CN103246526B - Client prestrain method and client pre-load means - Google Patents
Client prestrain method and client pre-load means Download PDFInfo
- Publication number
- CN103246526B CN103246526B CN201210025953.4A CN201210025953A CN103246526B CN 103246526 B CN103246526 B CN 103246526B CN 201210025953 A CN201210025953 A CN 201210025953A CN 103246526 B CN103246526 B CN 103246526B
- Authority
- CN
- China
- Prior art keywords
- scene
- user
- prestrain
- data
- loading data
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a kind of client prestrain method and client pre-load means. This client prestrain method comprises: the usage behavior characteristic of obtaining in advance user under each scene; Obtain user's usage behavior characteristic corresponding to user's current scene, calculate user under current scene according to the predicting strategy setting in advance and respectively operate corresponding prestrain requirements; Obtain the required loading data of handoff scenario, described handoff scenario is the scene that prestrain requirements exceedes the operation correspondence of predefined prestrain demand threshold; Monitor user ' scene is switched, and determines that the scene scene corresponding with the loading data of obtaining after switching matches, and shows the loading data of obtaining. Application the present invention, can reduce the client load time.
Description
Technical field
The present invention relates to computer communication technology, particularly a kind of client prestrain method and clientPre-load means.
Background technology
In current most client, when loading, need client to obtain corresponding loading number from serverAccording to, and due to the restriction of network transfer speeds and the loading data volume that need to obtain, client is to clothesBusiness device sends and loads request of data, responds and return to load required data to server, may need etc.Treat tens milliseconds to several seconds, the even time of tens seconds, if the response time is long, will make to loadTime is longer.
Generally, client need to be obtained and load data majorities and occur in that scene switches from serverTime, for example, for common website, it may be the redirect between the page that scene is switched, for richnessClient, scene switch may be calling or switching between different scenes, or in certain scene notFor example, with the switching (, tab label switching, upper nextpage switching etc.) of little scene. When user passes through clientWhen end is carried out the operation of scene switching, need waiting for server to return to new scene and play up required loading numberAccording to, or wait for the loading data that need from buffer memory (cache) again packing, and like this, if userEntering before new scene, expend the long data load time, will make that scene switch speed is slow, pageFace shows discontinuous, and user experiences may be subject to larger impact, thereby, while how reducing scene switchingThe data load time, be the emphasis that current client loads research.
For current most of client, general employing in the time carrying out scene handover operation triggered to serverThe mode of request of loading data, only has in the time that user determines execution scene handover operation, just to serverRequest is played up and is switched the rear needed loading data of scene, although this mode can be switched demand by sceneObtain loading data, and can reduce the waste of the unnecessary network bandwidth, but cut carrying out scene at every turnWhile changing, user such as needs at the loading of to be switched back court scape desired data, cannot make full use of user and browseThe prestrain that the time of the front scene of switching is switched rear scene, and because the load time is longer, it is to usingThe impact that family is experienced also becomes inevitable.
Based on above-mentioned technical problem, prior art has proposed the method for several client prestrains, briefly retouchesState as follows.
One, as required or by the client prestrain mode of experience:
In the time that user browses web sites first, according to developer's experience, the scene that some users are comparatively commonly usedCarry out prestrain in client, like this, in the time that user enters these scenes, without obtaining from serverLoad data, thereby reduced period of reservation of number, improved client loading velocity. But the method existsWhile loading first, may need the data content of several default scenes to load together, increase startupThe time of client; Further, by the default scene that needs prestrain of developer's experience, be difficult to doTo the use habit that accurately meets user; And this prestrain method also cannot make full use of user and browseThe time of current scene is carried out the prestrain that scene may appear in future.
Two, the client prestrain mode that picture rolls:
At present, much there is the client application of a large amount of image contents all to adopt the side of picture rolling prestrainFormula, for example, the microblogging picture presentation function of Tengxun's microblogging, Sina's microblogging. The method is for same webpageThe page by the viewing area of the current browser of preferential Web page loading, and adds one outside viewing areaPicture within the scope of the prestrain of individual setting or word, to accelerate the subsequent load response speed of webpage, byIn having added certain prestrain scope, in the time that user uses the slow scroll through pages of scroll bar, due in advancePicture in loading range has just been loaded in the time that user browses current viewing area, thereby, userCannot discover because picture loads the delay sensation producing, although this prestrain scheme realizes letterSingle, but can only apply to the application under single game scape, and be not suitable for complicated many scenes switching prestrains,When many scenes are switched, still need to pull required loading data from server, the load time is longer.
Summary of the invention
In view of this, main purpose of the present invention is to propose a kind of client prestrain method, and reduction addsThe time of carrying.
Another object of the present invention is to propose a kind of client pre-load means, reduce the load time.
For achieving the above object, the invention provides a kind of client prestrain method, the method comprises:
Obtain in advance the usage behavior characteristic of user under each scene;
Obtain user's usage behavior characteristic corresponding to user's current scene, according to the predicting strategy setting in advanceCalculate user under current scene and respectively operate corresponding prestrain requirements;
Obtain the required loading data of handoff scenario, described handoff scenario is that prestrain requirements exceedes in advanceThe scene of the operation correspondence of the prestrain demand threshold of setting;
Monitor user ' scene is switched, and determines the scene scene phase corresponding with the loading data of obtaining after switchingCoupling, shows the loading data of obtaining.
Described counting user usage behavior characteristic comprises:
Taking user as mark, add up respectively each user's usage behavior characteristic; Or
Add up all users' usage behavior characteristic, obtain all users' usage behavior characteristic averageValue, as user's usage behavior characteristic.
Described before obtaining user's usage behavior characteristic corresponding to user's current scene, further comprise:
Client to user first usage behavior characteristic sort, choose the field of default number before sequenceScape, pulls the required loading data of scene of this default number, forms and loads data set storage;
Determine that current scene is the scene that user browses first, from pre-stored loading data centralization, looks intoInquiry is obtained loading data corresponding to current scene and is shown.
Described user's usage behavior characteristic comprises: time response and operating characteristic, wherein,
Described time response comprises: mean value and the mean square deviation of user's time of staying in different scenes;
Described operating characteristic comprises that user switches to the probability of another scene from a scene.
The described predicting strategy according to setting in advance calculates user under current scene and respectively operates corresponding adding in advanceCarrying requirements comprises:
Calculate the time of staying of user in current scene by the mean value weight calculation function setting in advanceMean value weight, multiply each other with mean value weight coefficient;
Calculate the time of staying of user in current scene by the mean square deviation weight calculation function setting in advanceMean square deviation weight, multiply each other with mean square deviation weight coefficient;
Calculating user's executable operations under current scene by the probability right computing function setting in advance cutsChange the probability right to handoff scenario, with probability right multiplication; Or by the probability setting in advanceWeight calculation function calculates user and under current scene, passes through the scene probability of handoff scenario again after switchingWeight, multiplies each other with the number of plies power of the prediction of probability right coefficient and decay factor; And
The long-pending addition of respectively multiplying each other is obtained to prestrain requirements corresponding to user's executable operations under current scene.
After the step of described and probability right multiplication, further comprise:
Calculate user under current scene by the time Estimate value weight calculation function that pulls setting in advanceAfter executable operations, switch to the required loading data of handoff scenario corresponding pull time Estimate value weight, and drawGet time Estimate value weight coefficient and multiply each other, and carry out the step of the long-pending addition of respectively multiplying each other;
The step multiplying each other at the number of plies power of the prediction of described and probability right coefficient and decay factor itAfter, further comprise:
Current scene is estimated to the acquisition time by way of the required loading data of scene of last handoff scenarioThe summation of value, with the number of plies power phase of prediction that pulls time Estimate value weight coefficient and decay factorTake advantage of, and carry out the step of the long-pending addition of respectively multiplying each other.
The described required loading data of handoff scenario of obtaining comprise:
Inquire about pre-stored loading data set, if exist the required loading data of handoff scenario and this to addIt is up-to-date carrying data, obtains the required loading data of this handoff scenario from loading data centralization; Otherwise,
Send data acquisition request to server, pull prestrain requirements from server and exceed and presetThe required loading data of handoff scenario of operation correspondence of prestrain demand threshold, and be stored in loading numberAccording to concentrating; Or, sending data preparation request to server, server receives data preparation request, inquiryObtain the required loading data of handoff scenario, and encapsulate loading data.
Further comprise:
Be greater than in current time stamp and the difference of the timestamp that loads data the time renewal threshold value setting in advanceTime, client initiatively pulls corresponding loading data and the loading data of storage is carried out more to serverNewly.
Further comprise:
Determine that the scene scene corresponding with the loading data of obtaining after switching do not match, and interrupts enteringThe preloading data of row transmission, the required loading data of scene from server pulls switching.
Further comprise:
Determine that user exits scene, switch the use of user under each scene of upgrading storage according to user's sceneBehavioral trait.
A kind of client pre-load means, this device comprises: usage behavior statistics of features module, prestrainRequirements computing module, preloading data acquisition module and scene matching module, wherein,
Usage behavior statistics of features module, obtains the usage behavior characteristic of user under each scene;
Prestrain requirements computing module, obtains user's usage behavior characteristic corresponding to user's current scene,Calculate user under current scene according to the predicting strategy setting in advance and respectively operate corresponding prestrain requirements;
Preloading data acquisition module, obtains the required loading data of handoff scenario, and described handoff scenario isPrestrain requirements exceedes the scene of the operation correspondence of predefined prestrain demand threshold;
Scene matching module, monitor user ' scene is switched, and determines scene and the loading number obtaining after switchingMatch according to corresponding scene, show the loading data of obtaining.
Described prestrain requirements computing module comprises: mean value weight calculation unit, mean square deviation weight meterCalculate unit, probability right computing unit and prestrain requirements computing unit, wherein,
Mean value weight calculation unit, calculates user by the mean value weight calculation function setting in advance and existsThe mean value weight of the time of staying of current scene;
Mean square deviation weight calculation unit, calculates user by the mean square deviation weight calculation function setting in advance and existsThe mean square deviation weight of the time of staying of current scene;
Probability right computing unit, calculates user current by the probability right computing function setting in advanceUnder scene, executable operations switches to the probability right of handoff scenario;
Prestrain requirements computing unit, the mean value weight that mean value weight calculation unit is calculatedMultiply each other with mean value weight coefficient, the mean square deviation weight that mean square deviation weight calculation unit is calculated is with equalVariance weight coefficient multiplies each other, the probability right that probability right computing unit is calculated and probability right systemNumber multiplies each other, and by the long-pending addition of respectively multiplying each other.
Described scene matching module comprises: monitoring means, scene matching unit, load data set unit withAnd load data display unit, wherein,
Monitoring means, monitor user ' scene is switched, and exports the user's scene handover information monitoring to fieldScape matching unit;
Scene matching unit, receives user's scene handover information, with the loading that loads the storage of data set unitScene corresponding to data matches, if coupling, to the switching field that loads data display unit output matchingScape information;
Load data display unit, according to the handoff scenario information receiving, obtain from loading data set unitThe loading data that handoff scenario is required are also shown.
Described scene matching module further comprises: interrupt location and loading data pull unit, wherein,
Scene matching unit, is further used at the user's scene handover information receiving and loading data set listWhen the scene corresponding to loading data of unit's storage do not mated, export unmatched handoff scenario to interrupt locationInformation;
Interrupt location, according to the handoff scenario information receiving, the preloading data that interruption is being transmitted,Send a notification message to loading data pull unit;
Load data pull unit, receiving notice message, pulls the required loading of handoff scenario from serverData.
As seen from the above technical solutions, a kind of client prestrain method that the embodiment of the present invention provides andClient pre-load means, adds up and stores the usage behavior characteristic of user under each scene in advance; Obtain useUser's usage behavior characteristic that family current scene is corresponding, calculates and works as front court according to the predicting strategy setting in advanceUnder scape, user respectively operates corresponding prestrain requirements; Obtaining prestrain requirements exceedes predefined pre-Load the required loading data of handoff scenario of the operation correspondence of demand threshold; Monitor user ' scene is switched,Determine that the scene scene corresponding with the loading data of obtaining after switching matches, and shows the loading number obtainingAccording to. Like this, by adding up in advance and store the usage behavior characteristic of user under each scene, according to working as front courtUnder scape, user's usage behavior characteristic is predicted user's subsequent operation, and browses current scene userProcess in obtain prediction the required loading data of scene carry out prestrain, thereby switch in advance userWhile surveying scene, the load time of having reduced client.
Brief description of the drawings
Fig. 1 is the statistics schematic diagram of embodiment of the present invention user usage behavior characteristic.
Fig. 2 is the client prestrain method overall procedure schematic diagram of the embodiment of the present invention.
Fig. 3 is the statistics schematic diagram of embodiment of the present invention time response.
Fig. 4 is embodiment of the present invention scene operation characteristic schematic diagram.
Fig. 5 is the client prestrain method idiographic flow schematic diagram of the embodiment of the present invention.
Fig. 6 is that the embodiment of the present invention is obtained the required loading data flow schematic diagram of handoff scenario.
Fig. 7 is that the embodiment of the present invention is interrupted invalid prestrain schematic flow sheet.
Fig. 8 is the client pre-load means structural representation of the embodiment of the present invention.
Detailed description of the invention
For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing and concreteThe present invention is described in further detail for embodiment.
No matter existing client prestrain method, adopt as required or by experience prestrain mode or figureSheet rolling prestrain mode, in the time that many scenes are switched, client need to pull required loading from serverData, make to load required time longer, and the personalization that can not meet user loads demand. Practical applicationIn, for different scenes, each user's usage behavior is different, user stops in different sceneThe time of staying may be different, and the probability that is switched to different scenes from a scene is also notThe same. In the embodiment of the present invention, consider the user usage behavior of user in different scenes, user existsTime of staying (browsing time) of scene and carry out different operatings in different scenes and jump to corresponding sceneProbability, a kind of client prestrain method based on statistical forecast is proposed, the user who obtains according to statisticsIn average and the mean square deviation of browsing time of different scenes, and carry out different operating redirect in different scenesTo the probability of corresponding scene, predictive user is carried out a certain operation and is entered another scene under current sceneProbability, judges whether to utilize the browsing time of user in current scene with this, to entering next sceneDesired data carries out prestrain, obtains in the time entering next scene so that obtain user without waiting for from serverThe time of fetching data.
The scene described in the embodiment of the present invention that it should be noted that switch, comprising: redirect between the page,Between different scenes call or switching, scene in different little scenes switching (for example, tab label switch,Upper nextpage switching etc.), the switching of picture browsing and redirect of link information etc.
Fig. 1 is the statistics schematic diagram of embodiment of the present invention user usage behavior characteristic. Referring to Fig. 1, userUsage behavior characteristic comprises: time response and operating characteristic, wherein, when time response stops with userBetween represent, the probability that operating characteristic switches to another scene with user from a scene represents, each scene canWith a corresponding Website page.
Through statistics, user, under scene C1, stops the time that reaches 20s for browsing scene C1In content. Meanwhile, under scene C1, there is respectively the probability executable operations Q up to 60%1, fromScene C1 jumps to scene C2, has 20% probability executable operations Q2, show up from scene C1 redirectScape C3, there is 10% probability executable operations Q3, from scene C1 jump to scene C4 and, 10%Probability executable operations Q4, from scene C1, exit;
Under scene C2, stop the time of 10s for browsing the content of scene C2;
Under scene C3, stop the time of 5s for browsing the content of scene C3;
Under scene C4, stop the time of 2s for browsing the content of scene C4.
In the embodiment of the present invention, consider in the time jumping to scene C2 from scene C1, can be by from sceneC1 jumps to the needed loading data of scene C2, browses in the process of scene C1 user, passes throughUser end to server sends data acquisition request, and server response data is obtained request, by neededLoad data (the needed loading data of scene C2) and transfer to client, or trigger and seal from buffer memoryNeeded loading data when dress jumps to scene C2 from scene C1. Like this, by this prestrainMode, in the time that user will enter scene C2 from scene C1, can be without waiting pending data to load, directlyShow the data of scene C2.
Fig. 2 is the client prestrain method overall procedure schematic diagram of the embodiment of the present invention. Referring to Fig. 2,This flow process comprises:
Step 201, adds up and stores the usage behavior characteristic of user under each scene in advance;
In this step, make the technical scheme of the embodiment of the present invention can realize Accurate Prediction and makeThe prestrain decision-making of effect, first needs user to add up in the usage behavior characteristic of client.
User's usage behavior characteristic comprises: time response and operating characteristic, wherein,
Time response mainly comprises: the mean value of user's time of staying in different scenesAnd mean square deviationσ. Wherein,For reflecting that user is in the average characteristics of the scene time of staying, and σ is used for reflecting useThe stability of family time of staying in the time of different time access Same Scene. Certainly,, in practical application, also canWith according to the different of application and the information type that can obtain, corresponding change time response comprise inHold.
Fig. 3 is the statistics schematic diagram of embodiment of the present invention time response. Referring to Fig. 3, obtain user and enterScene C1～Cn, counting user enters scene C for the l timeiTime of staying Tl, according to TlObtaining user entersEnter scene CiThe mean value of the time of stayingAnd meansquaredeviationσi, wherein, n, l, i are natural number,1≤i≤n，
In formula,
M enters scene C for useriTotal degree.
Operating characteristic is for representing that user switches to the probability of another scene from a scene, mainly comprise userAt scene CiUnder carried out operation QkThereby, be switched to another scene CjProbability, can be expressed asp(Qk，Cj/Ci), wherein, k, j is natural number, in practical application, if carry out same under same sceneThe scene that operation is switched to is constant, from scene CiBe switched to scene CjProbability can be expressed asp(Qk/Ci)＝p(Cj/Ci)。
In the time of counting user usage behavior characteristic, can adopt overall statistical, with all usersUsage behavior characteristic be objects of statistics, the statistics obtaining is applied to all users; Also can adoptPersonalized statistical, taking the usage behavior characteristic of unique user as objects of statistics, the statistics knot obtainingFruit is applied to user self; Can also adopt the statistical of both combinations, first pass through global statisticsMode obtains effective overall user's usage behavior statistical property, and using this statistical property result asThe initialization value of all users' usage behavior characteristic, then starts personalized statistical, according to userThe usage behavior of self goes to revise the statistical property result as initialization value. That is to say counting userUsage behavior characteristic comprises:
Taking user as mark, add up respectively each user's usage behavior characteristic;
In this step, can be taking client as statistical unit, it is self-contained that each client is added up respectivelyEach user's usage behavior characteristic, like this, without consumption of network resources; Also can divide in each clientDo not add up after self-contained each user's usage behavior characteristic, each user's usage behavior characteristic is reportedTo server, the same user's that server reports each client usage behavior characteristic averages, and doesFor this user's usage behavior characteristic, and be issued to each client and store, like this, need to consume oneFixed Internet resources, but more can reflect by the usage behavior characteristic that server averages the user who obtainsUser's individual demand.
Or,
Add up all users' usage behavior characteristic, obtain all users' usage behavior characteristic averageValue, as user's usage behavior characteristic.
In this step, after the self-contained all users' of client statistics usage behavior characteristic, will ownThe total usage behavior characteristic of user reports to server, total use row that server reports each clientFor characteristic averages according to number of users, as user's usage behavior characteristic, and be issued to each clientStore.
Step 202, obtains user's usage behavior characteristic corresponding to user's current scene, according to setting in advancePredicting strategy calculate user under current scene and respectively operate corresponding prestrain requirements;
In this step, if current scene is the scene that user browses first, obtaining user's current sceneBefore corresponding user's usage behavior characteristic, further comprise:
Determine that current scene is the scene that user browses first, from pre-stored loading data centralization, looks intoLoading data corresponding to current scene are obtained in inquiry, and show. That is to say, client obtains respectively in statisticsUnder scene after user's usage behavior characteristic, to user first usage behavior characteristic sort, the row of choosingBefore order, preset the scene of number, pull the required loading data of scene of this default number, form and load numberAccording to collecting and storing, after determining that current scene is the scene browsed first of user, from pre-stored loadingData centralization, inquiry is obtained loading data corresponding to current scene and shows. Like this, also can haveEffect reduces the load time of client.
In the time that user browses current scene, need to determine whether need to be required to the scene of the follow-up switching of userData carry out prestrain, carry out the detection of prestrain demand, to reduce user in the time of handoff scenarioThe load time of stand-by period and client.
In the embodiment of the present invention, carry out the detection of prestrain demand according to user's usage behavior characteristic, weIn case, meet the situation of prestrain demand, what obtain according to the score policy calculation setting in advance is currentUnder scene, user's prestrain requirements comprises two aspects: be wherein that user is in current scene on the one handThe time of staying is wanted long enough, if the time of staying is shorter, carries out prestrain the minimizing of load time is doneWith not quite; That the handoff predictions that to be switched scene is carried out has larger confidence level on the other hand, asThe confidence level of fruit handoff predictions is lower, can make the scene of actual switching and the handoff scenario of prediction differCause, cause the unavailable of preloading data, and increased unnecessary network traffics expense.
Want sufficiently long condition for user in the time of staying of current scene, can be current according to what add upThe mean value of the time of staying of sceneMeansquaredeviationσ is determined, for to be switched scene is enteredThe handoff predictions of row has the condition of larger confidence level, can be by carry out different behaviour under this current sceneSwitch to the Probability p (Q of corresponding scenek/Ci) determine, like this, comprehensiveσ and p (Qk/Ci)，Can calculate the different operating that may carry out under current scene user prestrain requirements whetherMeet prestrain demand.
In the embodiment of the present invention, provide two kinds for the treatment of mechanisms of calculating prestrain requirements, i.e. predicting strategyComprise: individual layer predicting strategy and multilayer predicting strategy.
1) individual layer predicting strategy
Individual layer predicting strategy refers to only considers that the time of staying characteristic of user under current scene is (when stopBetween mean value and mean square deviation) and under current scene, carry out all operations and switch to the general of corresponding sceneRate is calculated the prestrain requirements of user under current scene. Under this mechanism, meet prestrain demandScene must meet and can directly switch from current scene.
In the embodiment of the present invention, adopt linear weighted function method to calculate at scene CiLower user's prestrain demandValue, certainly, also can adopt exponential weighting method or other nonlinear weight method to calculate, linearityMethod of weighting computing formula is as follows:
In formula,
pexkThat user is at current scene CiLower executable operations QkSwitch to scene CjPrestrain requirements;
Represent that the user who calculates by mean value weight calculation function is at current scene CiStopStay the mean value weight of time;
wσ(σi) represent that the user who calculates by mean square deviation weight calculation function is at current scene CiStopStay the mean square deviation weight of time,And wσ(σi) can determine according to actual needs;
wo(p(Qk/Ci) represent that the user who calculates by probability right computing function is at current scene CiUnderExecutable operations QkSwitch to scene CjProbability right;
α, β, γ are corresponding mean value weight coefficient, mean square deviation weight coefficient and probability right systemNumber, preferably, can arrange alpha+beta+γ=1.
In practical application, it is also conceivable that the scene (scene after switching) that next time may occur pullingLoad the spent time Estimate value of data as the foundation of calculating prestrain demand, load data spentTime Estimate value can experience given or set by the statistical conditions of previous operation, like this, formula (1)Can be amended as follows:
In formula,
w(Lj) represent working as front court by pulling the user that time Estimate value weight calculation function calculatesScape CiLower executable operations QkAfter will enter scene Cj(scene after switching) corresponding drawing of required loading dataGet time Estimate value weight;
ξ is for pulling accordingly time Estimate value weight coefficient.
LjSize on prestrain demand calculate impact depending on concrete application.
2) multilayer predicting strategy
Multilayer predicting strategy refer to not only consider the time of staying characteristic of user under current scene andUnder current scene, carry out the probability (operating characteristic) that all operations switches to corresponding scene, also look to the futureTime response and the operating characteristic of the scene (scene after switching) that may occur. Under this mechanism, symbolClose the scene of prestrain demand except can reaching from the direct switching of current scene, also comprised and worked asFront scene cannot directly arrive but by the scene scene that may arrive after switching. By above-mentioned formula (1)(2) time response and the operating characteristic of rear scene switched in consideration, modifies respectively, and it is right to be converted intoFormula (3) and (4) of answering, as follows respectively:
In formula,
p(Cj/Ci) represent from current scene CiSwitch to again scene C by the scene after switchingjProbability;
N represents the number of plies of prediction;
λ is decay factor, and 0 < λ < 1 introduces λnIn order to weaken gradually deeper prediction scene quiltThe possibility of prestrain;
Represent from scene CiTo scene CjThe acquisition time by way of the required loading data of sceneThe summation of estimated value.
Step 203, obtaining prestrain requirements, to exceed the operation of predefined prestrain demand threshold rightThe required loading data of handoff scenario of answering;
In this step, obtain the required loading data of handoff scenario, described handoff scenario is prestrain demandValue exceedes the scene of the operation correspondence of predefined prestrain demand threshold, specifically, if:
pexk＞pth, obtain user at current scene CiLower executable operations QkSwitch to scene CjRequiredLoad data.
In formula,
pthIt is predefined prestrain demand threshold.
When prestrain requirements corresponding to arbitrary operation GreatlyIn pthTime, think that it meets the demand of prestrain, pthBe worth greatlyr, represent calculating to prestrain demandStricter, the probability of carrying out prestrain is lower.
Obtain prestrain requirements and exceed the switching of the operation correspondence of predefined prestrain demand thresholdThe required loading data of scene comprise:
Inquire about pre-stored loading data set, if exist the required loading data of handoff scenario and this to addIt is up-to-date carrying data, obtains the required loading data of this handoff scenario from loading data centralization; Otherwise,
Send data acquisition request to server, pull prestrain requirements from server and exceed and presetThe required loading data of handoff scenario of operation correspondence of prestrain demand threshold, and be stored in loading numberAccording to concentrating; Or, sending data preparation request to server, server receives data preparation request, inquiryObtain the required loading data of handoff scenario, and encapsulate to reduce Internet Transmission to loading data as far as possibleThe application of amount.
Determine to load whether data are up-to-date, can be according to the timestamp information of these loading data of storage and currentTimestamp information is determined, if current time stamp is not more than in advance with the difference of the timestamp that loads dataThe time arranging is upgraded threshold value, determines that these loading data are up-to-date. In practical application, can also work asWhen front timestamp is greater than with the difference of the timestamp of loading data the time renewal threshold value setting in advance, clientInitiatively pull corresponding loading data and the loading data of storage are upgraded to server.
Fig. 4 is embodiment of the present invention scene operation characteristic schematic diagram. Referring to Fig. 4, suppose that user enters intoAfter scene C1, there is respectively 60% probability by executable operations Q1Be switched to scene C2, have 20%Probability by executable operations Q2Be switched to scene C3, there is 10% probability by executable operations Q3CutChange to scene C4, there is 10% probability by executable operations Q8Exit; In scene C2, tool respectivelyThere is 90% probability by executable operations Q4Be switched to scene C5, there is 10% probability and grasp by executionMake Q5Be switched to scene C6; In scene C6, there is respectively 50% probability by executable operations Q6CutChange to scene C3, there is 50% probability by executable operations Q7Be switched to scene C4. Like this, can be pre-Survey user under current scene C1, have 54% probability can enter scene C5, have 6% probability to enterEnter scene C6, have 20% probability can enter scene C3, have 10% probability can enter scene C4,There is 10% probability to exit, if α=β=0, predefined prestrain demand threshold are 50%, logicalCross employing multilayer prediction (it is 2 that the maximum number of plies is set), so when user enters after scene C1, sceneC2 and C5 all can meet prestrain demand.
Step 204, monitor user ' scene is switched, and determines scene and the loading data pair of obtaining after switchingThe scene of answering matches, and shows the loading data of obtaining.
In this step, if user is by executable operations, current scene is switched, obtained executionOperate the scene after corresponding switching, and judge the scene field corresponding with the loading data of obtaining after switchingWhether scape matches, if coupling is shown the loading data that pull.
In the embodiment of the present invention, the loading data of obtaining comprise: pre-stored adding in loading data centralizationCarry the loading data that data and step 203 are obtained.
As previously mentioned, in step 203, if send data preparation request to server, showThe loading data of obtaining comprise:
Send data acquisition request to server, server receives data acquisition request, by the loading of encapsulationData are sent to client and show.
This step further comprises:
Determine that the scene scene corresponding with the loading data of obtaining after switching do not match, and interrupts enteringThe preloading data of row transmission, required loading data or the interruption of scene from server pulls switchingThe preloading data that the preloading data transmitting and deletion have completed, pulls from serverThe required loading data of scene after switching.
Further, the method also comprises:
Step 205, determines that user exits scene, switches under each scene of upgrading storage according to user's sceneUser's usage behavior characteristic.
In this step, in the time that user exits scene, time response and operation according to user in each sceneCharacteristic is upgraded respectively the user's usage behavior characteristic under the corresponding scene of storage, for example, recalculates stopThe mean value of time and mean square deviation.
Fig. 5 is the client prestrain method idiographic flow schematic diagram of the embodiment of the present invention. Referring to Fig. 5,This flow process comprises:
Step 501, enters new scene;
Step 502, carries out the detection of prestrain demand;
In this step, obtain user's usage behavior characteristic that new scene is corresponding, according to user's usage behavior spyProperty is calculated user under new scene and is respectively operated corresponding prestrain requirements.
Step 503, judges whether to trigger prestrain, if so, and execution step 504, otherwise, carry outStep 521;
In this step, the prestrain demand obtaining according to detection, under the new scene calculating, user is eachOperate corresponding prestrain requirements, judge whether to meet predefined prestrain demand threshold, ifMeet, trigger prestrain.
Step 504, starts prestrain;
Step 505, judges whether user carries out predicted operation, if so, and execution step 506, otherwise,Execution step 511;
In this step, user can be arranged on when scene is switched whether carry out predicted operation, if carry out pre-Survey operation, browse user that in the process of current scene, to pull the corresponding scene of predicted operation from server requiredLoading data and storage.
Step 506, obtains the loading data that need from the loading data centralization of storage;
Step 507, enters next new scene, returns to execution step 502;
In this step, client, according to user's operation, is shown the loading data of obtaining to user.
Further, client is obtained the link information (scene) of user's operation correspondence, and obtainsLoad link information (scene) corresponding to data and mate, if the match is successful, by the loading of obtainingData show to user, if mate unsuccessfully, process according to existing procedure, to serverSend request, pull user and operate the required loading data of corresponding link information.
Step 511, starts and interrupts invalid prestrain;
In this step, if user does not carry out predicted operation, the operation that user carries out and the operation of predictionInconsistent, to need interruption transmitting preloading data, i.e. invalid prestrain.
Step 512 is obtained the loading data that need from server or buffer memory;
Step 513, enters next new scene, returns to execution step 502;
In this step, the scene that the new scene entering is corresponding from predicted operation is different.
Step 521, executable operations, carries out scene switching;
Step 522, enters next new scene, returns to execution step 502.
To obtaining prestrain requirements, to exceed the operation of predefined prestrain demand threshold right more belowThe required loading data of handoff scenario of answering are elaborated.
Fig. 6 is that the embodiment of the present invention is obtained the required loading data flow schematic diagram of handoff scenario. Referring to figure6, this flow process comprises:
Step 601, obtains a new task from prestrain task waiting list;
In this step, if exist multiple prestrain requirements to exceed predefined prestrain demand thresholdValue, corresponding multiple handoff scenario, multiple handoff scenario composition prestrain task waiting lists, often allThe corresponding new task of carry over scape.
Step 602, judgement completed in preloaded list or buffer memory (loading data set) in whether have thisThe loading data that new task is required, if so, execution step 603, otherwise, execution step 604;
Step 603, encapsulates and returns the loading data that need, execution step 605;
Step 604, request server returns to the required loading data of this new task;
Step 605, adds prestrain to complete list the loading data of returning;
Step 606, judges whether prestrain task waiting list is empty, if so, finishes this flow process,Otherwise, return to execution step 601.
Describe interrupting invalid prestrain below.
When the operation of operation and the prediction of carrying out as user is inconsistent, the data of previously carrying out prestrain are possibleWithout any effect, in the time that user operates handoff scenario, may occur two to the scene when advancing intoKind of situation: the data of a front prestrain are also being transmitted and the biography of the data of a front prestrainDefeatedly complete. For the first situation, if the data of a front prestrain are not carried out to any intervention, client need to be waited for after the transfer of data of a front prestrain completes and reloading when advancing into sceneNeeded loading data. And for the second situation, although can not advance into scene desired data to working asLoading time delay exert an influence, but the data of a front prestrain may be brought two kinds of problems: takeCertain memory headroom and comprise dirty data, dirty data refers to physically and exists, but do not deposit in logicData, for instance, in prestrain process, front end has carried out insertion, deletion or has upgraded operation,The data of buffer memory are changed, but be not also written to the data in disk or data file.
Fig. 7 is that the embodiment of the present invention is interrupted invalid prestrain schematic flow sheet. Referring to Fig. 7, it is invalid to interruptPrestrain comprises: interrupts ongoing invalid prestrain and deletes the data that completed prestrain,This flow process comprises:
Step 701, has judged whether ongoing prestrain task, if had, and execution step 702,Otherwise, execution step 703;
Step 702, determines the prestrain task that needs interruption in the task of carrying out prestrain, and willNeed the prestrain tasks carrying interrupting to interrupt processing, execution step 703;
In this step, need the prestrain task of interrupting meet following arbitrary condition:
Condition 1: the prestrain task residue load time that this need to interrupt is long, has blocked and has been about to enterThe loading of scene desired data;
In the embodiment of the present invention, can be by the task of all beginning prestrains is added to initial time stamp,In the time that needs interrupt this prestrain task, calculate this prestrain task load time, according to this prestrainThe estimated value of total load time of task estimates the residue load time of this prestrain task, if estimationThe residue load time has exceeded predefined load time threshold value,, by this prestrain tasks interrupt, addsCarrying time threshold can set according to user's demand for experience of different application.
Condition 2: the data that the prestrain task that this need to interrupt is loading, for the field that is about to enterScape needs the data possibility of prestrain lower.
In the embodiment of the present invention, using the scene that is about to enter as current scene, and adopt above-mentioned prestrainWhether the prestrain demand weight that demand computing formula is obtained the data that this task loading exceedes in advanceThe prestrain demand threshold p settingthIf do not had, by this prestrain tasks interrupt.
Step 703, determines the data that need deletion in the data that completed prestrain, and by its deletion.
In this step, the data of prestrain are completed for some, although can be to not being about to enter sceneData load impact, but may bring two kinds of problems: one, has taken certain internal memory skyBetween; Its two, comprise dirty data, for example, in prestrain process, front end carried out insertion, deletion orUpgrade operation, the data of server end storage are changed, and the data that completed prestrain alsoDo not carry out corresponding variation.
Delete and need the data of deleting to comprise:
Step 1, deletes the dirty data that has completed prestrain;
In this step, if can be by the dirty data that completes prestrain is inserted accordingly, deletedRemove or upgrade operation, without deletion.
Step 2, adopts above-mentioned prestrain demand computing formula to obtain to complete data pre-of prestrainWhether loading demand weight exceedes predefined load time threshold value, if exceeded, retains data,If do not exceeded, perform step 3;
Step 3, assesses the current shared memory headroom of data that has completed prestrain and whether exceedes tolerance limitDegree (rule of thumb set), if exceeded, deletes all prestrain demand weights and does not exceed in advance and establishThe data of fixed load time threshold value.
From above-mentioned, the client prestrain method of the embodiment of the present invention, by adding up in advance and storingUser's usage behavior characteristic under each scene, according to the usage behavior characteristic of user under current scene to userSubsequent operation predicts, and it is required in user browses the process of current scene, to obtain the scene of predictionLoad data and carry out prestrain, thereby in the time that user switches to prediction scene, without obtaining phase from service areaShould load data, reduce the load time of client, thereby improve user's experience; Further,Carry out prestrain according to user's usage behavior characteristic, can accurately meet user use habit, meetUser's individual demand; And, can be applied to the prestrain under many scenes switchings.
Fig. 8 is the client pre-load means structural representation of the embodiment of the present invention. Referring to Fig. 8, this dressPut and comprise: usage behavior statistics of features module, prestrain requirements computing module, preloading data obtainModule and scene matching module, wherein,
Usage behavior statistics of features module, adds up and stores the usage behavior characteristic of user under each scene;
Prestrain requirements computing module, obtains user's usage behavior characteristic corresponding to user's current scene,Calculate user under current scene according to the predicting strategy setting in advance and respectively operate corresponding prestrain requirements;
Preloading data acquisition module, obtains prestrain requirements and exceedes predefined prestrain demand thresholdThe required loading data of handoff scenario of the operation correspondence of value;
Scene matching module, monitor user ' scene is switched, and determines scene and the loading number obtaining after switchingMatch according to corresponding scene, show the loading data of obtaining.
Prestrain requirements computing module comprises: mean value weight calculation unit, mean square deviation weight calculation listUnit, probability right computing unit and prestrain requirements computing unit (not shown), wherein,
Mean value weight calculation unit, calculates user by the mean value weight calculation function setting in advance and existsThe mean value weight of the time of staying of current scene;
Mean square deviation weight calculation unit, calculates user by the mean square deviation weight calculation function setting in advance and existsThe mean square deviation weight of the time of staying of current scene;
Probability right computing unit, calculates user current by the probability right computing function setting in advanceUnder scene, executable operations switches to the probability right of handoff scenario;
Prestrain requirements computing unit, the mean value weight that mean value weight calculation unit is calculatedMultiply each other with mean value weight coefficient, the mean square deviation weight that mean square deviation weight calculation unit is calculated is with equalVariance weight coefficient multiplies each other, the probability right that probability right computing unit is calculated and probability right systemNumber multiplies each other, and by the long-pending addition of respectively multiplying each other.
Scene matching module comprises: monitoring means, scene matching unit, loading data set unit, loadingData display unit, interrupt location and loading data pull unit (not shown), wherein,
Monitoring means, monitor user ' scene is switched, and exports the user's scene handover information monitoring to fieldScape matching unit;
Scene matching unit, receives user's scene handover information, with the loading that loads the storage of data set unitScene corresponding to data matches, if coupling, to the switching field that loads data display unit output matchingScape information, if do not mated, exports unmatched handoff scenario information to interrupt location;
Load data display unit, according to the handoff scenario information receiving, obtain from loading data set unitThe loading data that handoff scenario is required are also shown;
Interrupt location, according to the handoff scenario information receiving, the preloading data that interruption is being transmitted,Send a notification message to loading data pull unit;
Load data pull unit, receiving notice message, pulls the required loading of handoff scenario from serverData.
The foregoing is only preferred embodiment of the present invention, be not intended to limit protection model of the present inventionEnclose. Within the spirit and principles in the present invention all, any amendment of doing, be equal to and replace and improvement etc.,Within all should being included in protection scope of the present invention.
Claims (13)
1. a client prestrain method, is characterized in that, the method comprises:
Obtain in advance the usage behavior characteristic of user under each scene; Described user's usage behavior characteristic bagDraw together: time response and operating characteristic; Described time response comprises: when user stops in different scenesBetween mean value and mean square deviation; Described operating characteristic comprises that user switches to another scene from a sceneProbability;
Obtain user corresponding to user's current scene usage behavior characteristic, according to the prediction plan setting in advanceSlightly, according to user corresponding to described user's current scene usage behavior characteristic, calculate under current scene and useFamily respectively operates corresponding prestrain requirements;
Obtain the required loading data of handoff scenario, described handoff scenario is that prestrain requirements exceedes in advanceThe scene of the operation correspondence of the prestrain demand threshold of setting;
Monitor user ' scene is switched, and determines the scene scene phase corresponding with the loading data of obtaining after switchingCoupling, shows the loading data of obtaining.
2. the method for claim 1, is characterized in that, described in obtain under the each scene of user and makeComprise with behavioral trait:
Taking user as mark, add up respectively each user's usage behavior characteristic; Or
Add up all users' usage behavior characteristic, obtain all users' usage behavior characteristic averageValue, as user's usage behavior characteristic.
3. the method for claim 1, is characterized in that, in the described user's current scene of obtainingBefore corresponding user's usage behavior characteristic, further comprise:
Client to user first usage behavior characteristic sort, choose the field of default number before sequenceScape, pulls the required loading data of scene of this default number, forms and loads data set storage;
Determine that current scene is the scene that user browses first, from pre-stored loading data centralization, looks intoInquiry is obtained loading data corresponding to current scene and is shown.
4. the method for claim 1, is characterized in that, the described prediction according to setting in advanceUnder policy calculation current scene, user respectively operates corresponding prestrain requirements and comprises:
Calculate the time of staying of user in current scene by the mean value weight calculation function setting in advanceMean value weight, multiply each other with mean value weight coefficient;
Calculate the time of staying of user in current scene by the mean square deviation weight calculation function setting in advanceMean square deviation weight, multiply each other with mean square deviation weight coefficient;
Calculating user's executable operations under current scene by the probability right computing function setting in advance cutsChange the probability right to handoff scenario, with probability right multiplication; Or by the probability setting in advanceWeight calculation function calculates user and under current scene, passes through the scene probability of handoff scenario again after switchingWeight, multiplies each other with the number of plies power of the prediction of probability right coefficient and decay factor; And
The long-pending addition of respectively multiplying each other is obtained to prestrain requirements corresponding to user's executable operations under current scene.
5. method as claimed in claim 4, is characterized in that, in described and probability right coefficient phaseAfter the step of taking advantage of, further comprise:
Calculate user under current scene by the time Estimate value weight calculation function that pulls setting in advanceAfter executable operations, switch to the required loading data of handoff scenario corresponding pull time Estimate value weight, and drawGet time Estimate value weight coefficient and multiply each other, and carry out the step of the long-pending addition of respectively multiplying each other;
The step multiplying each other at the number of plies power of the prediction of described and probability right coefficient and decay factor itAfter, further comprise:
Current scene is estimated to the acquisition time by way of the required loading data of scene of last handoff scenarioThe summation of value, with the number of plies power phase of prediction that pulls time Estimate value weight coefficient and decay factorTake advantage of, and carry out the step of the long-pending addition of respectively multiplying each other.
6. the method as described in claim 1 to 5 any one, is characterized in that, described in obtain switchingThe required loading data of scene comprise:
Inquire about pre-stored loading data set, if exist the required loading data of handoff scenario and this to addIt is up-to-date carrying data, obtains the required loading data of this handoff scenario from loading data centralization; Otherwise,
Send data acquisition request to server, pull prestrain requirements from server and exceed and presetThe required loading data of handoff scenario of operation correspondence of prestrain demand threshold, and be stored in loading numberAccording to concentrating; Or, sending data preparation request to server, server receives data preparation request, inquiryObtain the required loading data of handoff scenario, and encapsulate loading data.
7. method as claimed in claim 6, is characterized in that, further comprises:
Be greater than in current time stamp and the difference of the timestamp that loads data the time renewal threshold value setting in advanceTime, client initiatively pulls corresponding loading data and the loading data of storage is carried out more to serverNewly.
8. method as claimed in claim 6, is characterized in that, further comprises:
Determine that the scene scene corresponding with the loading data of obtaining after switching do not match, and interrupts enteringThe preloading data of row transmission, the required loading data of scene from server pulls switching.
9. method as claimed in claim 6, is characterized in that, further comprises:
Determine that user exits scene, switch the use of user under each scene of upgrading storage according to user's sceneBehavioral trait.
10. a client pre-load means, is characterized in that, this device comprises: usage behavior characteristicStatistical module, prestrain requirements computing module, preloading data acquisition module and scene matching module,Wherein,
Usage behavior statistics of features module, obtains the usage behavior characteristic of user under each scene; Described userUsage behavior characteristic comprise: time response and operating characteristic; Described time response comprises: Yong HuMean value and the mean square deviation of the time of staying in different scenes; Described operating characteristic comprises that user is from a sceneSwitch to the probability of another scene;
Prestrain requirements computing module, obtains user corresponding to user's current scene usage behavior spyProperty, according to the predicting strategy setting in advance, according to user corresponding to described user's current scene use rowFor characteristic, calculate user under current scene and respectively operate corresponding prestrain requirements;
Preloading data acquisition module, obtains the required loading data of handoff scenario, and described handoff scenario isPrestrain requirements exceedes the scene of the operation correspondence of predefined prestrain demand threshold;
Scene matching module, monitor user ' scene is switched, and determines scene and the loading number obtaining after switchingMatch according to corresponding scene, show the loading data of obtaining.
11. devices as claimed in claim 10, is characterized in that, described prestrain requirements calculatesModule comprises: mean value weight calculation unit, mean square deviation weight calculation unit, probability right computing unitAnd prestrain requirements computing unit, wherein,
Mean value weight calculation unit, calculates user by the mean value weight calculation function setting in advance and existsThe mean value weight of the time of staying of current scene;
Mean square deviation weight calculation unit, calculates user by the mean square deviation weight calculation function setting in advance and existsThe mean square deviation weight of the time of staying of current scene;
Probability right computing unit, calculates user current by the probability right computing function setting in advanceUnder scene, executable operations switches to the probability right of handoff scenario;
Prestrain requirements computing unit, the mean value weight that mean value weight calculation unit is calculatedMultiply each other with mean value weight coefficient, the mean square deviation weight that mean square deviation weight calculation unit is calculated is with equalVariance weight coefficient multiplies each other, the probability right that probability right computing unit is calculated and probability right systemNumber multiplies each other, and by the long-pending addition of respectively multiplying each other.
12. devices as described in claim 10 or 11, is characterized in that described scene matching moduleComprise: monitoring means, scene matching unit, loading data set unit and loading data display unit,Wherein,
Monitoring means, monitor user ' scene is switched, and exports the user's scene handover information monitoring to fieldScape matching unit;
Scene matching unit, receives user's scene handover information, with the loading that loads the storage of data set unitScene corresponding to data matches, if coupling, to the switching field that loads data display unit output matchingScape information;
Load data display unit, according to the handoff scenario information receiving, obtain from loading data set unitThe loading data that handoff scenario is required are also shown.
13. devices as claimed in claim 12, is characterized in that, described scene matching module enters oneStep comprises: interrupt location and loading data pull unit, wherein,
Scene matching unit, is further used at the user's scene handover information receiving and loading data set listWhen the scene corresponding to loading data of unit's storage do not mated, export unmatched handoff scenario to interrupt locationInformation;
Interrupt location, according to the handoff scenario information receiving, the preloading data that interruption is being transmitted,Send a notification message to loading data pull unit;
Load data pull unit, receiving notice message, pulls the required loading of handoff scenario from serverData.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210025953.4A CN103246526B (en) | 2012-02-07 | 2012-02-07 | Client prestrain method and client pre-load means |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210025953.4A CN103246526B (en) | 2012-02-07 | 2012-02-07 | Client prestrain method and client pre-load means |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103246526A CN103246526A (en) | 2013-08-14 |
| CN103246526B true CN103246526B (en) | 2016-05-04 |
Family
ID=48926060
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210025953.4A Active CN103246526B (en) | 2012-02-07 | 2012-02-07 | Client prestrain method and client pre-load means |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103246526B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107133094A (en) * | 2017-06-05 | 2017-09-05 | 努比亚技术有限公司 | Application management method, mobile terminal and computer-readable recording medium |
| CN107679176A (en) * | 2017-09-29 | 2018-02-09 | 努比亚技术有限公司 | It is a kind of to improve using the method for opening speed, terminal, server, computer-readable recording medium |
Families Citing this family (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103455353B (en) * | 2013-08-30 | 2017-11-14 | 广州市动景计算机科技有限公司 | The requesting method and device of web data |
| CN104657372B (en) | 2013-11-20 | 2018-11-06 | 腾讯科技（深圳）有限公司 | page operation data processing method and device |
| CN103729558A (en) * | 2013-12-26 | 2014-04-16 | 北京像素软件科技股份有限公司 | Scene change method |
| CN104765746B (en) * | 2014-01-06 | 2020-02-07 | 腾讯科技（深圳）有限公司 | Data processing method and device for mobile communication terminal browser |
| CN105786839A (en) * | 2014-12-22 | 2016-07-20 | 阿里巴巴集团控股有限公司 | Application data acquisition method and apparatus |
| CN104753922A (en) * | 2015-03-10 | 2015-07-01 | 百度在线网络技术（北京）有限公司 | Method for pre-loading, server side, client and system |
| CN104794004B (en) * | 2015-03-17 | 2018-09-04 | 中国石油天然气集团公司 | The method that information preloads |
| CN104750824A (en) * | 2015-03-31 | 2015-07-01 | 北京百度网讯科技有限公司 | Application functional data processing method and device |
| CN105260406A (en) * | 2015-09-23 | 2016-01-20 | 北京金山安全软件有限公司 | Method and device for determining loading speed of browser |
| CN106610887B (en) * | 2015-10-26 | 2019-02-22 | 北京国双科技有限公司 | The determination method and apparatus of webpage loading velocity |
| CN106202368B (en) * | 2016-07-07 | 2021-11-23 | 贵州白山云科技股份有限公司 | Preloading method and device |
| CN106791911A (en) * | 2016-12-28 | 2017-05-31 | 北京风行在线技术有限公司 | A kind of loading method of picture and video |
| CN106708583A (en) * | 2016-12-30 | 2017-05-24 | 广州华多网络科技有限公司 | Application loading method and device |
| CN107203629B (en) * | 2017-05-31 | 2021-02-26 | 北京京东尚科信息技术有限公司 | Page presentation method, system and device |
| CN107229397A (en) * | 2017-06-08 | 2017-10-03 | 惠州Tcl移动通信有限公司 | A kind of method, system, terminal and storage device for improving terminal fluency |
| CN110020133B (en) * | 2017-11-07 | 2023-04-07 | 腾讯科技（深圳）有限公司 | Content recommendation processing method and device, computer equipment and storage medium |
| CN108280168B (en) * | 2018-01-19 | 2022-03-08 | 中国科学院上海高等研究院 | Webpage processing method/system, computer readable storage medium and electronic device |
| CN108595230B (en) * | 2018-05-15 | 2020-03-27 | Oppo广东移动通信有限公司 | Application program preloading method and device, storage medium and terminal |
| CN109144357B (en) * | 2018-06-08 | 2019-07-02 | 北京小蓦机器人技术有限公司 | Command processing method, device, equipment and computer readable storage medium |
| CN109033147B (en) * | 2018-06-12 | 2021-10-22 | 网宿科技股份有限公司 | Data display method, terminal and computer storage medium |
| CN109445581B (en) * | 2018-10-17 | 2021-04-06 | 北京科技大学 | Large-scale scene real-time rendering method based on user behavior analysis |
| CN110347491B (en) * | 2019-07-15 | 2020-08-25 | 贵阳动视云科技有限公司 | Application processing method, device and medium |
| CN111275607B (en) * | 2020-01-17 | 2022-05-24 | 腾讯科技（深圳）有限公司 | Interface display method and device, computer equipment and storage medium |
| CN111835833A (en) * | 2020-06-23 | 2020-10-27 | 武汉菲奥达物联科技有限公司 | LPWAN-based scene fast switching method and device |
| CN115016855B (en) * | 2021-11-17 | 2023-05-09 | 荣耀终端有限公司 | Method, apparatus and storage medium for application preloading |
| CN114827646B (en) * | 2022-03-23 | 2023-12-12 | 百果园技术(新加坡)有限公司 | Method, device, equipment and storage medium for preloading live broadcasting room in video stream |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101656869A (en) * | 2008-08-21 | 2010-02-24 | 华为技术有限公司 | Method, device and system for switching channels |
| CN101754437A (en) * | 2008-12-19 | 2010-06-23 | 英特尔公司 | Handling sensors in a context aware platform |
| CN102033935A (en) * | 2010-12-20 | 2011-04-27 | 深圳市五巨科技有限公司 | Preloading method, preloading device and preloading system of mobile terminal browser |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20010049615A1 (en) * | 2000-03-27 | 2001-12-06 | Wong Christopher L. | Method and apparatus for dynamic business management |
| US7984000B2 (en) * | 2007-10-31 | 2011-07-19 | Microsoft Corporation | Predicting and using search engine switching behavior |
-
2012
- 2012-02-07 CN CN201210025953.4A patent/CN103246526B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101656869A (en) * | 2008-08-21 | 2010-02-24 | 华为技术有限公司 | Method, device and system for switching channels |
| CN101754437A (en) * | 2008-12-19 | 2010-06-23 | 英特尔公司 | Handling sensors in a context aware platform |
| CN102033935A (en) * | 2010-12-20 | 2011-04-27 | 深圳市五巨科技有限公司 | Preloading method, preloading device and preloading system of mobile terminal browser |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107133094A (en) * | 2017-06-05 | 2017-09-05 | 努比亚技术有限公司 | Application management method, mobile terminal and computer-readable recording medium |
| CN107133094B (en) * | 2017-06-05 | 2021-11-02 | 努比亚技术有限公司 | Application management method, mobile terminal and computer readable storage medium |
| CN107679176A (en) * | 2017-09-29 | 2018-02-09 | 努比亚技术有限公司 | It is a kind of to improve using the method for opening speed, terminal, server, computer-readable recording medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103246526A (en) | 2013-08-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103246526B (en) | Client prestrain method and client pre-load means | |
| US10332009B2 (en) | Predicting user navigation events | |
| US10896285B2 (en) | Predicting user navigation events | |
| JP4746712B2 (en) | Calculate document importance by historical importance factoring | |
| US10026103B2 (en) | Time-triggered advertisement replacement | |
| EP2756422B1 (en) | Predicting user navigation events | |
| TWI492175B (en) | Information processing apparatus, information processing method, information processing program product and recording medium with information processing program | |
| RU2720954C1 (en) | Search index construction method and system using machine learning algorithm | |
| US20100082637A1 (en) | Web Page and Web Site Importance Estimation Using Aggregate Browsing History | |
| US8566332B2 (en) | Populating variable content slots on web pages | |
| CN104584008A (en) | Heuristic caching to personalize applications | |
| CN109274987A (en) | A kind of video collection sort method, server and readable storage medium storing program for executing | |
| Yang et al. | Analysis of M/G/1 stochastic clearing systems | |
| Badescu et al. | On the analysis of a multi-threshold Markovian risk model | |
| CN104283934A (en) | WEB service pushing method and device based on reliability prediction and server | |
| JP2020154880A (en) | Calculation device, calculation method, and calculation program | |
| JP2007503651A5 (en) | ||
| JP7010870B2 (en) | Decision device, decision method and decision program | |
| CN114358850A (en) | Intelligent advertisement putting management method and related device | |
| EP2830290B1 (en) | Communication of Pages via a Digital Cellular Telecommunication Network | |
| US20210334734A1 (en) | Method and system for presenting digital task implemented in computer-implemented crowd-sourced environment | |
| JP5428918B2 (en) | Document distribution device | |
| KR20060060783A (en) | Method for providing a search service and system for executing the method | |
| CN116186366A (en) | Web crawler scheduling method for associating user behavior hotness | |
| Lu et al. | Performance–energy trade-off in controllable servers with set-up delays |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |