using first kind prediction model;

The target candidate set of applications is determined using the corresponding second class prediction model of the target candidate set of applications In include destination application to be preloaded;

The destination application is preloaded.

Second aspect, the embodiment of the present application provide a kind of application program pre-load means, including:

Set of applications determining module, for detect be triggered using pre-loaded event when, using the first kind predict mould Type determines target candidate set of applications;

Intended application determining module, for being determined using the corresponding second class prediction model of the target candidate set of applications The destination application to be preloaded for including in the target candidate set of applications;

Preloaded components, for being preloaded to the destination application.

The third aspect, the embodiment of the present application provide a kind of computer readable storage medium, are stored thereon with computer journey Sequence realizes the application program pre-add support method as described in the embodiment of the present application when the program is executed by processor.

Fourth aspect, the embodiment of the present application provide a kind of terminal, including memory, and processor and storage are on a memory And the computer program that can be run in processor, the processor realize such as the embodiment of the present application when executing the computer program The application program pre-add support method.

The application program that is there is provided in the embodiment of the present application preloads scheme, detects when being triggered using pre-loaded event, Target candidate set of applications is determined using first kind prediction model, then using the corresponding second class prediction of target candidate set of applications Model determines the destination application to be preloaded for including in target candidate set of applications, and pre-add is carried out to destination application It carries.By using above-mentioned technical proposal, the different prediction model of two classes may be used and predicted stage by stage, improves application program The accuracy of prediction.

Description of the drawings

Fig. 1 is a kind of flow diagram of application program pre-add support method provided by the embodiments of the present application;

Fig. 2 is a kind of preloading active window storehouse provided by the embodiments of the present application and shows the opposite of on-screen display (osd) area Position relationship schematic diagram;

Fig. 3 is another preloading active window storehouse provided by the embodiments of the present application and shows the phase of on-screen display (osd) area To position relationship schematic diagram;

Fig. 4 is that a kind of application interface provided by the embodiments of the present application migrates schematic diagram;

Fig. 5 is the flow diagram of another application program pre-add support method provided by the embodiments of the present application;

Fig. 6 is the flow diagram of another application program pre-add support method provided by the embodiments of the present application;

Fig. 7 is a kind of structure diagram of application program pre-load means provided by the embodiments of the present application;

Fig. 8 is a kind of structural schematic diagram of terminal provided by the embodiments of the present application;

Fig. 9 is the structural schematic diagram of another terminal provided by the embodiments of the present application.

Specific implementation mode

Further illustrate the technical solution of the application below with reference to the accompanying drawings and specific embodiments.It is appreciated that It is that specific embodiment described herein is used only for explaining the application, rather than the restriction to the application.It further needs exist for illustrating , illustrate only for ease of description, in attached drawing and the relevant part of the application rather than entire infrastructure.

It should be mentioned that some exemplary embodiments are described as before exemplary embodiment is discussed in greater detail The processing described as flow chart or method.Although each step is described as the processing of sequence, many of which by flow chart Step can be implemented concurrently, concomitantly or simultaneously.In addition, the sequence of each step can be rearranged.When its operation The processing can be terminated when completion, it is also possible to the additional step being not included in attached drawing.The processing can be with Corresponding to method, function, regulation, subroutine, subprogram etc..

Fig. 1 is a kind of flow diagram of application program pre-add support method provided by the embodiments of the present application, and this method can be with It is executed by application program pre-load means, wherein the device can generally be integrated in the terminal by software and or hardware realization. As shown in Figure 1, this method includes:

Step 101 detects when being triggered using pre-loaded event, determines that target candidate is answered using first kind prediction model With set.

Illustratively, the terminal in the embodiment of the present application may include mobile phone, tablet computer, laptop and intelligent family The terminal devices such as electricity.Operating system is mounted in terminal.

Illustratively, can be arranged according to actual conditions using the trigger condition of pre-loaded event, the embodiment of the present application is not Make specific limit.For example, can meet preset condition in the action for detecting user (such as picks up terminal, entr screen unlock behaviour Make or input terminal unlock operation etc.) when, pre-loaded event is applied in triggering;Or it can detect foreground application When change, pre-loaded event is applied in triggering;Or it can be after the pre- flow gauge of preloaded applications, immediately (or by pre- If after duration) it triggers and applies pre-loaded event;Or can be triggered with fixed time interval etc..It is triggered when using pre-loaded event Afterwards, system can be detected by the modes such as reading flag bit or reception triggering command and has been triggered using pre-loaded event, be had Detection mode the embodiment of the present application of body does not also limit.

In the embodiment of the present application, preloaded before detecting and being triggered using pre-loaded event, or in detection application Before whether event is triggered, the candidate application program installed in terminal can be divided, obtain multiple candidate application programs Set.Candidate application program may include all application programs installed in terminal, may also comprise certain applications program.Certain applications Program may include the application program that user is commonly used, and may also comprise third party application, that is to say, that certain applications program The application program that user is rarely employed can not included, can not also include system application.The embodiment of the present application is to candidate The method of determination and quantity of application program do not limit.It optionally, can be according to respectively being answered in the preset period of time before current time Candidate application program is determined with the access times of program and/or using duration, and preset period of time is, for example, 1 month, works as access times And/or when using duration more than corresponding threshold value, corresponding application program is determined as candidate application program, alternatively, according to Access times and/or each application program is ranked up using duration, application program in the top is determined as candidate application Program.

In the embodiment of the present application, the dividing mode of candidate set of applications is not limited.For example, can be according to application class Type divides, that is, belongs to same type of application program and be divided in the corresponding candidate pool of applications of the type, wherein this Application does not also limit the division rule of Application Type, such as can be according in the demands of individuals of user or application shop Default categories are divided into such as social class, office class, game class, shopping class, property class, photography and the class and educational of recording a video; Also can according to history during use in application program frequency of usage divide, multiple frequency of usage sections are such as set, according to application Frequency of usage of the program in historical period (in 1 month such as nearly) is included in some frequency section, and each frequency section corresponds to one A candidate's pool of applications;Training sample for model training can also be clustered, such as use K arest neighbors (k- NearestNeighbor, KNN) algorithm clustered, the division of candidate set of applications is carried out according to cluster result;Can also be It is divided according to file, that is, the application program belonged in same file folder is divided to the corresponding candidate of this document folder and answers With in collection of programs, can be named with the entitled corresponding candidate pool of applications of file;It can also be desk interface It is divided, that is, belonging to the corresponding application program of application icon of the same desk interface, to be divided to the desk interface corresponding It, can be with the corresponding candidate pool of applications name of the serial number of desktop in candidate pool of applications.In addition, candidate application The dividing mode of collection of programs can also be freely arranged by user, such as safeguard a division list, user Ke Gen in the terminal It is that each candidate pool of applications adds application program according to the actual demand of oneself.

In the embodiment of the present application, first kind prediction model is for predicting which the application program that will be most likely initiated by a user belongs to In candidate set of applications.First kind prediction model can be machine learning model, and used algorithm may include cycle nerve Network (Recurrent neural networks, RNN), shot and long term remember (LongShort-Term Memory, LSTM) net Network, deep neural network (Deep Neural Network, DNN), thresholding cycling element, simple cycle unit, autocoding Device, decision tree, random forest, characteristic mean classification, post-class processing, Hidden Markov, K arest neighbors (k- NearestNeighbor, KNN) algorithm, Logic Regression Models, Bayesian model, Gauss model and KL divergences (Kullback- Leibler divergence) etc..

Optionally, the first training sample can be acquired, and record the first training sample pair during user's using terminal The sample labeling answered, sample labeling are the candidate set of applications belonging to the application program of training sample acquisition moment opening, or The candidate application collection belonging to application program that (can be in setting duration, in 10 minutes) opens after the sample collection moment It closes, default initial model is trained using the first training sample and corresponding sample labeling, finally obtain for predicting to wait for Preloaded applications program belongs to the first kind prediction model of which candidate set of applications.Illustratively, it is wrapped in the first training sample The element contained may include time, place and the frequency etc. that application program is opened;It may include the operating status of terminal, it is such as mobile The on off state of data network, the connection status of hotspot, the hotspot connected identity information, currently running answer Duration, the current application program the last time quilt stopped on backstage with program, a upper foreground application program, current application program Switch to the time on backstage, the plug state of earphone jack, charged state, battery level information and screen display duration etc. Deng;It may also include the collected data of sensor integrated in terminal, such as motion sensor, light sensor, temperature sensor And humidity sensor etc..

Illustratively, suitable sample elements can be selected according to selected machine learning model, it also can be according to selected Sample elements determine selected machine learning model, may also be combined with the demand to prediction accuracy and pre-set velocity etc. The selection etc. of model and sample elements is carried out, the embodiment of the present application does not limit.

Illustratively, the acquisition of the first current sample can be carried out according to the sample elements for including in the first training sample, it will First current sample is input in first kind prediction model, determines that target candidate is answered according to the output result of first kind prediction model With set.Optionally, the output result of first kind prediction model can be the hit probability of each candidate set of applications, will hit Probability is higher or highest candidate set of applications is determined as target candidate set of applications.Target candidate set of applications can be one Or it is multiple, the embodiment of the present application does not limit.

Step 102 determines the target candidate using the corresponding second class prediction model of the target candidate set of applications The destination application to be preloaded for including in set of applications.

In the embodiment of the present application, preloaded before detecting and being triggered using pre-loaded event, or in detection application Before whether event is triggered, model training is carried out for each candidate set of applications, obtains each candidate set of applications difference Corresponding second class prediction model, what the second class prediction model was used to predict including in corresponding candidate set of applications will be by The application program that user starts.Second class prediction model can be machine learning model, and used algorithm may include cycle Neural network (Recurrent neural networks, RNN), deep neural network (Deep Neural Network, DNN), shot and long term memory (Long Short-Term Memory, LSTM) network, thresholding cycling element, simple cycle unit, from Dynamic encoder, decision tree, random forest, characteristic mean classification, post-class processing, Hidden Markov, K arest neighbors (k- NearestNeighbor, KNN) algorithm, Logic Regression Models, Bayesian model, Gauss model and KL divergences (Kullback- Leibler divergence) etc..Algorithm used by second class prediction model can be identical as first kind prediction model, Can be different, the embodiment of the present application does not limit.

Optionally, the second training sample can be acquired, and record the second training sample pair during user's using terminal The sample labeling answered, sample labeling are the application program opened at the training sample acquisition moment or (can after the sample collection moment To be in setting duration, such as 10 minutes in) (application program opened belongs to corresponding candidate application and collects for the application program opened Close), using the first training sample and corresponding sample labeling default initial model corresponding to current candidate pool of applications It is trained, finally obtains the second class prediction model for predicting to wait for preloaded applications program.Illustratively, the second training sample The element for including in this may include time, place and the frequency etc. that application program is opened;It may include the operating status of terminal, The on off state of mobile data network, the connection status of hotspot, the identity information of the hotspot connected, current operation Application program, a upper foreground application program, current application program backstage stop duration, current application program nearest one The secondary time for being switched to backstage, the plug state of earphone jack, charged state, battery level information and screen display duration Etc.;It may also include the collected data of sensor integrated in terminal, such as motion sensor, light sensor, temperature sensing Device and humidity sensor etc..In the embodiment of the present application, the element for including in the second training sample can be with the first training sample The element for including in this is identical, also can be different.In some embodiments, the element species number for including in the first training sample is less than The element species number for including in second training sample, the advantages of this arrangement are as follows, since first kind prediction model is for rough Prediction can reduce model training by reducing the sample scale of construction and time cost when application model is predicted, improve Predetermined speed.

Illustratively, the acquisition of the second current sample can be carried out according to the sample elements for including in the second training sample, it will Second current sample is input in the corresponding second class prediction model of target candidate set of applications, according to the second class prediction model Output result determines destination application to be preloaded.Optionally, the output result of the second class prediction model can be corresponding The startup probability of each application program of candidate set of applications, will start that probability is higher or highest application program is determined as target Application program.The probability that starts includes the probability that application program will be opened.

Step 103 preloads the destination application.

In the embodiment of the present application, when destination application only there are one when, without consider application preloading sequence;And work as When destination application has multiple, multiple destination applications can be determined as one by one currently waiting for preloaded applications program, and according to 2 or more destination applications can also be determined as currently waiting for preloaded applications program by secondary carry out preload operation, and simultaneously Carry out preload operation, that is to say, that the preloading procedure of multiple application programs can carry out parallel.

The detailed process of preloading and the resource loaded are not limited in the embodiment of the present application, may be, for example, wait for it is pre- The destination application of load distributes corresponding hardware resource, and the load of the hardware resource based on distribution starts required dependency number According to.Illustratively, it may include application process startup, application service startup, Memory Allocation, file content is read, network data obtains Take and interface render etc..In addition, the resource preloaded can be determined according to the concrete type for waiting for preloaded applications program.Example Such as, if waiting for, preloaded applications program is a social software, can preload start picture, contact person's row in the application program Table and recent Message Record etc.;If waiting for, preloaded applications program is a game, can be preloaded in the application program Game background related data etc..

Optionally, after being preloaded to the destination application, in the operation for receiving destination application When instruction, the corresponding destination application of the operating instruction is started based on the resource preloaded.

The application program pre-add support method provided in the embodiment of the present application detects when being triggered using pre-loaded event, Target candidate set of applications is determined using first kind prediction model, then using the corresponding second class prediction of target candidate set of applications Model determines the destination application to be preloaded for including in target candidate set of applications, and pre-add is carried out to destination application It carries.By using above-mentioned technical proposal, the different prediction model of two classes may be used and predicted stage by stage, improves application program The accuracy of prediction.

In some embodiments, described that target candidate set of applications is determined using first kind prediction model, including:Before acquisition The currently used sequential correlation sequence of platform application program, the currently used sequential correlation sequence include the foreground application And the sequence that at least one application program used before the foreground application is formed sequentially in time;By institute It states in currently used sequential correlation sequence inputting to first kind prediction model, according to the output result of the first kind prediction model Determine target candidate set of applications.Illustratively, when detection application program B is switched to application program A, show to detect foreground The handover operation of application program is run, at this point, currently running first application program is application program A, last moment operation Application program is application program B, then the currently used sequential correlation sequence of currently running first application program A is:Using journey Sequence B- application programs A.Again illustrative, currently running first application program is application program C, the application of last moment operation Program is application program D, is application program E in the application program that the previous moment of last moment is run.It is understood that first Application program D is switched to by application program E, then application program C is switched to by application program D, then currently running application program C Currently used sequential correlation sequence can be expressed as:Application program E- application program D- application programs C.It should be noted that The quantity of application program of the embodiment of the present application to including in the currently used sequential correlation sequence of foreground application does not limit It is fixed.The advantages of this arrangement are as follows when changing foreground application, the corresponding preamble of current foreground application program makes The sequence that will be changed with sequence, and be opened based on application program, can rapidly and accurately predict user will make Candidate set of applications belonging to application program.

In some embodiments, it is described detect be triggered using pre-loaded event before, further include:In preset time The history of collecting sample application program uses sequential correlation sequence, as the first training sample, first training sample in section Sample labeling be included in the candidate set of applications belonging to the application program used after the sample application program;It will be described First training sample and corresponding sample labeling are input in the first preset model, to be instructed to first preset model Practice, using the model after training as first kind prediction model.Optionally, it can respectively be applied in record terminal in default history cycle The sequence that program is opened, i.e., when foreground application switches, the application program after record switching, the program that is applied sequence Row.It is illustrated by taking RNN networks as an example, the step-length used is 2, therefore in sampling, every 2 application programs can be formed One training sample, application program followed by are the sample labeling of the training sample.For example, the application sequence of storage is 1, 2,2,5,3 ..., number therein represents the digital number of application program, then disassembled at sample (1,2 → 2), (2,2 → 5), (2,5 → 3) ..., wherein the first two number is input in each sample, the number after → symbol is corresponding sample labeling.It will train Sample is sequentially input into preset RNN networks, obtained first kind prediction model.

In some embodiments, described that institute is determined using the corresponding second class prediction model of the target candidate set of applications The destination application to be preloaded for including in target candidate set of applications is stated, including:Obtain the current state feature of terminal Information;The current state characteristic information is input in the corresponding second class prediction model of the target candidate set of applications, The mesh to be preloaded for including in the target candidate set of applications is determined according to the output result of the second class prediction model Mark application program.The advantages of this arrangement are as follows after predicting target candidate set of applications using application sequence, can be used more Abundant sample elements are added to carry out further precisely prediction.Wherein, current state characteristic information may include that application program is beaten Time, place and frequency for opening etc.;The operating status that may include terminal, on off state, wireless heat such as mobile data network The connection status of point, the identity information of the hotspot connected, the plug state of earphone jack, charged state, battery capacity Information and screen display duration etc.;May also include in terminal integrate the collected data of sensor, as motion sensor, Light sensor, temperature sensor and humidity sensor etc..

In some embodiments, it is described detect be triggered using pre-loaded event before, further include:In preset time The historic state characteristic information of sample application program in section in acquisition current candidate set of applications terminal when being used, makees For the second training sample, the sample labeling of second training sample includes the sample application program or in the sample application The application program used after program;Second training sample and corresponding sample labeling are input to the second preset model In, to be trained to second preset model, collect the second preset model after training as current candidate application Close corresponding second class prediction model.The advantages of this arrangement are as follows the sample elements progress model training that use is abundanter, Improve precision of prediction and accuracy.

In the embodiment of the present application, the training process of first kind prediction model and the second class prediction model can be in terminal local It carries out, can also be carried out in the other equipments such as server, the embodiment of the present application does not limit.

In some embodiments, described that the destination application is preloaded, including:It is pre- based on what is be pre-created Loading activity window stack preloads the corresponding application interface of the destination application, wherein the preloading active window The corresponding boundary coordinate of storehouse is located at except the coordinate range of display screen.The advantages of this arrangement are as follows to application interface into Row preloads, and can largely complete to promote the startup speed for waiting for preloaded applications program using the preparation before starting Degree, and do not interfere with the display of the display content of foreground application on the display screen.

In the embodiment of the present application, active window can be regarded as one it is independent be directly facing user provide interaction and operation Interface may name the interface in different operating system using different nouns.In order to make it easy to understand, below with Android (Android) following explanation is carried out for operating system.

In android system, active window is referred to as Activity.Activity, which is one, to be responsible for interacting with user Component provides screen (can be understood as screen interface, and the display screen of non-physical), certain is completed so that user interacts Item task.In an android application program, Activity is exactly usually an individual screen, it is above can be with Show that some controls can also be monitored and handle the event of user.In being managed to Activity, there is two such concept: Task (task stack) and Stack (active window storehouse).Task corresponds to an application program, and Task is used to store Activity, One or more Activity can be stored in one Task, and these Activity are followed " first-in last-out, last in, first out " Principle.And Stack is used to be managed Task, in general, each Activity shown needed for one Stack pairs screen Affiliated Task is managed, and a Stack can manage one or more Task, and certainly, Stack similarly follows storehouse Basic management principle.The not necessarily complete independent display screen of screen described here, by taking " two screens " as an example, the two Screen may be the region that two independent displays respectively show content in a full display screen.Certainly, if terminal has Two even more than two separate display screens, then " two screens " can also be two independent display screens.

In android system, multiwindow pattern is supported, it may include span mode, picture-in-picture mode and free schema (FreeForm).Under multiwindow pattern, the Stack where application program can have the size (size) of oneself, may include Using the terminal screen upper left corner as the coordinate up and down in the coordinate system of origin.For example, (a, b, c, d), general description is one The coordinate on the boundary of a rectangle, coordinate and the lower right corner that the rectangle upper left corner can be used is indicated, i.e. the top left co-ordinate of rectangle For (a, b), bottom right angular coordinate is (c, d), and such rectangular area just corresponds to the size of Stack.The interior layout of application in Stack It is to be subject to the size of Stack, that is to say, that the corresponding application interfaces of Activity are carried out in the bounds of size Display.

Under multiwindow pattern, multiple application programs can be allowed to be in visible state, including system and use it is visible per family Only system is visible.System and refer to being shown on the display screen with visible per family, and user it can be seen that;Only system It can be seen that refer to operating system visible, and user is invisible, may be blocked by the application on foreground or as the application to be realized Display screen outside show.

In the embodiment of the present application, the application interface of destination application is preloaded outside display screen, it can be with base It realizes in the multiwindow mechanism of operating system, the corresponding size of application program is located at outside display screen by multiwindow mechanism And reach to the sightless purpose of user, to not interfere with the display content of foreground application on the display screen aobvious Show.

In general, under multiwindow pattern, may exist a plurality of types of Stack, as Home Stack indicate desktop application The Stack of display, App Stack indicates the Stack that third-party application is shown, can also there is other split screens Stack etc., above-mentioned three The content for including in kind Stack can be shown on the display screen, and application activity window stack is referred to as in the embodiment of the present application.This Apply in embodiment, preloading active window storehouse (preloading Stack) has been increased newly, for indicating what preloaded applications were shown Stack, and the boundary coordinate for preloading Stack is set except the coordinate range of display screen, application journey to be preloaded Sequence may be displayed on the Stack.For android system, it can be created based on the multiwindow mechanism of android system One Stack dedicated for display preloaded applications.In the embodiment of the present application, newly-built Stack is newly-built because of that can allow It preloads Stack and possesses the size and visibility of oneself, to realize the purpose preloaded outside display screen.

In the embodiment of the present application, the establishment opportunity to preloading Stack does not limit, and can be the acquiescence before terminal is dispatched from the factory Setting preloads Stack and is in resident state, that is, preloads Stack and always exist;It can also be in starting up of terminal or terminal solution It is created after locking successfully;It can also (can be before determining destination application) establishment etc. after being triggered using pre-loaded event Deng.Optionally, described to preload that the destination application is corresponding to answer based on the preloading active window storehouse that is pre-created With interface, including:Judge whether the preloading active window storehouse being pre-created;If being not present, according to preset rules It creates and preloads active window storehouse;The destination application pair is preloaded based on the preloading active window storehouse created The application interface answered.The advantages of this arrangement are as follows after determining destination application to be preloaded, judge to preload Stack whether there is, and if it exists, is then not necessarily to create, if being not present, is created, can save system resource.It is appreciated that , when destination application includes multiple, namely need when continuously preloading multiple application programs in a short time, the Before one destination application starts load, preloading Stack, which has been created, to be finished, then second target application program starts to add Before load, it is existing to preload Stack, it may not be necessary to carry out above-mentioned judgement.

In the embodiment of the present application, to based on the tool for preloading the corresponding application interface of Stack preload target application programs Body process does not limit, such as the drafting of application interface can be carried out based on the size for preloading Stack and shown etc..

In some embodiments, described that the intended application is preloaded based on the preloading active window storehouse being pre-created The corresponding application interface of program, including:Create the corresponding target process of the destination application;In the preloading being pre-created The corresponding task stack of the destination application is created in active window storehouse;Based on the target process in the task stack Start the corresponding active window of the destination application;It is drawn based on the active window started and shows the intended application The corresponding application interface of program.The advantages of this arrangement are as follows can be based on the preloading active window except screen coordinate range Mouth storehouse is drawn and is shown to the application interface of destination application, will not interfere the operation of foreground application and show Show, ensure system stability, while effectively improving speed when destination application starts.While creating target process, It may also include the initialization procedure of target process.In the implementation procedure of above-mentioned steps, it is also possible to be related to the pre-add of other resources It carries, such as application service startup, Memory Allocation, file content reading and Network Data Capture, the embodiment of the present application provide other The preloading procedure in source does not limit.

In some embodiments, further include:It is sent to the destination application and forges focus notice, and be based on the puppet Make lasting drafting and display that focus notice keeps the corresponding application interface of the destination application within a preset period of time Update.The advantages of this arrangement are as follows focus can be obtained in destination application and answered the visible completion of system With the drafting and display at interface, the completeness of preloading is improved, and do not had an impact to the focus of foreground application.The application is implemented Focus in example is also known as input focus, and it is mutual indepedent to forge focus focus corresponding with foreground application.In general, for current Android system for, focus is unique, for example, touch etc. inputs operation a focus point come into force, for input focus believe Breath, system end and application end are consistent, and system end will send input focus letter once changing input focus information to application Changed information is ceased, ensures that system end is consistent with application end input focus information in this approach.In the embodiment of the present application, lead to The mode of separation system end and application end input focus information is crossed, realizes that application end can forge the purpose of focus.Specifically, this Shen Focus notice please be forged to preloaded applications, make preloaded applications that there is focus information in embodiments, and the focus of system end is believed Breath is still correct, and processing in this way can enable preloaded applications normally to draw, and achieve the purpose that all to load.Focus It is present in system end and application end, it is believed that be server-side (server) and client (client), system end record possesses coke The application of point, application end preserve flag bit (flag) and identify whether oneself has focus.The opportunity for forging input focus can be worked as The window system of Android increases window newly, when needing to update focus, generates forgery focus and notifies and sent.Forge focus Method can be call window the ends client change window focus method so that the window obtain focus.Specifically, It can carry out forging the transmission of focus notice based on Binder mechanism, Binder mechanism is the interprocess communication of android system Most common mode, using c/s frameworks, i.e. client/service architecture.

In the embodiment of the present application, preset time period can be designed according to actual conditions, such as can be started after preloading Can also be to start to be pre-loaded to the period etc. for completing to preload within the scope of fixed duration.In some embodiments, described default The length of period includes starting advertisement in the destination application or starting the playing duration of animation.Some application programs exist During startup, it will usually play some advertisements or animation, the playing duration of advertisement or animation be usually 3 seconds to more than ten seconds not Deng during playing advertisement or animation, user may can not carry out any operation, can only wait for and finish, and waste user Quality time.The embodiment of the present application the advantages of this arrangement are as follows, can destination application startup before, outside screen Play start advertisement or start animation, when destination application starts, can be directly entered application program homepage or its The stand-by period is reduced to further shift to an earlier date the operable time point of destination application in the operable interface of his user.

In some embodiments, the intended application journey is being preloaded based on the preloading active window storehouse being pre-created After the corresponding application interface of sequence, further include:When receiving the operating instruction of the destination application, by the preloading The corresponding application interface of destination application corresponding with the operating instruction for including in active window storehouse is migrated to described Display screen is shown.The advantages of this arrangement are as follows when destination application really needs startup, it directly will Rendered good application interface moves to display screen and is shown, can reach being switched fast for application interface, improves and starts Speed.

In some embodiments, described corresponding with the operating instruction by include in the preloading active window storehouse The corresponding application interface of destination application migrate to the display screen and shown, including:By the preloading activity The corresponding task stack of destination application corresponding with the operating instruction for including in window stack is migrated to application activity window The top of mouth storehouse;The dimension information, configuration information and visibility for updating the task stack, to realize the destination application Corresponding application interface is shown in the display screen.The advantages of this arrangement are as follows can ensure interfacial migration process Stability, ensure that recovery process is not in card screen, the problems such as blank screen or migration velocity are slow.

For some terminals, especially terminals such as mobile phone and tablet computer, in order to facilitate the use of user, display screen The display mode of curtain generally includes vertical screen display and transverse screen display, and many application program acquiescences are shown using vertical screen mode, And some application programs acquiescence is shown (such as some online games) using transverse screen mode, during the use of terminal, is had A little application programs can also hold the direction of terminal with user and carry out the switching that horizontal/vertical screen is shown.In some implementations of the application In example, the corresponding boundary coordinate of the preloadings active window storehouse is (H, 0,2H, H), the corresponding coordinate of the boundary coordinate System is system coordinates, and the origin of the system coordinates is the upper left corner of the display screen, and H is the viewing area of the display screen The long side length in domain.That is, H it is corresponding while be show screen display area it is maximum while, be aobvious in vertical screen display The height of display screen curtain is the width of display screen in transverse screen display.The purpose being arranged in this way allows for display screen transverse screen, pre-add Carry the normal display using transverse screen display and some application programs.Fig. 2 is that a kind of preloading provided by the embodiments of the present application is lived Dynamic window stack and the relative position relation schematic diagram for showing on-screen display (osd) area.As shown in Fig. 2, showing that screen is vertical screen at this time Mode, terminal system origin are the left vertex (0,0) for showing screen 201, and the width direction of display screen 201 is X-axis, Short transverse is Y-axis, and it is (H, 0,2H, H) to preload the corresponding boundary coordinates of Stack202, and H is that screen is high, i.e. left side solid line square Region within the scope of shape is main on-screen display (osd) area, and the region within the scope of the dashed rectangle of the right is to preload display area.Fig. 3 is The relative position relation of another preloading active window storehouse and display on-screen display (osd) area provided by the embodiments of the present application shows It is intended to.As shown in figure 3, showing that screen is transverse screen mode at this time, terminal system origin is the left vertex for showing screen 301 The short transverse of (0,0), display screen 301 is X-axis, and width direction is Y-axis, preloads the corresponding boundary coordinates of Stack202 and is (H, 0,2H, H), H are that screen is high, i.e., the region within the scope of the solid-line rectangle of the left side is main on-screen display (osd) area, the right dashed rectangle Region in range is to preload display area.

The reason of each boundaries Stack are arranged such is preloaded to be:

The abscissa in the upper left corner is H, be in order to prevent transverse screen when show that screen (be alternatively referred to as main screen) is shown to pre-add The interface of application is carried, because main screen is in addition to vertical screen mode, also transverse screen mode, when main screen transverse screen, in order to prevent Main screen display area domain shows the part of the application of preloading, so the upper left that will preload the corresponding rectangular areas Stack Angle abscissa is set as screen height.

The ordinate in the upper left corner is 0, is in order to which preloaded applications can be computed correctly status bar height.Android is applied For better design user interface (User Interface, UI), the self-defined top state column of meeting, if top is corresponding vertical Coordinate is not equal to 0, then the height of status bar may mistake.

The abscissa in the lower right corner is 2H (2 times of screen height), that is, the width=screen for preloading the corresponding rectangles of Stack is high, is (i.e. application interface is transverse screen display mode for transverse screen application when size in order to preload Stack can include preloading Application program).

The ordinate in the lower right corner is H, that is, the height=screen for preloading the corresponding rectangles of Stack is high, is to preload The size of Stack can include vertical screen application when preloading.

Based on above-mentioned reason, inventor sets (H, 0,2H, H) size for preloading Stack to.

In addition, Fig. 4, which is a kind of application interface provided by the embodiments of the present application, migrates schematic diagram, as shown in figure 4, receiving It is when the operating instruction of destination application, the destination application for including in preloading active window storehouse is corresponding using boundary Face 401 migrates to display screen 201 and is shown, lives specifically, migrating the task belonging to preloaded applications interface to application The top of dynamic window Stack, and update the dimension information, configuration information and visibility of the task, to which application interface can be It is normally shown on display screen.

Fig. 5 is the flow diagram of another application program pre-add support method provided by the embodiments of the present application, this method packet Include following steps:

Step 501 detects when being triggered using pre-loaded event, acquires current sample.

The current sample is input in the first kind prediction model based on RNN networks by step 502, according to the first kind The output result of prediction model determines target candidate set of applications.

The current sample is input to corresponding the second class based on DNN networks of target candidate set of applications by step 503 In prediction model, include in target candidate set of applications to be preloaded is determined according to the output result of the second class prediction model Destination application.

Step 504, to preload the destination application based on the preloading active window storehouse that is pre-created corresponding Application interface, wherein the corresponding boundary coordinate of active window storehouse that preloads is located at except the coordinate range of display screen.

Step 505, in the operating instruction for receiving destination application, will preload active window storehouse in include The application interface of the corresponding destination application of the operating instruction is migrated to display screen and is shown.

The application program pre-add support method provided in the embodiment of the present application, can be used identical training sample respectively to RNN Network and DNN networks are trained, and when being trained for RNN networks, the corresponding sample labeling of training sample is answered for candidate With set, and when being trained for DNN networks, the corresponding sample labeling of training sample is specific application program.In this way Training stage and the time of forecast period sample collection can be saved.When being triggered using pre-loaded event, by current sample It is first input in the first kind prediction model based on RNN networks, predicts target candidate set of applications, then current sample is inputted To in the second class prediction model based on DNN networks, application program to be preloaded is predicted, passes through the side preloaded outside screen Formula avoids interference of the preloading procedure to foreground application, when user really starts destination application, can directly by The application interface preloaded outside screen is migrated to display screen and is shown, effectively promotes the startup speed of destination application Degree.

Fig. 6 is the flow diagram of another application program pre-add support method provided by the embodiments of the present application, this method packet It includes:

The history of step 601, within a preset period of time collecting sample application program uses sequential correlation sequence, as first Training sample, the historic state characteristic information of terminal when collecting sample application program is used, as the second training sample.

Wherein, the sample labeling of first training sample is included in the application used after the sample application program Candidate set of applications belonging to program;The sample labeling of second training sample is included in quilt after the sample application program The application program used.

Illustratively, it when carrying out the first training sample and the second training sample acquires, can not temporarily differentiate between belonging to it Candidate set of applications only records the sample labeling of the second training sample, that is, is recorded in answering of being used after sample application program Use program.

Illustratively, the length of preset time period for example can be 2 months, after training sample acquires, according to this 2 Sequence is numbered in the frequency that each application program is used in a month.Application program point highest for the frequency, i.e., most-often used It is numbered with maximum ID, is that the frequency is minimum, i.e., the most slight ID numbers of the application assigned being least often used.Assuming that candidate Application program totally 100, final number is 1~100, then this 100 ID are divided into n class applications, is such as divided into 4 classes:The most frequently used, It commonly uses, be of little use, seldom use, per class 25.That is, being divided into 4 candidate set of applications according to the frequency, each candidate answers Include 25 candidate application programs with set.

First training sample and corresponding sample labeling are input in the first preset model by step 602, with pre- to first If model is trained, using the model after training as first kind prediction model.

Illustratively, the first preset model can be the machine learning model based on RNN networks.

Step 603, for each candidate set of applications, will the second training sample corresponding with current candidate set of applications and Corresponding sample labeling is input in the second preset model, to be trained to the second preset model, second after training is pre- If model is as the corresponding second class prediction model of current candidate set of applications.

Illustratively, the second preset model can be the machine learning model based on DNN networks.In this step, with above It is illustrated for citing, 4 the second class prediction models, i.e., the candidate application collection of each of 4 candidate set of applications can be obtained It closes and all there is oneself corresponding second class prediction model.

Step 604 detects when being triggered using pre-loaded event, acquires the first current sample and the second current sample.

First current sample is input to first kind prediction model by step 605, according to the output knot of first kind prediction model Fruit determines target candidate set of applications.

Illustratively, determining target candidate set of applications is most common candidate set of applications.

Step 606 determines corresponding second class prediction model according to target candidate set of applications, and the second current sample is defeated Enter to the second class prediction model, destination application to be preloaded is determined according to the output result of the second class prediction model.

Illustratively, the second current sample is input to the corresponding second class prediction model of most common candidate set of applications In, result is exported according to model and selects one or more destination applications from 25 application programs.

Step 607, to preload the destination application based on the preloading active window storehouse that is pre-created corresponding Application interface.

Wherein, the corresponding boundary coordinate of active window storehouse that preloads is located at except the coordinate range of display screen.

Step 608, in the operating instruction for receiving destination application, will preload active window storehouse in include The application interface of the corresponding destination application of the operating instruction is migrated to display screen and is shown.

Application program pre-add support method provided by the embodiments of the present application, terminal historical period acquire the first training sample and Second training sample, and mould is predicted based on the first kind that the first training sample trains to obtain for predicting target candidate set of applications Type trains to obtain the second class prediction model for predicting to wait for preloaded applications program, to reach based on the second training sample The purpose of sample collection and model training is carried out in terminal local, and reduces data transmission so that sample collection and prediction mould The update of type much sooner, while can save flow, be applicable to the mobile terminals such as mobile phone.It is touched using pre-loaded event When hair, segmented prediction is carried out using two class prediction models, improves precision of prediction and the accuracy of application program.It is waited for predicting After the application program of preloading, by way of being preloaded outside screen, interference of the preloading procedure to foreground application is avoided, When user really starts destination application, directly the application interface preloaded outside screen can be migrated to display screen Curtain is shown, the startup speed of destination application is effectively promoted.

Fig. 7 is a kind of structure diagram of application program pre-load means provided by the embodiments of the present application, which can be by soft Part and/or hardware realization, are typically integrated in terminal, can carry out application program by executing application pre-add support method It preloads.As shown in fig. 7, the device includes:

Set of applications determining module 701, for detect be triggered using pre-loaded event when, predicted using the first kind Model determines target candidate set of applications;

Intended application determining module 702, for using the corresponding second class prediction model of the target candidate set of applications Determine the destination application to be preloaded for including in the target candidate set of applications;

Preloaded components 703, for being preloaded to the destination application.

The application program pre-load means provided in the embodiment of the present application detect when being triggered using pre-loaded event, Target candidate set of applications is determined using first kind prediction model, then using the corresponding second class prediction of target candidate set of applications Model determines the destination application to be preloaded for including in target candidate set of applications, and pre-add is carried out to destination application It carries.By using above-mentioned technical proposal, the different prediction model of two classes may be used and predicted stage by stage, improves application program The accuracy of prediction.

Optionally, described that target candidate set of applications is determined using first kind prediction model, including:

The currently used sequential correlation sequence of foreground application is obtained, the currently used sequential correlation sequence includes institute State foreground application and at least one application program for being used before the foreground application sequentially in time The sequence of formation;

By in the currently used sequential correlation sequence inputting to first kind prediction model, mould is predicted according to the first kind The output result of type determines target candidate set of applications.

Optionally, which further includes:

First training sample acquisition module, for it is described detect be triggered using pre-loaded event before, default The history of collecting sample application program uses sequential correlation sequence in period, as the first training sample, first training The sample labeling of sample is included in the candidate set of applications belonging to the application program used after the sample application program;

First model training module is preset for first training sample and corresponding sample labeling to be input to first In model, to be trained to first preset model, using the model after training as first kind prediction model.

Optionally, described to determine that the target is waited using the corresponding second class prediction model of the target candidate set of applications The destination application to be preloaded for including in set of applications is selected, including:

Obtain the current state characteristic information of terminal;

The current state characteristic information is input to the corresponding second class prediction model of the target candidate set of applications In, include in the target candidate set of applications to be preloaded is determined according to the output result of the second class prediction model Destination application.

Optionally, which further includes:

Second training sample acquisition module, for it is described detect be triggered using pre-loaded event before, default The historic state feature letter of sample application program in period in acquisition current candidate set of applications terminal when being used Breath, as the second training sample, the sample labeling of second training sample is included in the sample application program to be made later Application program;

Second model training module is preset for second training sample and corresponding sample labeling to be input to second In model, to be trained to second preset model, the second preset model after training is answered as the current candidate With the corresponding second class prediction model of set.

Optionally, the division rule of candidate set of applications includes being divided according to Application Type or according to history validity period The frequency of usage of interior application program divides.

Optionally, described that the destination application is preloaded, including:

The corresponding application interface of the destination application is preloaded based on the preloading active window storehouse being pre-created, Wherein, the corresponding boundary coordinate of active window storehouse that preloads is located at except the coordinate range of display screen.

The embodiment of the present application also provides a kind of storage medium including computer executable instructions, and the computer is executable Instruction is used for executing application pre-add support method when being executed by computer processor, and this method includes:

It detects when being triggered using pre-loaded event, target candidate set of applications is determined using first kind prediction model;

The target candidate set of applications is determined using the corresponding second class prediction model of the target candidate set of applications In include destination application to be preloaded;

The destination application is preloaded.

Storage medium --- any various types of memory devices or storage device.Term " storage medium " is intended to wrap It includes:Install medium, such as CD-ROM, floppy disk or magnetic tape equipment;Computer system memory or random access memory, such as DRAM, DDRRAM, SRAM, EDORAM, blue Bath (Rambus) RAM etc.;Nonvolatile memory, such as flash memory, magnetic medium (example Such as hard disk or optical storage);The memory component etc. of register or other similar types.Storage medium can further include other types Memory or combinations thereof.In addition, storage medium can be located at program in the first computer system being wherein performed, or It can be located in different second computer systems, second computer system is connected to the first meter by network (such as internet) Calculation machine system.Second computer system can provide program instruction to the first computer for executing.Term " storage medium " can To include two or more that may reside in different location (such as in different computer systems by network connection) Storage medium.Storage medium can store the program instruction that can be executed by one or more processors and (such as be implemented as counting Calculation machine program).

Certainly, a kind of storage medium including computer executable instructions that the embodiment of the present application is provided, computer The application program preload operation that executable instruction is not limited to the described above can also be performed the application any embodiment and be provided Application program pre-add support method in relevant operation.

The embodiment of the present application provides a kind of terminal, and it is pre- that application program provided by the embodiments of the present application can be integrated in the terminal Loading device.Fig. 8 is a kind of structural schematic diagram of terminal provided by the embodiments of the present application.Terminal 600 may include:Memory 801, processor 802 and it is stored in the computer program that can be run on memory 801 and in processor 802, the processor 802 The application program pre-add support method as described in the embodiment of the present application is realized when executing the computer program.

Terminal provided by the embodiments of the present application may be used the different prediction model of two classes and be predicted stage by stage, improve The accuracy of application program prediction.

Fig. 9 is the structural schematic diagram of another terminal provided by the embodiments of the present application, which may include:Shell (figure In be not shown), memory 901, central processing unit (central processing unit, CPU) 902 (also known as processor, with Lower abbreviation CPU), circuit board (not shown) and power circuit (not shown).The circuit board is placed in the shell The space interior surrounded;The CPU902 and the memory 901 are arranged on the circuit board;The power circuit, is used for It powers for each circuit or device of the terminal;The memory 901, for storing executable program code;It is described CPU902 is run and the executable program code pair by reading the executable program code stored in the memory 901 The computer program answered, to realize following steps:

It detects when being triggered using pre-loaded event, target candidate set of applications is determined using first kind prediction model;

The target candidate set of applications is determined using the corresponding second class prediction model of the target candidate set of applications In include destination application to be preloaded;

The destination application is preloaded.

The terminal further includes:Peripheral Interface 903, RF (Radio Frequency, radio frequency) circuit 905, voicefrequency circuit 906, loud speaker 911, power management chip 908, input/output (I/O) subsystem 909, other input/control devicess 910, touch Touch screen 912, other input/control devicess 910 and outside port 904, these components by one or more communication bus or Signal wire 907 communicates.

It should be understood that graphic terminal 900 is only an example of terminal, and terminal 900 can have than figure Shown in more or less components, two or more components can be combined, or can have different portions Part configures.Various parts shown in the drawings can be including one or more signal processings and/or application-specific integrated circuit Hardware, software or hardware and software combination in realize.

Just the terminal provided in this embodiment preloaded for application program is described in detail below, and the terminal is with hand For machine.

Memory 901, the memory 901 can be by access such as CPU902, Peripheral Interfaces 903, and the memory 901 can Can also include nonvolatile memory to include high-speed random access memory, such as one or more disk memory, Flush memory device or other volatile solid-state parts.

The peripheral hardware that outputs and inputs of equipment can be connected to CPU902 and deposited by Peripheral Interface 903, the Peripheral Interface 903 Reservoir 901.

I/O subsystems 909, the I/O subsystems 909 can be by the input/output peripherals in equipment, such as touch screen 912 With other input/control devicess 910, it is connected to Peripheral Interface 903.I/O subsystems 909 may include 9091 He of display controller One or more input controllers 9092 for controlling other input/control devicess 910.Wherein, one or more input controls Device 9092 processed receives electric signal from other input/control devicess 910 or sends electric signal to other input/control devicess 910, Other input/control devicess 910 may include physical button (pressing button, rocker buttons etc.), dial, slide switch, behaviour Vertical pole clicks idler wheel.It is worth noting that input controller 9092 can with it is following any one connect:Keyboard, infrared port, The indicating equipment of USB interface and such as mouse.

Touch screen 912, the touch screen 912 are the input interface and output interface between user terminal and user, can It is shown to user depending on output, visual output may include figure, text, icon, video etc..

Display controller 9091 in I/O subsystems 909 receives electric signal from touch screen 912 or is sent out to touch screen 912 Electric signals.Touch screen 912 detects the contact on touch screen, and the contact detected is converted to and is shown by display controller 9091 The interaction of user interface object on touch screen 912, that is, realize human-computer interaction, the user interface being shown on touch screen 912 Object can be the icon of running game, be networked to the icon etc. of corresponding network.It is worth noting that equipment can also include light Mouse, light mouse are the extensions for the touch sensitive surface for not showing the touch sensitive surface visually exported, or formed by touch screen.

RF circuits 905 are mainly used for establishing the communication of mobile phone and wireless network (i.e. network side), realize mobile phone and wireless network The data receiver of network and transmission.Such as transmitting-receiving short message, Email etc..Specifically, RF circuits 905 receive and send RF letters Number, RF signals are also referred to as electromagnetic signal, and RF circuits 905 convert electrical signals to electromagnetic signal or electromagnetic signal is converted to telecommunications Number, and communicated with communication network and other equipment by the electromagnetic signal.RF circuits 905 may include for executing The known circuit of these functions comprising but it is not limited to antenna system, RF transceivers, one or more amplifiers, tuner, one A or multiple oscillators, digital signal processor, CODEC (COder-DECoder, coder) chipset, user identifier mould Block (Subscriber Identity Module, SIM) etc..

Voicefrequency circuit 906 is mainly used for receiving audio data from Peripheral Interface 903, which is converted to telecommunications Number, and the electric signal is sent to loud speaker 911.

Loud speaker 911, the voice signal for receiving mobile phone from wireless network by RF circuits 905, is reduced to sound And play the sound to user.

Power management chip 908, the hardware for being connected by CPU902, I/O subsystem and Peripheral Interface are powered And power management.

It is arbitrarily real that application program pre-load means, storage medium and the terminal provided in above-described embodiment can perform the application The application program pre-add support method that example is provided is applied, has and executes the corresponding function module of this method and advantageous effect.Not upper The technical detail of detailed description in embodiment is stated, reference can be made to the application program preloading side that the application any embodiment is provided Method.

Note that above are only preferred embodiment and the institute's application technology principle of the application.It will be appreciated by those skilled in the art that The application is not limited to specific embodiment described here, can carry out for a person skilled in the art it is various it is apparent variation, The protection domain readjusted and substituted without departing from the application.Therefore, although being carried out to the application by above example It is described in further detail, but the application is not limited only to above example, in the case where not departing from the application design, also May include other more equivalent embodiments, and scope of the present application is determined by scope of the appended claims.

Claims (10)

1. a kind of application program pre-add support method, which is characterized in that including:
It detects when being triggered using pre-loaded event, target candidate set of applications is determined using first kind prediction model;
It is determined in the target candidate set of applications and is wrapped using the corresponding second class prediction model of the target candidate set of applications The destination application to be preloaded contained;
The destination application is preloaded.
2. according to the method described in claim 1, it is characterized in that, described determine that target candidate is answered using first kind prediction model With set, including:
The currently used sequential correlation sequence of foreground application is obtained, before the currently used sequential correlation sequence includes described Platform application program and at least one application program used before the foreground application are formed sequentially in time Sequence;
By in the currently used sequential correlation sequence inputting to first kind prediction model, according to the first kind prediction model Output result determines target candidate set of applications.
3. according to the method described in claim 2, it is characterized in that, being triggered it using pre-loaded event in described detect Before, further include:
The history of collecting sample application program uses sequential correlation sequence within a preset period of time, as the first training sample, institute The sample labeling for stating the first training sample is included in candidate belonging to the application program used after the sample application program Set of applications;
First training sample and corresponding sample labeling are input in the first preset model, with to the described first default mould Type is trained, using the model after training as first kind prediction model.
4. according to the method described in claim 2, it is characterized in that, described using the target candidate set of applications corresponding the Two class prediction models determine the destination application to be preloaded for including in the target candidate set of applications, including:
Obtain the current state characteristic information of terminal;
The current state characteristic information is input in the corresponding second class prediction model of the target candidate set of applications, root The target to be preloaded for including in the target candidate set of applications is determined according to the output result of the second class prediction model Application program.
5. according to the method described in claim 4, it is characterized in that, being triggered it using pre-loaded event in described detect Before, further include:
The history of the terminal when sample application program in current candidate set of applications is used is acquired within a preset period of time State characteristic information, as the second training sample, the sample labeling of second training sample includes the sample application program Or the application program used after the sample application program;
Second training sample and corresponding sample labeling are input in the second preset model, with to the described second default mould Type is trained, and mould is predicted using the second preset model after training as corresponding second class of the current candidate set of applications Type.
6. according to the method described in claim 1, it is characterized in that, the division rule of candidate set of applications includes according to using journey Sequence Type division or according to history during use in application program frequency of usage divide.
7. according to the method described in claim 1, it is characterized in that, described preload the destination application, packet It includes:
The corresponding application interface of the destination application is preloaded based on the preloading active window storehouse being pre-created, In, the corresponding boundary coordinate of active window storehouse that preloads is located at except the coordinate range of display screen.
8. a kind of application program pre-load means, which is characterized in that including:
Set of applications determining module, for detect be triggered using pre-loaded event when, it is true using first kind prediction model Set the goal candidate set of applications;
Intended application determining module, described in being determined using the corresponding second class prediction model of the target candidate set of applications The destination application to be preloaded for including in target candidate set of applications;
Preloaded components, for being preloaded to the destination application.
9. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is held by processor The application program pre-add support method as described in any in claim 1-7 is realized when row.
10. a kind of terminal, which is characterized in that including memory, processor and storage can be run on a memory and in processor Computer program, the application journey as described in claim 1-7 is any is realized when the processor executes the computer program Sequence pre-add support method.
CN201810532724.9A 2018-05-29 2018-05-29 Application program preloads method, apparatus, storage medium and terminal CN108762844A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810532724.9A CN108762844A (en) 2018-05-29 2018-05-29 Application program preloads method, apparatus, storage medium and terminal

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201810532724.9A CN108762844A (en) 2018-05-29 2018-05-29 Application program preloads method, apparatus, storage medium and terminal
PCT/CN2019/085506 WO2019228134A1 (en) 2018-05-29 2019-05-05 Application program pre-loading method and device, storage medium, and terminal

Publications (1)

Publication Number Publication Date
CN108762844A true CN108762844A (en) 2018-11-06

Family

ID=64003702

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810532724.9A CN108762844A (en) 2018-05-29 2018-05-29 Application program preloads method, apparatus, storage medium and terminal

Country Status (2)

Country Link
CN (1) CN108762844A (en)
WO (1) WO2019228134A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019228134A1 (en) * 2018-05-29 2019-12-05 Oppo广东移动通信有限公司 Application program pre-loading method and device, storage medium, and terminal

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103376974A (en) * 2013-06-28 2013-10-30 广州市动景计算机科技有限公司 Window switching method and device
CN106406966A (en) * 2016-10-31 2017-02-15 维沃移动通信有限公司 Preloading method for application and mobile terminal
CN106919313A (en) * 2017-04-21 2017-07-04 维沃移动通信有限公司 The startup method and mobile terminal of a kind of application program
CN107783801A (en) * 2017-11-06 2018-03-09 广东欧珀移动通信有限公司 Application program forecast model is established, preloads method, apparatus, medium and terminal
CN107807935A (en) * 2016-09-09 2018-03-16 腾讯科技(深圳)有限公司 Using recommendation method and device

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2783284B1 (en) * 2011-11-22 2019-03-13 Solano Labs, Inc. System of distributed software quality improvement
US9508040B2 (en) * 2013-06-12 2016-11-29 Microsoft Technology Licensing, Llc Predictive pre-launch for applications
CN105446988B (en) * 2014-06-30 2018-10-30 华为技术有限公司 The method and apparatus for predicting classification
CN105939416A (en) * 2016-05-30 2016-09-14 努比亚技术有限公司 Mobile terminal and application prestart method thereof
CN108762844A (en) * 2018-05-29 2018-11-06 Oppo广东移动通信有限公司 Application program preloads method, apparatus, storage medium and terminal

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103376974A (en) * 2013-06-28 2013-10-30 广州市动景计算机科技有限公司 Window switching method and device
CN107807935A (en) * 2016-09-09 2018-03-16 腾讯科技(深圳)有限公司 Using recommendation method and device
CN106406966A (en) * 2016-10-31 2017-02-15 维沃移动通信有限公司 Preloading method for application and mobile terminal
CN106919313A (en) * 2017-04-21 2017-07-04 维沃移动通信有限公司 The startup method and mobile terminal of a kind of application program
CN107783801A (en) * 2017-11-06 2018-03-09 广东欧珀移动通信有限公司 Application program forecast model is established, preloads method, apparatus, medium and terminal

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019228134A1 (en) * 2018-05-29 2019-12-05 Oppo广东移动通信有限公司 Application program pre-loading method and device, storage medium, and terminal

Also Published As

Publication number Publication date
WO2019228134A1 (en) 2019-12-05

Similar Documents

Publication Publication Date Title
CN205038630U (en) Electronic equipment
KR20170072904A (en) Method, system for implementing operation of mobile terminal according to touching signal and mobile terminal
CN106415431B (en) For sending method, computer-readable medium and the electronic equipment of instruction
WO2016169465A1 (en) Method, apparatus and system for displaying screen information
CA2813172C (en) Multiple-access-level lock screen
US9009630B2 (en) Above-lock notes
US9779527B2 (en) Method, terminal device and storage medium for processing image
EP2981104A1 (en) Apparatus and method for providing information
ES2643176T3 (en) Method and apparatus for providing independent view activity reports that respond to a tactile gesture
CN108769373A (en) Equipment, method and graphic user interface for providing notice and being interacted with notice
US20140025371A1 (en) Method and apparatus for recommending texts
US20140019912A1 (en) System and method for processing sliding operations on portable terminal devices
CN103902640B (en) Portable electronic devices, content recommendation method and computer-readable media
EP3035283A1 (en) Image processing method and apparatus, and terminal device
US9588642B2 (en) Information processing apparatus and application controlling method
US10739854B2 (en) Terminal and touch response method and device
US20140101617A1 (en) Method and apparatus for generating task recommendation icon in a mobile device
US10547729B2 (en) Electronic device and method of executing function of electronic device
WO2015127825A1 (en) Expression input method and apparatus and electronic device
US8893054B2 (en) Devices, systems, and methods for conveying gesture commands
US20160063339A1 (en) Scrapped information providing method and apparatus
CN102033710B (en) Method for managing file folder and related equipment
CN104077515A (en) Terminal device and terminal control program
KR20130127349A (en) System and control method for character make-up
CN105009033A (en) Interactive inputs for background task

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination