CN107329882A - Obtain method and apparatus, storage medium and the electronic installation of online data - Google Patents
Obtain method and apparatus, storage medium and the electronic installation of online data Download PDFInfo
- Publication number
- CN107329882A CN107329882A CN201710439790.7A CN201710439790A CN107329882A CN 107329882 A CN107329882 A CN 107329882A CN 201710439790 A CN201710439790 A CN 201710439790A CN 107329882 A CN107329882 A CN 107329882A
- Authority
- CN
- China
- Prior art keywords
- period
- line duration
- online data
- duration section
- section
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/34—Recording or statistical evaluation of computer activity, e.g. of down time, of input/output operation ; Recording or statistical evaluation of user activity, e.g. usability assessment
- G06F11/3466—Performance evaluation by tracing or monitoring
- G06F11/3476—Data logging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3003—Monitoring arrangements specially adapted to the computing system or computing system component being monitored
- G06F11/302—Monitoring arrangements specially adapted to the computing system or computing system component being monitored where the computing system component is a software system
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/30—Monitoring
- G06F11/3089—Monitoring arrangements determined by the means or processing involved in sensing the monitored data, e.g. interfaces, connectors, sensors, probes, agents
- G06F11/3093—Configuration details thereof, e.g. installation, enabling, spatial arrangement of the probes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2201/00—Indexing scheme relating to error detection, to error correction, and to monitoring
- G06F2201/865—Monitoring of software
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- Mathematical Physics (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The invention discloses a kind of method and apparatus, storage medium and electronic installation for obtaining online data.Wherein, this method includes:Receive and obtain request, wherein, obtaining request is used for target online data of the acquisition request intended application in the first line duration section, and the first line duration section is later than the request time that generation obtains request;In the case where the first line duration section is the particular type period, according to the acquisition strategy matched with the particular type period, the history online data in the second line duration section corresponding with the first line duration section is obtained, wherein, the second line duration section is earlier than request time;The target online data in the first line duration section is obtained using the history online data in the second line duration section.The present invention solves the relatively low technical problem of the data accuracy got in the presence of current available data technology.
Description
Technical field
The present invention relates to computer realm, it is situated between in particular to a kind of method and apparatus for obtaining online data, storage
Matter and electronic installation.
Background technology
Nowadays, many application developers generally require the history online data by obtaining terminal applies, and are gone through to above-mentioned
History online data is analyzed, to predict online data that above-mentioned terminal applies are following, in order to according to predicting in line number
Improve according to terminal applies are carried out with correspondence adjustment, so that realizing that the more users of attraction download uses above-mentioned terminal applies, reach
Expand the purpose of the scope of application of above-mentioned terminal applies.
Wherein, the method that prior art provides many acquisition online datas, mainly including following two:
1) GBDT (Gradient Boosting Decision Tree), GBDT algorithms are generally used for predicting each user
Whether can log in, that is to say, that carry out online data prediction using GBDT algorithms, be only capable of prediction and obtain whether a certain user can step on
Record, and be difficult predict the specific login time of user, i.e., it is unpredictable go out accurate online data;In addition, this algorithm meter
Calculation amount is huge, in addition it is also necessary to substantial amounts of computing resource.
2) Arima (Auto Regressive integrated Moving Average), Arima models are generally used for
The serializing online data of terminal applies is obtained, then tranquilization (stationary) sequence, is AR (Auto respectively
Regressive) converted with MA (Moving Average).However, for some terminal applies because data variation is more frequent,
Sequence stationary is difficult to ensure that, further, the problem of by leading to not by predicting to get accurate online data.
For it is above-mentioned the problem of, effective solution is not yet proposed at present.
The content of the invention
The embodiments of the invention provide it is a kind of obtain online data method and apparatus, storage medium and electronic installation, with
At least solve the relatively low technical problem of the data accuracy got in the presence of current available data technology.
One side according to embodiments of the present invention there is provided a kind of target online data acquisition methods, including:Reception is obtained
Request is taken, wherein, the target online data in the first line duration section for acquisition request intended application is asked in above-mentioned acquisition,
Above-mentioned first line duration section is later than the request time for generating above-mentioned acquisition request;It is certain kinds in above-mentioned first line duration section
In the case of the type period, according to the acquisition strategy matched with the above-mentioned particular type period, obtain and above-mentioned first is online
History online data in period corresponding second line duration section, wherein, above-mentioned second line duration section is asked earlier than above-mentioned
Seeking time;Obtain upper in above-mentioned first line duration section using the above-mentioned history online data in above-mentioned second line duration section
State target online data.
Another aspect according to embodiments of the present invention, additionally provides a kind of device for obtaining online data, including:Receive single
Member, request is obtained for receiving, wherein, above-mentioned acquisition is asked for acquisition request intended application in the first line duration section
Target online data, above-mentioned first line duration section is later than the request time for generating above-mentioned acquisition request;First acquisition unit, is used
In the case of being the particular type period in above-mentioned first line duration section, match according to the above-mentioned particular type period
Acquisition strategy, obtain with the history online data in above-mentioned first line duration section corresponding second line duration section, wherein,
Above-mentioned second line duration section is earlier than above-mentioned request time;Second acquisition unit, for using in above-mentioned second line duration section
Above-mentioned history online data obtain above-mentioned target online data in above-mentioned first line duration section.
Another aspect according to embodiments of the present invention, additionally provides a kind of storage medium, and above-mentioned storage medium includes storage
Program said procedure operation when perform above-mentioned acquisition online data method.
Another aspect according to embodiments of the present invention, additionally provides a kind of electronic installation, including memory, processor and deposits
Store up the computer program that can be run on above-mentioned memory and on above-mentioned processor, it is characterised in that above-mentioned processor passes through
The method that above computer program performs above-mentioned acquisition online data.
In embodiments of the present invention, it is determined that first where obtaining the target online data of request institute acquisition request is online
After the type of period, according to the acquisition strategy with the type matching of the first line duration section, obtain and the first line duration
History online data in the corresponding second line duration section of section, during predicting that acquisition first is online using the history online data
Between target online data in section, different acquisition plans are respectively configured for different types of first line duration section so as to realize
Slightly, corresponding target online data is got according to different acquisition strategy Accurate Predictions to reach, and is no longer according to unified
Rule carry out online data prediction, and then reach improve obtain online data accuracy effect, overcome correlation technique
In the presence of the data accuracy got it is relatively low the problem of.
Brief description of the drawings
Accompanying drawing described herein is used for providing a further understanding of the present invention, constitutes the part of the application, this hair
Bright schematic description and description is used to explain the present invention, does not constitute inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is a kind of schematic diagram of the application environment of the method for acquisition online data according to embodiments of the present invention;
Fig. 2 is a kind of schematic flow sheet of optional method for obtaining online data according to embodiments of the present invention;
Fig. 3 is the schematic flow sheet of another optional method for obtaining online data according to embodiments of the present invention;
Fig. 4 is a kind of schematic diagram of optional device for obtaining online data according to embodiments of the present invention;
Fig. 5 is a kind of schematic diagram of optional electronic installation according to embodiments of the present invention.
Embodiment
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, rather than whole embodiments.Based on the embodiment in the present invention, ordinary skill people
The every other embodiment that member is obtained under the premise of creative work is not made, should all belong to the model that the present invention is protected
Enclose.
It should be noted that term " first " in description and claims of this specification and above-mentioned accompanying drawing, "
Two " etc. be for distinguishing similar object, without for describing specific order or precedence.It should be appreciated that so using
Data can exchange in the appropriate case, so as to embodiments of the invention described herein can with except illustrating herein or
Order beyond those of description is implemented.In addition, term " comprising " and " having " and their any deformation, it is intended that cover
Lid is non-exclusive to be included, for example, the process, method, system, product or the equipment that contain series of steps or unit are not necessarily limited to
Those steps or unit clearly listed, but may include not list clearly or for these processes, method, product
Or the intrinsic other steps of equipment or unit.
Embodiment 1
According to embodiments of the present invention there is provided a kind of method for obtaining online data, as an alternative embodiment,
The method of the acquisition online data can be, but not limited to be applied in application environment as shown in Figure 1, the receiving terminal of server 102
106 acquisitions sent by network 104 are asked, wherein obtaining request is used for acquisition request intended application in the first line duration section
Interior target online data, the first line duration section is later than the request time that generation obtains request, and server 102 is online first
In the case that period is the particular type period, according to the acquisition strategy matched with the particular type period, obtain and the
History online data in the corresponding second line duration section of one line duration section, wherein, the second line duration section is earlier than request
Time;Using the target in history online data acquisition the first line duration section in above-mentioned second line duration section in line number
According to.
In the present embodiment, it is used for target of the acquisition request intended application in the first line duration section in line number in reception
According to acquisition request after, judge the first line duration section be the particular type period in the case of, according to certain kinds
The acquisition strategy that the type period matches, the history obtained in the second line duration section corresponding with the first line duration section exists
Line, to obtain the target online data in the first line duration section using the history online data, wherein, the first line duration section
It is later than request time, the second line duration section is earlier than above-mentioned request time.That is, according to be predicted first it is online when
Between the time segment type of section determine acquisition strategy, obtained so as to realize according to the acquisition strategy matched with the first line duration section
History online data in the second line duration section corresponding with the first line duration section, with the history got using classifying type
Online data is accurately obtained the target online data in the first line duration to be predicted section, and is no longer according to unified
Rule carries out online data prediction, realizes and improves the accuracy that online data is obtained, and then overcomes in correlation technique and deposited
The data accuracy got it is relatively low the problem of.
Alternatively, in the present embodiment, above-mentioned terminal can include but is not limited at least one of:Mobile phone, flat board electricity
Brain, notebook computer, desktop PC, intelligent television and other the hardware device of data conversion client is installed.Above-mentioned network
At least one of can be included but is not limited to:Wide area network, Metropolitan Area Network (MAN), LAN.Above-mentioned simply a kind of example, the present embodiment pair
This does not do any restriction.
According to embodiments of the present invention there is provided a kind of method for obtaining online data, as shown in Fig. 2 this method includes:
S202, receives and obtains request, wherein, obtaining request is used for acquisition request intended application in the first line duration section
Target online data, the first line duration section be later than generation obtain request request time;
S204, the first line duration section be the particular type period in the case of, according to particular type period phase
The acquisition strategy of matching, obtains the history online data in the second line duration section corresponding with the first line duration section, wherein,
Second line duration section is earlier than request time;
S206, the target obtained using the history online data in the second line duration section in the first line duration section is online
Data.
Alternatively, in the present embodiment, the method for above-mentioned acquisition online data can be, but not limited to be applied to answer target
During online data in is predicted, wherein, above-mentioned intended application can include but is not limited to mobile terminal application,
PC (Personal Computer, abbreviation PC) end application, for example, above-mentioned intended application can be game application, video
Using, shopping application, news application and music application etc..That is, by the scheme of above-mentioned acquisition online data, it is determined that
After the type for obtaining the first line duration section where the target online data of request institute acquisition request, according to first it is online when
Between section type matching acquisition strategy, obtain and exist with the history in the first line duration section corresponding second line duration section
Line number evidence, to predict the target online data obtained in the first line duration section using the history online data, so as to realize pin
Different acquisition strategies are respectively configured to different types of first line duration section, it is accurate according to different acquisition strategies to reach
Prediction gets corresponding target online data, and is no longer to carry out online data prediction according to unified rule, and then reaches
The effect for the accuracy for obtaining online data is improved, the data accuracy got overcome in the presence of correlation technique is relatively low
The problem of.
Alternatively, in the present embodiment, the above-mentioned particular type period includes:When first kind period, Second Type
Between section and the 3rd type of time section, wherein, intended application the first kind period online data rate of change higher than first pre-
Determine threshold value, the Second Type period is adjacent with the first kind period, the 3rd type of time section repeats according to the cycle.Example
Such as, by taking game application as an example, the first kind period can be, but not limited to be holiday time section (when in hereafter alternatively referred to as saving
Between section), the period in section) the online data rate of change of game application that gets is higher;The Second Type period can with but
It is not limited to front/rear period festivals or holidays (following article is alternatively referred to as before saving the period after period or section);During three types
Between section can be, but not limited to as the cycle events period in game application.
Alternatively, in the present embodiment, for the first kind period, identified first strategy can be, but not limited to refer to
Show the online data obtained in the period for belonging to the first kind period in very first time interval, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, in very first time interval the 3rd line duration section with
Time interval between first line duration section is one or more period 1.
For example, still by taking game application as an example, it is assumed that the request time for obtaining request is 2017.5.15.The mesh to be obtained
The first line duration section is above-mentioned by taking vacation New Year's Day in next year (assuming that 2017.12.31-2018.1.2) as an example where marking online data
With the first strategy for matching of the first line duration section can be, but not limited to for indicate before request time the 3rd it is online when
Between section (such as New Year's Day in this year vacation 2016.12.31-2017.1.2) between the very first time between request time (2017.5.15)
Every it is middle acquisition belong to the first kind period (as save in the period) period in online data, as with first it is online when
Between section corresponding second line duration section (such as the second line duration section includes:Vacation this Spring Festival 2017.1.27-2017.2.2,
Clear and Bright in this year vacation 2017.4.2-2017.4.4 and May Day in this year vacation 2017.4.30-2017.5.2) in history in line number
According to.Further, the first line duration section (vacation New Year's Day in next year is obtained using above-mentioned history online data amount forceasted
2017.12.31-2018.1.2) in target online data.
Alternatively, in the present embodiment, for the Second Type period, identified second strategy can be, but not limited to refer to
Show the online data for obtaining and belonging in the second time interval in period of Second Type period, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, the above-mentioned Second Type period can include but not limit
In:And first sub- period (as saved before period) positioned at first kind period before adjacent with the first kind period,
Second sub- period (period after as saved) adjacent with the first kind period and after the first kind period.Its
In, time interval between the 4th line duration in the second time interval section and the first line duration section is one or more the
Two cycles.
It should be noted that above-mentioned second round can be, but not limited to it is identical with the period 1, that is to say, that second strategy
Can serve to indicate that the online data of period after period before the section for obtaining upper one or more festivals or holidays or section, as with institute
Before the section of the corresponding festivals or holidays to be predicted the period or section after the period history online data;In addition, above-mentioned second round
It can also but be not limited to be less than the period 1, that is to say, that the second strategy can serve to indicate that acquisition upper one or more second
The online data in cycle, is used as the history online data with the period after period before the section of the festivals or holidays to be predicted or section.
For example, it is assumed that before section the period be Monday to Wednesday, then can obtain the online data of Monday to Wednesday of one week as going through
Period before history online data, rather than the section of a upper festivals or holidays.Above-mentioned is only to this in a kind of optional example, the present embodiment
Any restriction is not done.
Alternatively, in the present embodiment, for the 3rd type of time section, identified 3rd strategy can be, but not limited to refer to
Show obtain the 3rd time interval in belong to the 3rd type of time section period in online data, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, above-mentioned 3rd type of time section can include but not limit
In:Repeat according to the period 3, by taking game application as an example, such as period racing season in cycle.
Alternatively, in the present embodiment, the first line duration section be the non specified type period in the case of, can with but
The acquisition model matched according to characteristic determination of the intended application before request time with the first line duration section is not limited to,
To obtain the above-mentioned target online data to be predicted using the acquisition model.Wherein, in the present embodiment, features described above data
It can be, but not limited to include at least one of:The registration amount of intended application, the login amount of intended application, the retention of intended application
The capacity of returns of amount, the number of dropouts of intended application and intended application.
Alternatively, in the present embodiment, above-mentioned acquisition model can include but is not limited at least one of:Linear regression
Model, Random Forest model, the acquisition model of customization.Above-mentioned is only a kind of example, trains what is drawn according to features described above data
Obtain in the method that model goes for the acquisition online data in the present embodiment, the present embodiment and do not do any limit to this
It is fixed.
Specific to be illustrated with reference to Fig. 3 example showns, such as step S302-S308 after receiving and obtaining request, obtains the
One line duration section, and judge whether the first line duration section is the particular type period, is being judged for the particular type time
In the case of section, step S306-1 is performed, step S306-2 is otherwise performed.Further after step S306-1 is performed, perform
Subsequent step, such as performs step S308-1, really in the case where judging that the first line duration section belongs to the first kind period
Make the first strategy;Step S308-2 is performed in the case where judging that the first line duration section belongs to the Second Type period,
Determine the second strategy;Step S308- is performed in the case where judging that the first line duration section belongs to the 3rd type of time section
3, determine the 3rd strategy.If in addition, in the case of judging for the non specified type period, performing step S306-2, according to
Characteristic of the intended application before request time determines acquisition model, and then upper using what is determined according to step S308-4
State acquisition model and obtain target online data.
It should be noted that above-mentioned steps S306-1 and step 306-2 are regardless of sequencing is performed, after being judgment step
Two different branches situations, above-mentioned steps S308-1 to step 308-4, regardless of sequencing is performed, is three kinds of juxtaposed condition correspondences
Three branches arranged side by side.Any restriction is not done in the present embodiment to this.
The embodiment provided by the application, determines to obtain according to the time segment type of the first line duration to be predicted section
Strategy is taken, obtains corresponding with the first line duration section according to the acquisition strategy that the first line duration section matches so as to realize
History online data in second line duration section, is accurately obtained with the history online data got using classifying type and wanted
Target online data in the first line duration section of prediction, and be no longer to carry out online data prediction according to unified rule,
Realize and improve the accuracy that online data is obtained, and then overcome the data accuracy got in the presence of correlation technique
Relatively low the problem of.
As a kind of optional scheme, according to the acquisition strategy matched with the particular type period, obtain with first
History online data in line period corresponding second line duration section includes:
S1, in the case where the first line duration section is the first kind period, determines that acquisition strategy is and the first kind
The first strategy that period matches, wherein, the particular type period includes the first kind period, and intended application is in the first kind
The online data rate of change of type period is higher than the first predetermined threshold;
S2, obtains online in the period for belonging to the first kind period according to the first strategy out of the very first time interval
Data, as the history online data in the second line duration section corresponding with the first line duration section, wherein, between the very first time
The time interval between the 3rd line duration section and request time is divided into, the 3rd line duration section occurs before request time,
And the time interval between the first line duration section is one or more period 1.
It should be noted that in the present embodiment, above-mentioned very first time interval includes one or more period 1, its
In, in the case where the first line duration section and request time are closer to the distance, it can be, but not limited in above-mentioned very first time interval
Include a period 1, that is to say, that can just be obtained in distance first line duration section period 1 forward
History online data before to request time;Or, in the case where the first line duration section and request time are distant,
It can be, but not limited to include multiple period 1 in above-mentioned very first time interval, that is to say, that in the line duration of distance first section
The history online data before request time can be got in multiple period 1 forward.Wherein, during the above-mentioned first kind
Between section can be, but not limited to as festivals or holidays type, the above-mentioned period 1 can be, but not limited to as a calendar year.In other words, pass through
The period produces in the section in 1 year before request time history online data is obtained come after predicting and obtaining request time
The target online data that the period produces in section.
In addition, in the present embodiment, above-mentioned second line duration section can be, but not limited to include one or more periods,
Such as the period in period in a section or multiple sections.Above-mentioned is only not do any restriction to this in a kind of example, the present embodiment.
Alternatively, in the present embodiment, above-mentioned online data rate of change can be, but not limited to for indicating online data
Fluctuation situation.It should be noted that according to produced by the period in big data statistical analysis section online data change fluctuation compared with
Greatly, thus according to the characteristic of period in section, to determine whether the first line duration to be predicted section is the period in saving,
Determine whether that the history that corresponding second line duration section is obtained using the first strategy corresponding with the period in section is existed
Line number evidence.
The embodiment provided by the application, in the case where the first line duration section is defined as the first kind period,
The first strategy according to matching with the first kind period is obtained out of the very first time interval belongs to the first kind period
Online data in period, as the history online data in the second line duration section corresponding with the first line duration section,
So as to realize accurate target online data is got according to the first strategy with the type matching of the first line duration section.
As a kind of optional scheme, the first line duration is obtained using the history online data in the second line duration section
Target online data in section includes:
S1, in the case where the first line duration section is the first kind period, second in very first time interval is existed
History online data in the line period is divided into multiple N tuple sequences, wherein, each N tuple sequence includes N number of member
Element, each element is used for the first same period rate of change of log history online data, and the first same period rate of change is current N tuples
Online data rate of change of i-th of element relative to the i-th element in previous N tuple sequences in sequence, i be more than or equal to
1 and less than or equal to N integer;
S2, obtains the element in first object N tuple sequences corresponding with the 3rd line duration section;
S3, is compared in the element and first object N tuple sequences in multiple N tuple sequences in very first time interval successively
Element, obtained from multiple N tuple sequences with the closest object N tuple sequences of first object N tuple sequences, be used as the
First reference sequences of one line duration section;
S4, target online data is obtained according to the first reference sequences and history online data.
Alternatively, in the present embodiment, element in multiple N tuple sequences in very first time interval and the are compared successively
Element in one target N tuple sequences, is obtained and closest pair of first object N tuple sequences from multiple N tuple sequences
As N tuple sequences, the reference sequences as the first line duration section include:
S31, repeats following steps, until multiple N tuple sequences in traversal very first time interval:
S31-2, obtains current N tuple sequences from multiple N tuple sequences;
S31-4, obtains i-th of element in current N tuple sequences and i-th of element in first object N tuple sequences
Difference;
S31-6, using the quadratic sum of the difference of N number of element obtain current N tuple sequences and first object N tuple sequences it
Between distance;
S31-8, determines object N tuple sequences in multiple N tuple sequences as ginseng according to the quadratic sum of minimum difference
Examine sequence.
Alternatively, in the present embodiment, above-mentioned N tuple sequences can be, but not limited to minimum festivals or holidays unit, for example, 3
My god.That is, N number of element in N tuple sequences can be 3 elements.Above-mentioned is only to this in a kind of example, the present embodiment
Any restriction is not done.
Specific to be illustrated with reference to the example below, it is assumed that still by taking game application as an example, the request time for obtaining request is
2017.5.15, the first line duration section where the target online data to be obtained with vacation in this year 11 (assuming that
2017.10.1-2017.10.7) exemplified by, above-mentioned the first strategy matched with the first line duration section can be, but not limited to be used to refer to
Show from the 3rd line duration section (such as vacation 2016.10.1-2017.10.7 of last year 11) before request time to request time
The very first time between (2017.5.15), which is obtained in interval, to be belonged in the period of first kind period (period in as saved)
Online data, as with the first line duration section corresponding second line duration section, wherein, above-mentioned second line duration section bag
Include:New Year's Day in this year vacation 2016.12.31-2017.1.2, vacation this Spring Festival 2017.1.27-2017.2.2, this year are clear and bright false
Phase 2017.4.2-2017.4.4 and May Day in this year vacation 2017.4.30-2017.5.2) in history online data.Further,
The target online data in the first line duration section (vacation in this year 11) is obtained using above-mentioned history online data amount forceasted.
For example, the history online data in very first time interval in above-mentioned second line duration section is divided into multiple 3 yuan
Group sequence, for example, 7 days of vacation this Spring Festival are divided into 3 triad sequences, such as (1-27,1-28,1-29), (1-30,
1-31,2-1), (1-31,2-1,2-2), wherein, above-mentioned division principle can be, but not limited to be adjusted according to different application scene
Whole, above-mentioned is only not do any restriction to this in a kind of example, the present embodiment.
It should be noted that above-mentioned daily as an element, the same period for recording corresponding history online data becomes
Rate, is such as illustrated by taking clear and bright vacation in this year as an example, and vacation Clear and Bright in this year, corresponding triad sequence was (4-2,4-3,4-
4), the 1st element 4-2 is used to recording relative to a upper triad sequence, last triple sequence in such as Spring Festival holiday
The 1st element 1-31 online data rate of change x1, the 2nd element 4-3 is used for record relative to a upper triple sequence in row
2nd element 2-1 online data rate of change x2, the 3rd element in last triad sequence in row, such as Spring Festival holiday
4-4 is used to record relative to a upper triad sequence, the 3rd element in last triad sequence in such as Spring Festival holiday
2-2 online data rate of change x3.
Further, corresponding 3 triad sequences of above-mentioned 3rd line duration section, such as (10-1,10-2,10-3), (10-
4,10-5,10-6), (10-5,10-6,10-7), it is assumed that with first triad sequence (10-1,10-2,10-3) for the first mesh
Mark triad sequence to illustrate, then can compare each triad sequence and above-mentioned first in above-mentioned second line duration section
Target triad sequence (10-1,10-2,10-3).Specifically, comparing each element in above-mentioned triad sequence, repeat
Following steps, until multiple triad sequences in traversal very first time interval:Obtained from multiple triad sequences and work as first three
Tuple sequence (assuming that triad sequence (4-2,4-3,4-4) of clear and bright vacation), obtains each element in current triad sequence
With the difference of each element in first object triad sequence (10-1,10-2,10-3), the corresponding difference of above three element is obtained
Square root sum square of value, is used as distance therebetween.Multiple triad sequences out of above-mentioned second line duration section
The middle acquisition triad sequence minimum with first object triad sequence distance refers to sequence to realize as reference sequences according to this
Row obtain target online data.
For example, i-th of triad sequence HiIt is expressed as follows:
Hi=(xi1,xi2,xi3)
Wherein, xi1For representing the 1st element, x in i-th of triad sequencei2For representing i-th of triple sequence
The 2nd element, x in rowi3For representing the 3rd element in i-th of triad sequence.
For example, first object triad sequence HjWith i-th of triad sequence HiComparison process can be such as below equation:
The embodiment provided by the application, by the way that history online data to be divided into the N tuple sequences of equal length, with
Realization usually determines similar sequences by comparing the member in N tuple sequences, so that when realizing using similar sequences to predict in section
Between section target online data, it is ensured that the accuracy of the target online data got.
As a kind of optional scheme, according to the acquisition strategy matched with the particular type period, obtain with first
History online data in line period corresponding second line duration section includes:
S1, in the case where the first line duration section is the Second Type period, determines that acquisition strategy is and Second Type
The second strategy that period matches, wherein, the particular type period includes Second Type period, Second Type period bag
Include:First sub- period and first kind time adjacent with the first kind period and before the first kind period
Duan Xianglin and the second sub- period after the first kind period;
S2, obtains out of second time interval according to the second strategy and belongs to online in the period of Second Type period
Data, as the history online data in the second line duration section corresponding with the first line duration section, wherein, between the second time
The time interval between the 4th line duration section and request time is divided into, the 4th line duration section occurs before request time,
And the time interval between the first line duration section is one or more second rounds.
It should be noted that above-mentioned second round can be, but not limited to it is identical with the period 1, that is to say, that second strategy
Can serve to indicate that the online data of period after period before the section for obtaining upper one or more festivals or holidays or section, as with institute
Before the section of the corresponding festivals or holidays to be predicted the period or section after the period history online data;In addition, above-mentioned second round
It can also but be not limited to be less than the period 1, that is to say, that the second strategy can serve to indicate that acquisition upper one or more second
The online data in cycle, is used as the history online data with the period after period before the section of the festivals or holidays to be predicted or section.
For example, it is assumed that before section the period be Monday to Wednesday, then can obtain the online data of Monday to Wednesday of one week as going through
Period before history online data, rather than the section of a upper festivals or holidays.Above-mentioned is only to this in a kind of optional example, the present embodiment
Any restriction is not done.
It should be noted that in the present embodiment, above-mentioned second time interval includes one or more second rounds, from
And by obtaining the average data of second round, to ensure the accuracy of accessed target online data.
The embodiment provided by the application, in the case where the first line duration section is defined as the Second Type period,
The second strategy according to matching with the Second Type period is obtained out of second time interval belongs to the Second Type period
Online data in period, as the history online data in the second line duration section corresponding with the first line duration section,
So as to realize accurate target online data is got according to the second strategy with the type matching of the second line duration section.
As a kind of optional scheme, the first line duration is obtained using the history online data in the second line duration section
Target online data in section includes:
S1, is the Second Type period in the first line duration section, the first line duration section includes the situation of M element
Under, acquisition history corresponding with j-th of element in M element exists respectively from each second round in the second time interval
Line number evidence, wherein, second same period that history online data corresponding with j-th of element in M element includes j-th of element becomes
Rate, the second same period rate of change of j-th of element is j-th of element in current second round relative to upper one second
The online data rate of change of j-th of element in cycle, j is the integer more than or equal to 1 and less than or equal to M, and M is less than or equal to second
Cycle;
S2, according to history online data corresponding with j-th of element in M element in the second time interval, obtains jth
The average same period rate of change of individual element;
S3, the target of j-th of element in the first line duration section in M element is obtained according to average same period rate of change
Online data.
Specific to be illustrated with reference to the example below, still by taking game application as an example, the request time for obtaining request is
2017.5.15, the first line duration section was with three days before vacation in this year 11 where the target online data to be obtained
Exemplified by example (assuming that Thursday to Saturday 2017.9.28-2017.9.30), it is above-mentioned matched with the first line duration section it is second tactful
It can be, but not limited to for indicating the 4th line duration section (such as 11 vacation of the last year 2016.9.28- before request time
2017.9.30) obtained into the second time interval between request time (2017.5.15) and belong to the Second Type period (such as
Period in section) period in online data, as with the first line duration section corresponding second line duration section, its
In, above-mentioned second line duration section can include:Vacation New Year's Day in this year first three day, first three day of vacation this Spring Festival, this year
Clear and bright vacation first three day and first three day of vacation May Day in this year) in history online data.In addition, it is above-mentioned with first
The line period matching the second strategy can with but be not limited to use in indicate before request time the 4th line duration section (such as
September in this year Thursday of 2nd week reciprocal is to Saturday 2016.9.21-2017.9.23) to the between request time (2017.5.15)
The online data belonged in the period of Second Type period is obtained in two time intervals, as right with the first line duration section
The the second line duration section answered, wherein, above-mentioned second line duration section can also include:September in this year all Thursdays second from the bottom
To Saturday, the history online data of September in this year all Thursdays to Saturday third from the bottom, etc..
Further, M with target online data are obtained respectively from the history online data in above-mentioned multiple second rounds
Corresponding multiple second same period rates of change of j-th of element in element, according to the average value of above-mentioned multiple second same period rates of change, such as
Average same period rate of change, the corresponding target online data of j-th of element in M element is obtained to predict.
The embodiment provided by the application, by obtaining in the recent period, averagely same period amount of increase is obtained when first is online to predict
Between section (as save before the period or section after the period) target online data so that reach improve acquired in target online data
Accuracy.
As a kind of optional scheme, according to the acquisition strategy matched with the particular type period, obtain with first
History online data in line period corresponding second line duration section includes:
S1, in the case of the type of time of the first line duration Duan Wei tri- section, determines that acquisition strategy is and the 3rd type
The 3rd strategy that period matches, wherein, the particular type period includes the 3rd type of time section, and the 3rd type of time section is pressed
Repeat according to the period 3;
S2, obtains online in the period for belonging to the 3rd type of time section according to the 3rd strategy out of the 3rd time interval
Data, as the history online data in the second line duration section corresponding with the first line duration section, wherein, between the 3rd time
The time interval between the 5th line duration section and request time is divided into, the 5th line duration section occurs before request time,
And the time interval between the first line duration section is one or more period 3.
Alternatively, in the present embodiment, for the 3rd type of time section, identified 3rd strategy can be, but not limited to refer to
Show obtain the 3rd time interval in belong to the 3rd type of time section period in online data, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, above-mentioned 3rd type of time section can include but not limit
In:Repeat according to the period 3, by taking game application as an example, such as period racing season in cycle.
The embodiment provided by the application, in the case where the first line duration section is defined as the 3rd type of time section,
The 3rd strategy according to matching with the 3rd type of time section is obtained out of the 3rd time interval belongs to the 3rd type of time section
Online data in period, as the history online data in the second line duration section corresponding with the first line duration section,
So as to realize accurate target online data is got according to the 3rd strategy with the type matching of the first line duration section.
As a kind of optional scheme, the first line duration is obtained using the history online data in the second line duration section
Target online data in section includes:
S1, in the case of the type of time of the first line duration Duan Wei tri- section, the first line duration section includes P member
In the case of element, obtain corresponding with k-th of element in P element respectively from each period 3 in the 3rd time interval
History online data, wherein, history online data corresponding with k-th of element in P element includes the of k-th element
Three same period rates of change, the 3rd same period rate of change of k-th of element is k-th of element in the current period 3 relative to upper
The online data rate of change of k-th of element in one period 3, k is the integer more than or equal to 1 and less than or equal to P;
S2, according to history online data corresponding with k-th of element in P element in the 3rd time interval, obtains kth
The average same period rate of change of individual element;
S3, the target of k-th of element in the first line duration section in P element is obtained according to average same period rate of change
Online data.
It should be noted that in the present embodiment, the backflow largely improved occurs in the cycle events in intended application
Customer volume, natural trend of this amplification considerably beyond online data.Therefore, in the present embodiment, it is same in the recent period by obtaining
Phase is averaged amount of increase to obtain the target online data of the first line duration section for belonging to the 3rd type of time section to be predicted.Example
Such as, the mode for obtaining the corresponding target online data of the 3rd type of time section is referred to above-mentioned second strategy, wherein, the 3rd week
Phase, which can be not limited to greatly be more than in the self-defined cycle obtained according to actual scene, the present embodiment, does not do any restriction to this.
The example provided by the application, by obtaining in the recent period, averagely same period amount of increase is obtained in the first line duration to predict
The target online data of section (such as cycle time section), so as to reach the accuracy of the target online data acquired in improving.
As a kind of optional scheme, after acquisition request is received, in addition to:
S1, in the case where the first line duration section is the non specified type period, according to intended application in request time
Characteristic before, obtains the target online data in the first line duration section, wherein, characteristic includes following at least one
Kind:The registration amount of intended application, the login amount of intended application, the retention amount of intended application, the number of dropouts of intended application and target
The capacity of returns of application.
Alternatively, in the present embodiment, the first line duration section be the non specified type period in the case of, can with but
The acquisition model matched according to characteristic determination of the intended application before request time with the first line duration section is not limited to,
To obtain the above-mentioned target online data to be predicted using the acquisition model.
Alternatively, in the present embodiment, the characteristic according to intended application before request time, obtains first online
Target online data in period includes:
S12, determines what is matched with the first line duration section according to characteristic of the intended application before request time
Obtain model;
S14, using the target online data obtained in model acquisition the first line duration section determined, wherein, in spy
Levy in the case that data are designated as linear relationship, use the linear regression model (LRM) determined to obtain the mesh in the first line duration section
Mark online data;Or, in the case where characteristic is designated as non-linear relation, obtained using the Random Forest model determined
Take the target online data in the first line duration section;Or, in the case where customization obtains model, use the acquisition mould of customization
Type obtains the target online data in the first line duration section.
Alternatively, in the present embodiment, it can be, but not limited to determine to obtain model based on features described above data, further,
Use the target online data in the first line duration section for obtaining the model acquisition non specified type period determined.It is above-mentioned
Characteristic can include but is not limited to:Month feature, Zhou Tezheng and day feature etc..
Alternatively, in the present embodiment, above-mentioned acquisition model can include but is not limited to:Traditional linear regression model (LRM)
(Linear Regression), the acquisition model of Random Forest model (Random Forest) customization, using this flexible
Model is obtained, can not only ensure the precision of prediction, the algorithm complexity for obtaining model can also be reduced, reduces and obtains model
Maintenance cost.
The embodiment provided by the application, it is proposed that the thinking that strategy can match somebody with somebody, greatly simplifies acquisition model, so that
The business of realization is quickly accessed, the effect that low cost is safeguarded.
It should be noted that for foregoing each method embodiment, in order to be briefly described, therefore it is all expressed as a series of
Combination of actions, but those skilled in the art should know, the present invention is not limited by described sequence of movement because
According to the present invention, some steps can be carried out sequentially or simultaneously using other.Secondly, those skilled in the art should also know
Know, embodiment described in this description belongs to preferred embodiment, involved action and module is not necessarily of the invention
It is necessary.
Through the above description of the embodiments, those skilled in the art can be understood that according to above-mentioned implementation
The method of example can add the mode of required general hardware platform to realize by software, naturally it is also possible to by hardware, but a lot
In the case of the former be more preferably embodiment.Understood based on such, technical scheme is substantially in other words to existing
The part that technology contributes can be embodied in the form of software product, and the computer software product is stored in a storage
In medium (such as ROM/RAM, magnetic disc, CD), including some instructions are to cause a station terminal equipment (can be mobile phone, calculate
Machine, server, or network equipment etc.) perform method described in each of the invention embodiment.
Embodiment 2
According to embodiments of the present invention, a kind of acquisition of the method for implementing above-mentioned acquisition online data is additionally provided online
The device of data, as shown in figure 4, the device includes:
1) receiving unit 402, request is obtained for receiving, wherein, obtaining request is used for acquisition request intended application the
Target online data in one line duration section, the first line duration section is later than the request time that generation obtains request;
2) first acquisition unit 404, for the first line duration section be the particular type period in the case of, according to
The acquisition strategy that the particular type period matches, obtains going through in the second line duration section corresponding with the first line duration section
History online data, wherein, the second line duration section is earlier than request time;
3) second acquisition unit 406, it is online for obtaining first using the history online data in the second line duration section
Target online data in period.
Alternatively, in the present embodiment, the method for above-mentioned acquisition online data can be, but not limited to be applied to answer target
During online data in is predicted, wherein, above-mentioned intended application can include but is not limited to mobile terminal application,
PC (Personal Computer, abbreviation PC) end application, for example, above-mentioned intended application can be game application, video
Using, shopping application, news application and music application etc..That is, by the scheme of above-mentioned acquisition online data, it is determined that
After the type for obtaining the first line duration section where the target online data of request institute acquisition request, according to first it is online when
Between section type matching acquisition strategy, obtain and exist with the history in the first line duration section corresponding second line duration section
Line number evidence, to predict the target online data obtained in the first line duration section using the history online data, so as to realize pin
Different acquisition strategies are respectively configured to different types of first line duration section, it is accurate according to different acquisition strategies to reach
Prediction gets corresponding target online data, and is no longer to carry out online data prediction according to unified rule, and then reaches
The effect for the accuracy for obtaining online data is improved, the data accuracy got overcome in the presence of correlation technique is relatively low
The problem of.
Alternatively, in the present embodiment, the above-mentioned particular type period includes:When first kind period, Second Type
Between section and the 3rd type of time section, wherein, intended application the first kind period online data rate of change higher than first pre-
Determine threshold value, the Second Type period is adjacent with the first kind period, the 3rd type of time section repeats according to the cycle.Example
Such as, by taking game application as an example, the first kind period can be, but not limited to be holiday time section (when in hereafter alternatively referred to as saving
Between section), the period in section) the online data rate of change of game application that gets is higher;The Second Type period can with but
It is not limited to front/rear period festivals or holidays (following article is alternatively referred to as before saving the period after period or section);During three types
Between section can be, but not limited to as the cycle events period in game application.
Alternatively, in the present embodiment, for the first kind period, identified first strategy can be, but not limited to refer to
Show the online data obtained in the period for belonging to the first kind period in very first time interval, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, in very first time interval the 3rd line duration section with
Time interval between first line duration section is one or more period 1.
For example, still by taking game application as an example, it is assumed that the request time for obtaining request is 2017.5.15.The mesh to be obtained
The first line duration section is above-mentioned by taking vacation New Year's Day in next year (assuming that 2017.12.31-2018.1.2) as an example where marking online data
With the first strategy for matching of the first line duration section can be, but not limited to for indicate before request time the 3rd it is online when
Between section (such as New Year's Day in this year vacation 2016.12.31-2017.1.2) between the very first time between request time (2017.5.15)
Every it is middle acquisition belong to the first kind period (as save in the period) period in online data, as with first it is online when
Between section corresponding second line duration section (such as the second line duration section includes:Vacation this Spring Festival 2017.1.27-2017.2.2,
Clear and Bright in this year vacation 2017.4.2-2017.4.4 and May Day in this year vacation 2017.4.30-2017.5.2) in history in line number
According to.Further, the first line duration section (vacation New Year's Day in next year is obtained using above-mentioned history online data amount forceasted
2017.12.31-2018.1.2) in target online data.
Alternatively, in the present embodiment, for the Second Type period, identified second strategy can be, but not limited to refer to
Show the online data for obtaining and belonging in the second time interval in period of Second Type period, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, the above-mentioned Second Type period can include but not limit
In:And first sub- period (as saved before period) positioned at first kind period before adjacent with the first kind period,
Second sub- period (period after as saved) adjacent with the first kind period and after the first kind period.Its
In, time interval between the 4th line duration in the second time interval section and the first line duration section is one or more the
Two cycles.
It should be noted that above-mentioned second round can be, but not limited to it is identical with the period 1, that is to say, that second strategy
Can serve to indicate that the online data of period after period before the section for obtaining upper one or more festivals or holidays or section, as with institute
Before the section of the corresponding festivals or holidays to be predicted the period or section after the period history online data;In addition, above-mentioned second round
It can also but be not limited to be less than the period 1, that is to say, that the second strategy can serve to indicate that acquisition upper one or more second
The online data in cycle, is used as the history online data with the period after period before the section of the festivals or holidays to be predicted or section.
For example, it is assumed that before section the period be Monday to Wednesday, then can obtain the online data of Monday to Wednesday of one week as going through
Period before history online data, rather than the section of a upper festivals or holidays.Above-mentioned is only to this in a kind of optional example, the present embodiment
Any restriction is not done.
Alternatively, in the present embodiment, for the 3rd type of time section, identified 3rd strategy can be, but not limited to refer to
Show obtain the 3rd time interval in belong to the 3rd type of time section period in online data, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, above-mentioned 3rd type of time section can include but not limit
In:Repeat according to the period 3, by taking game application as an example, such as period racing season in cycle.
Alternatively, in the present embodiment, the first line duration section be the non specified type period in the case of, can with but
The acquisition model matched according to characteristic determination of the intended application before request time with the first line duration section is not limited to,
To obtain the above-mentioned target online data to be predicted using the acquisition model.Wherein, in the present embodiment, features described above data
It can be, but not limited to include at least one of:The registration amount of intended application, the login amount of intended application, the retention of intended application
The capacity of returns of amount, the number of dropouts of intended application and intended application.
Alternatively, in the present embodiment, above-mentioned acquisition model can include but is not limited at least one of:Linear regression
Model, Random Forest model, the acquisition model of customization.Above-mentioned is only a kind of example, trains what is drawn according to features described above data
Obtain in the method that model goes for the acquisition online data in the present embodiment, the present embodiment and do not do any limit to this
It is fixed.
Specific to be illustrated with reference to Fig. 3 example showns, such as step S302-S308 after receiving and obtaining request, obtains the
One line duration section, and judge whether the first line duration section is the particular type period, is being judged for the particular type time
In the case of section, step S306-1 is performed, step S306-2 is otherwise performed.Further after step S306-1 is performed, perform
Subsequent step, such as performs step S308-1, really in the case where judging that the first line duration section belongs to the first kind period
Make the first strategy;Step S308-2 is performed in the case where judging that the first line duration section belongs to the Second Type period,
Determine the second strategy;Step S308- is performed in the case where judging that the first line duration section belongs to the 3rd type of time section
3, determine the 3rd strategy.If in addition, in the case of judging for the non specified type period, performing step S306-2, according to
Characteristic of the intended application before request time determines acquisition model, and then upper using what is determined according to step S308-4
State acquisition model and obtain target online data.
It should be noted that above-mentioned steps S306-1 and step 306-2 are regardless of sequencing is performed, after being judgment step
Two different branches situations, above-mentioned steps S308-1 to step 308-4, regardless of sequencing is performed, is three kinds of juxtaposed condition correspondences
Three branches arranged side by side.Any restriction is not done in the present embodiment to this.
The embodiment provided by the application, determines to obtain according to the time segment type of the first line duration to be predicted section
Strategy is taken, obtains corresponding with the first line duration section according to the acquisition strategy that the first line duration section matches so as to realize
History online data in second line duration section, is accurately obtained with the history online data got using classifying type and wanted
Target online data in the first line duration section of prediction, and be no longer to carry out online data prediction according to unified rule,
Realize and improve the accuracy that online data is obtained, and then overcome the data accuracy got in the presence of correlation technique
Relatively low the problem of.
As a kind of optional scheme, first acquisition unit 404 includes:
1) the first determining module, in the case of being the first kind period in the first line duration section, it is determined that obtaining
Strategy is the first strategy matched with the first kind period, wherein, the particular type period includes the first kind period,
Online data rate of change of the intended application in the first kind period is higher than the first predetermined threshold;
2) the first acquisition module, belongs to the first kind period for being obtained according to the first strategy out of the very first time interval
Period in online data, as with the history in the first line duration section corresponding second line duration section in line number
According to, wherein, the very first time exists at intervals of the time interval between the 3rd line duration section and request time, the 3rd line duration section
Occur before request time, and the time interval between the first line duration section is one or more period 1.
It should be noted that in the present embodiment, above-mentioned very first time interval includes one or more period 1, its
In, in the case where the first line duration section and request time are closer to the distance, it can be, but not limited in above-mentioned very first time interval
Include a period 1, that is to say, that can just be obtained in distance first line duration section period 1 forward
History online data before to request time;Or, in the case where the first line duration section and request time are distant,
It can be, but not limited to include multiple period 1 in above-mentioned very first time interval, that is to say, that in the line duration of distance first section
The history online data before request time can be got in multiple period 1 forward.Wherein, during the above-mentioned first kind
Between section can be, but not limited to as festivals or holidays type, the above-mentioned period 1 can be, but not limited to as a calendar year.In other words, pass through
The period produces in the section in 1 year before request time history online data is obtained come after predicting and obtaining request time
The target online data that the period produces in section.
In addition, in the present embodiment, above-mentioned second line duration section can be, but not limited to include one or more periods,
Such as the period in period in a section or multiple sections.Above-mentioned is only not do any restriction to this in a kind of example, the present embodiment.
Alternatively, in the present embodiment, above-mentioned online data rate of change can be, but not limited to for indicating online data
Fluctuation situation.It should be noted that according to produced by the period in big data statistical analysis section online data change fluctuation compared with
Greatly, thus according to the characteristic of period in section, to determine whether the first line duration to be predicted section is the period in saving,
Determine whether that the history that corresponding second line duration section is obtained using the first strategy corresponding with the period in section is existed
Line number evidence.
The embodiment provided by the application, in the case where the first line duration section is defined as the first kind period,
The first strategy according to matching with the first kind period is obtained out of the very first time interval belongs to the first kind period
Online data in period, as the history online data in the second line duration section corresponding with the first line duration section,
So as to realize accurate target online data is got according to the first strategy with the type matching of the first line duration section.
As a kind of optional scheme, second acquisition unit 406 is realized by following steps utilizes the second line duration section
Interior history online data obtains the target online data in the first line duration section:
S1, in the case where the first line duration section is the first kind period, second in very first time interval is existed
History online data in the line period is divided into multiple N tuple sequences, wherein, each N tuple sequence includes N number of member
Element, each element is used for the first same period rate of change of log history online data, and the first same period rate of change is current N tuples
Online data rate of change of i-th of element relative to the i-th element in previous N tuple sequences in sequence, i be more than or equal to
1 and less than or equal to N integer;
S2, obtains the element in first object N tuple sequences corresponding with the 3rd line duration section;
S3, is compared in the element and first object N tuple sequences in multiple N tuple sequences in very first time interval successively
Element, obtained from multiple N tuple sequences with the closest object N tuple sequences of first object N tuple sequences, be used as the
First reference sequences of one line duration section;
S4, target online data is obtained according to the first reference sequences and history online data.
Alternatively, in the present embodiment, step S3 realizes comparison process by following steps:
Following steps are repeated, until multiple N tuple sequences in traversal very first time interval:From multiple N tuples sequences
Current N tuple sequences are obtained in row;Obtain i-th element in current N tuple sequences and the in first object N tuple sequences
The difference of i element;Using the quadratic sum of the difference of N number of element obtain current N tuple sequences and first object N tuple sequences it
Between distance;Determine object N tuple sequences in multiple N tuple sequences as referring to sequence according to the quadratic sum of minimum difference
Row.
Alternatively, in the present embodiment, above-mentioned N tuple sequences can be, but not limited to minimum festivals or holidays unit, for example, 3
My god.That is, N number of element in N tuple sequences can be 3 elements.Above-mentioned is only to this in a kind of example, the present embodiment
Any restriction is not done.
Specific to be illustrated with reference to the example below, it is assumed that still by taking game application as an example, the request time for obtaining request is
2017.5.15, the first line duration section where the target online data to be obtained with vacation in this year 11 (assuming that
2017.10.1-2017.10.7) exemplified by, above-mentioned the first strategy matched with the first line duration section can be, but not limited to be used to refer to
Show from the 3rd line duration section (such as vacation 2016.10.1-2017.10.7 of last year 11) before request time to request time
The very first time between (2017.5.15), which is obtained in interval, to be belonged in the period of first kind period (period in as saved)
Online data, as with the first line duration section corresponding second line duration section, wherein, above-mentioned second line duration section bag
Include:New Year's Day in this year vacation 2016.12.31-2017.1.2, vacation this Spring Festival 2017.1.27-2017.2.2, this year are clear and bright false
Phase 2017.4.2-2017.4.4 and May Day in this year vacation 2017.4.30-2017.5.2) in history online data.Further,
The target online data in the first line duration section (vacation in this year 11) is obtained using above-mentioned history online data amount forceasted.
For example, the history online data in very first time interval in above-mentioned second line duration section is divided into multiple 3 yuan
Group sequence, for example, 7 days of vacation this Spring Festival are divided into 3 triad sequences, such as (1-27,1-28,1-29), (1-30,
1-31,2-1), (1-31,2-1,2-2), wherein, above-mentioned division principle can be, but not limited to be adjusted according to different application scene
Whole, above-mentioned is only not do any restriction to this in a kind of example, the present embodiment.
It should be noted that above-mentioned daily as an element, the same period for recording corresponding history online data becomes
Rate, is such as illustrated by taking clear and bright vacation in this year as an example, and vacation Clear and Bright in this year, corresponding triad sequence was (4-2,4-3,4-
4), the 1st element 4-2 is used to recording relative to a upper triad sequence, last triple sequence in such as Spring Festival holiday
The 1st element 1-31 online data rate of change x1, the 2nd element 4-3 is used for record relative to a upper triple sequence in row
2nd element 2-1 online data rate of change x2, the 3rd element in last triad sequence in row, such as Spring Festival holiday
4-4 is used to record relative to a upper triad sequence, the 3rd element in last triad sequence in such as Spring Festival holiday
2-2 online data rate of change x3.
Further, corresponding 3 triad sequences of above-mentioned 3rd line duration section, such as (10-1,10-2,10-3), (10-
4,10-5,10-6), (10-5,10-6,10-7), it is assumed that with first triad sequence (10-1,10-2,10-3) for the first mesh
Mark triad sequence to illustrate, then can compare each triad sequence and above-mentioned first in above-mentioned second line duration section
Target triad sequence (10-1,10-2,10-3).Specifically, comparing each element in above-mentioned triad sequence, repeat
Following steps, until multiple triad sequences in traversal very first time interval:Obtained from multiple triad sequences and work as first three
Tuple sequence (assuming that triad sequence (4-2,4-3,4-4) of clear and bright vacation), obtains each element in current triad sequence
With the difference of each element in first object triad sequence (10-1,10-2,10-3), the corresponding difference of above three element is obtained
Square root sum square of value, is used as distance therebetween.Multiple triad sequences out of above-mentioned second line duration section
The middle acquisition triad sequence minimum with first object triad sequence distance refers to sequence to realize as reference sequences according to this
Row obtain target online data.
For example, i-th of triad sequence HiIt is expressed as follows:
Hi=(xi1,xi2,xi3)
Wherein, xi1For representing the 1st element, x in i-th of triad sequencei2For representing i-th of triple sequence
The 2nd element, x in rowi3For representing the 3rd element in i-th of triad sequence.
For example, first object triad sequence HjWith i-th of triad sequence HiComparison process can be such as below equation:
The embodiment provided by the application, by the way that history online data to be divided into the N tuple sequences of equal length, with
Realization usually determines similar sequences by comparing the member in N tuple sequences, so that when realizing using similar sequences to predict in section
Between section target online data, it is ensured that the accuracy of the target online data got.
As a kind of optional scheme, first acquisition unit 404 includes:
1) the second determining module, in the case of being the Second Type period in the first line duration section, it is determined that obtaining
Strategy is the second strategy matched with the Second Type period, wherein, the particular type period includes the Second Type period,
The Second Type period includes:The first sub- time adjacent with the first kind period and before the first kind period
Section, the second sub- period adjacent with the first kind period and after the first kind period;
2) the second acquisition module, belongs to the Second Type period for being obtained according to the second strategy out of second time interval
Period in online data, as with the history in the first line duration section corresponding second line duration section in line number
According to, wherein, the second time interval is the time interval between the 4th line duration section and request time, and the 4th line duration section exists
Occur before request time, and the time interval between the first line duration section is one or more second rounds.
It should be noted that above-mentioned second round can be, but not limited to it is identical with the period 1, that is to say, that second strategy
Can serve to indicate that the online data of period after period before the section for obtaining upper one or more festivals or holidays or section, as with institute
Before the section of the corresponding festivals or holidays to be predicted the period or section after the period history online data;In addition, above-mentioned second round
It can also but be not limited to be less than the period 1, that is to say, that the second strategy can serve to indicate that acquisition upper one or more second
The online data in cycle, is used as the history online data with the period after period before the section of the festivals or holidays to be predicted or section.
For example, it is assumed that before section the period be Monday to Wednesday, then can obtain the online data of Monday to Wednesday of one week as going through
Period before history online data, rather than the section of a upper festivals or holidays.Above-mentioned is only to this in a kind of optional example, the present embodiment
Any restriction is not done.
It should be noted that in the present embodiment, above-mentioned second time interval includes one or more second rounds, from
And by obtaining the average data of second round, to ensure the accuracy of accessed target online data.
The embodiment provided by the application, in the case where the first line duration section is defined as the Second Type period,
The second strategy according to matching with the Second Type period is obtained out of second time interval belongs to the Second Type period
Online data in period, as the history online data in the second line duration section corresponding with the first line duration section,
So as to realize accurate target online data is got according to the second strategy with the type matching of the second line duration section.
As a kind of optional scheme, second acquisition unit 406 is realized by following steps utilizes the second line duration section
Interior history online data obtains the target online data in the first line duration section:
S1, is the Second Type period in the first line duration section, the first line duration section includes the situation of M element
Under, acquisition history corresponding with j-th of element in M element exists respectively from each second round in the second time interval
Line number evidence, wherein, second same period that history online data corresponding with j-th of element in M element includes j-th of element becomes
Rate, the second same period rate of change of j-th of element is j-th of element in current second round relative to upper one second
The online data rate of change of j-th of element in cycle, j is the integer more than or equal to 1 and less than or equal to M, and M is less than or equal to second
Cycle;
S2, according to history online data corresponding with j-th of element in M element in the second time interval, obtains jth
The average same period rate of change of individual element;
S3, the target of j-th of element in the first line duration section in M element is obtained according to average same period rate of change
Online data.
Specific to be illustrated with reference to the example below, still by taking game application as an example, the request time for obtaining request is
2017.5.15, the first line duration section was with three days before vacation in this year 11 where the target online data to be obtained
Exemplified by example (assuming that Thursday to Saturday 2017.9.28-2017.9.30), it is above-mentioned matched with the first line duration section it is second tactful
It can be, but not limited to for indicating the 4th line duration section (such as 11 vacation of the last year 2016.9.28- before request time
2017.9.30) obtained into the second time interval between request time (2017.5.15) and belong to the Second Type period (such as
Period in section) period in online data, as with the first line duration section corresponding second line duration section, its
In, above-mentioned second line duration section can include:Vacation New Year's Day in this year first three day, first three day of vacation this Spring Festival, this year
Clear and bright vacation first three day and first three day of vacation May Day in this year) in history online data.In addition, it is above-mentioned with first
The line period matching the second strategy can with but be not limited to use in indicate before request time the 4th line duration section (such as
September in this year Thursday of 2nd week reciprocal is to Saturday 2016.9.21-2017.9.23) to the between request time (2017.5.15)
The online data belonged in the period of Second Type period is obtained in two time intervals, as right with the first line duration section
The the second line duration section answered, wherein, above-mentioned second line duration section can also include:September in this year all Thursdays second from the bottom
To Saturday, the history online data of September in this year all Thursdays to Saturday third from the bottom, etc..
Further, M with target online data are obtained respectively from the history online data in above-mentioned multiple second rounds
Corresponding multiple second same period rates of change of j-th of element in element, according to the average value of above-mentioned multiple second same period rates of change, such as
Average same period rate of change, the corresponding target online data of j-th of element in M element is obtained to predict.
The embodiment provided by the application, by obtaining in the recent period, averagely same period amount of increase is obtained when first is online to predict
Between section (as save before the period or section after the period) target online data so that reach improve acquired in target online data
Accuracy.
As a kind of optional scheme, first acquisition unit 404 includes:
1) the 3rd determining module, in the case of the type of time of the first line duration Duan Wei tri- section, it is determined that obtaining
Strategy is the 3rd strategy matched with the 3rd type of time section, wherein, the particular type period includes the 3rd type of time section,
3rd type of time section repeats according to the period 3;
2) the 3rd acquisition module, belongs to the 3rd type of time section for being obtained according to the 3rd strategy out of the 3rd time interval
Period in online data, as with the history in the first line duration section corresponding second line duration section in line number
According to, wherein, the 3rd time interval is the time interval between the 5th line duration section and request time, and the 5th line duration section exists
Occur before request time, and the time interval between the first line duration section is one or more period 3.
Alternatively, in the present embodiment, for the 3rd type of time section, identified 3rd strategy can be, but not limited to refer to
Show obtain the 3rd time interval in belong to the 3rd type of time section period in online data, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, above-mentioned 3rd type of time section can include but not limit
In:Repeat according to the period 3, by taking game application as an example, such as period racing season in cycle.
The embodiment provided by the application, in the case where the first line duration section is defined as the 3rd type of time section,
The 3rd strategy according to matching with the 3rd type of time section is obtained out of the 3rd time interval belongs to the 3rd type of time section
Online data in period, as the history online data in the second line duration section corresponding with the first line duration section,
So as to realize accurate target online data is got according to the 3rd strategy with the type matching of the first line duration section.
As a kind of optional scheme, second acquisition unit 406 includes:
1) the 4th acquisition module, in the case of the type of time of the first line duration Duan Wei tri- section, first is online
In the case that period includes P element, obtained respectively and P element from each period 3 in the 3rd time interval
In the corresponding history online data of k-th of element, wherein, history online data corresponding with k-th of element in P element
Include the 3rd same period rate of change of k-th of element, the 3rd same period rate of change of k-th of element is in the current period 3
K-th of element is relative to the online data rate of change of k-th of element in a upper period 3, and k is more than or equal to 1 and to be less than
Integer equal to P;
2) the 5th acquisition module, for according to history corresponding with k-th of element in P element in the 3rd time interval
Online data, obtains the average same period rate of change of k-th of element;
3) the 6th acquisition module, for being obtained according to average same period rate of change in the first line duration section in P element
K-th of element target online data.
Alternatively, in the present embodiment, for the 3rd type of time section, identified 3rd strategy can be, but not limited to refer to
Show obtain the 3rd time interval in belong to the 3rd type of time section period in online data, as with the first line duration
History online data in the corresponding second line duration section of section.Wherein, above-mentioned 3rd type of time section can include but not limit
In:Repeat according to the period 3, by taking game application as an example, such as period racing season in cycle.
The embodiment provided by the application, in the case where the first line duration section is defined as the 3rd type of time section,
The 3rd strategy according to matching with the 3rd type of time section is obtained out of the 3rd time interval belongs to the 3rd type of time section
Online data in period, as the history online data in the second line duration section corresponding with the first line duration section,
So as to realize accurate target online data is got according to the 3rd strategy with the type matching of the first line duration section.
As a kind of optional scheme, in addition to:
1) the 3rd acquiring unit, for after acquisition request is received, when the first line duration section is non specified type
Between in the case of section, according to characteristic of the intended application before request time, obtain the target in the first line duration section
Online data, wherein, characteristic includes following at least one:The registration amount of intended application, login amount, the target of intended application
The capacity of returns of the retention amount of application, the number of dropouts of intended application and intended application.
Alternatively, in the present embodiment, the first line duration section be the non specified type period in the case of, can with but
The acquisition model matched according to characteristic determination of the intended application before request time with the first line duration section is not limited to,
To obtain the above-mentioned target online data to be predicted using the acquisition model.
Alternatively, in the present embodiment, the 3rd acquiring unit includes:
(1) the 4th determining module, for according to characteristic of the intended application before request time determine with first
The acquisition model that the line period matches;
(2) the 7th acquisition modules, exist for obtaining the target in the first line duration section using the acquisition model determined
Line number evidence, wherein, in the case where characteristic is designated as linear relationship, first is obtained using the linear regression model (LRM) determined
Target online data in line duration section;Or, in the case where characteristic is designated as non-linear relation, using determining
Random Forest model obtain the first line duration section in target online data;Or, in the case where customization obtains model,
Use the target online data obtained in model acquisition the first line duration section of customization.
Alternatively, in the present embodiment, it can be, but not limited to determine to obtain model based on features described above data, further,
Use the target online data in the first line duration section for obtaining the model acquisition non specified type period determined.It is above-mentioned
Characteristic can include but is not limited to:Month feature, Zhou Tezheng and day feature etc..
Alternatively, in the present embodiment, above-mentioned acquisition model can include but is not limited to:Traditional linear regression model (LRM)
(Linear Regression), the acquisition model of Random Forest model (Random Forest) customization, using this flexible
Model is obtained, can not only ensure the precision of prediction, the algorithm complexity for obtaining model can also be reduced, reduces and obtains model
Maintenance cost.
The embodiment provided by the application, it is proposed that the thinking that strategy can match somebody with somebody, greatly simplifies acquisition model, so that
The business of realization is quickly accessed, the effect that low cost is safeguarded.
Embodiment 3
According to embodiments of the present invention, a kind of electronics dress for being used to implement the method for above-mentioned acquisition online data is additionally provided
Put, as shown in figure 5, the electronic installation includes:
1) communication interface 502, are set to receive to obtain and ask, wherein, obtain request and exist for acquisition request intended application
Target online data in first line duration section, the first line duration section is later than the request time that generation obtains request;
2) processor 504, are connected with communication interface 502, and it is the particular type period to be set in the first line duration section
In the case of, according to the acquisition strategy matched with the particular type period, obtain and the first line duration section corresponding second
History online data in line duration section, wherein, the second line duration section is earlier than request time;Be also configured to using second
History online data in the line period obtains the target online data in the first line duration section.
3) memory 506, are connected with communication interface 502 with processor 504, are set to store history online data.
Alternatively, the specific example in the present embodiment may be referred to showing described in above-described embodiment 1 and embodiment 2
Example, the present embodiment will not be repeated here.
Embodiment 4
Embodiments of the invention additionally provide a kind of storage medium.Alternatively, in the present embodiment, above-mentioned storage medium can
With at least one network equipment in multiple network equipments in network.
Alternatively, in the present embodiment, storage medium is arranged to the program code that storage is used to perform following steps:
S1, receives and obtains request, wherein, obtaining request is used for acquisition request intended application in the first line duration section
Target online data, the first line duration section is later than the request time that generation obtains request;
S2, the first line duration section be the particular type period in the case of, according to particular type period phase
The acquisition strategy matched somebody with somebody, obtains the history online data in the second line duration section corresponding with the first line duration section, wherein, the
Two line durations section is earlier than request time;
S3, using the target in history online data acquisition the first line duration section in the second line duration section in line number
According to.
Alternatively, in the present embodiment, above-mentioned storage medium can include but is not limited to:USB flash disk, read-only storage (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disc or
CD etc. is various can be with the medium of store program codes.
Alternatively, the specific example in the present embodiment may be referred to showing described in above-described embodiment 1 and embodiment 2
Example, the present embodiment will not be repeated here.
The embodiments of the present invention are for illustration only, and the quality of embodiment is not represented.
If the integrated unit in above-described embodiment is realized using in the form of SFU software functional unit and is used as independent product
Sale or in use, the storage medium that above computer can be read can be stored in.Understood based on such, skill of the invention
The part or all or part of the technical scheme that art scheme substantially contributes to prior art in other words can be with soft
The form of part product is embodied, and the computer software product is stored in storage medium, including some instructions are to cause one
Platform or multiple stage computers equipment (can be personal computer, server or network equipment etc.) perform each embodiment institute of the invention
State all or part of step of method.
In the above embodiment of the present invention, the description to each embodiment all emphasizes particularly on different fields, and does not have in some embodiment
The part of detailed description, may refer to the associated description of other embodiment.
, can be by others side in several embodiments provided herein, it should be understood that disclosed client
Formula is realized.Wherein, device embodiment described above is only schematical, such as division of described unit, only one
Kind of division of logic function, can there is other dividing mode when actually realizing, such as multiple units or component can combine or
Another system is desirably integrated into, or some features can be ignored, or do not perform.It is another, it is shown or discussed it is mutual it
Between coupling or direct-coupling or communication connection can be the INDIRECT COUPLING or communication link of unit or module by some interfaces
Connect, can be electrical or other forms.
The unit illustrated as separating component can be or may not be it is physically separate, it is aobvious as unit
The part shown can be or may not be physical location, you can with positioned at a place, or can also be distributed to multiple
On NE.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs
's.
In addition, each functional unit in each embodiment of the invention can be integrated in a processing unit, can also
That unit is individually physically present, can also two or more units it is integrated in a unit.Above-mentioned integrated list
Member can both be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (15)
1. a kind of method for obtaining online data, it is characterised in that including:
Receive and obtain request, wherein, described obtain asks to be used for mesh of the acquisition request intended application in the first line duration section
Online data is marked, the first line duration section, which is later than, generates the request time for obtaining request;
First line duration section be the particular type period in the case of, according to the particular type period phase
The acquisition strategy matched somebody with somebody, obtains the history online data in the second line duration section corresponding with first line duration section, its
In, the second line duration section is earlier than the request time;
Described in history online data acquisition the first line duration section in second line duration section
Target online data.
2. according to the method described in claim 1, it is characterised in that the basis matches with the particular type period
Acquisition strategy, the history online data obtained in the second line duration section corresponding with first line duration section includes:
In the case where first line duration section is the first kind period, determine that the acquisition strategy is and described first
The first strategy that type of time section matches, wherein, the particular type period includes the first kind period, described
Online data rate of change of the intended application in the first kind period is higher than the first predetermined threshold;
According to the described first strategy obtained out of the very first time interval in the period for belonging to the first kind period
Line number evidence, as the history online data in the second line duration section corresponding with first line duration section,
Wherein, the very first time at intervals of the 3rd line duration section and the request time between time interval, the described 3rd
The line period occurs before the request time, and the time interval between first line duration section is one or many
The individual period 1.
3. method according to claim 2, it is characterised in that gone through described in the utilization second line duration section
The target online data that history online data is obtained in the first line duration section includes:
In the case where first line duration section is the first kind period, by the institute in the very first time interval
The history online data stated in the second line duration section is divided into multiple N tuple sequences, wherein, each described N tuple
Sequence includes N number of element, and each element is used for the first same period rate of change for recording the history online data, described first
Same period rate of change is i-th of element in the current N tuple sequences relative to i-th in the previous N tuple sequences
The online data rate of change of element, i is the integer more than or equal to 1 and less than or equal to N;
Obtain the element in first object N tuple sequences corresponding with the 3rd line duration section;
The element and the first object N tuples in the multiple N tuple sequences in the very first time interval are compared successively
Element in sequence, is obtained and the closest object N of the first object N tuple sequences from the multiple N tuple sequences
Tuple sequence, is used as the first reference sequences of first line duration section;
The target online data is obtained according to first reference sequences and the history online data.
4. method according to claim 3, it is characterised in that it is described compare successively in the very first time interval it is described
Element in multiple N tuple sequences and the element in the first object N tuple sequences, are obtained from the multiple N tuple sequences
The object N tuple sequence closest with the first object N tuple sequences is taken, the reference of first line duration section is used as
Sequence includes:
Following steps are repeated, until traveling through the multiple N tuple sequences in the very first time interval:
Current N tuple sequences are obtained from the multiple N tuple sequences;
Obtain i-th of element in the current N tuple sequences and i-th of element in the first object N tuple sequences
Difference;
The current N tuple sequences and the first object N tuples are obtained using the quadratic sum of the difference of N number of element
The distance between sequence;
The object N tuple sequence conducts in the multiple N tuple sequences are determined according to the quadratic sum of the minimum difference
The reference sequences.
5. method according to claim 2, it is characterised in that the basis matches with the particular type period
Acquisition strategy, the history online data obtained in the second line duration section corresponding with first line duration section includes:
In the case where first line duration section is the Second Type period, determine that the acquisition strategy is and described second
The second strategy that type of time section matches, wherein, the particular type period includes the Second Type period, described
The Second Type period includes:It is adjacent with the first kind period and before the first kind period first
Sub- period, the second sub- period adjacent with the first kind period and after the first kind period;
According to described second strategy out of second time interval obtain belong in the period of the Second Type period
Line number evidence, as the history online data in the second line duration section corresponding with first line duration section,
Wherein, second time interval be the 4th line duration section and the request time between time interval, the described 4th
The line period occurs before the request time, and the time interval between first line duration section is one or many
Individual second round.
6. the method stated according to claim 5, it is characterised in that the history using in second line duration section
The target online data that online data is obtained in the first line duration section includes:
It it is the Second Type period in first line duration section, the first line duration section includes M element
In the case of, obtained respectively and the jth in the M element from each described second round in second time interval
The corresponding history online data of individual element, wherein, it is corresponding with j-th of element in the M element described to go through
History online data includes the second same period rate of change of j-th of element, second same period rate of change of j-th of element
It was j-th of element in the current second round relative to described j-th in a upper second round
The online data rate of change of element, j is the integer more than or equal to 1 and less than or equal to M, and M is less than or equal to the second round;
According to the history online data corresponding with j-th of element in the M element in second time interval, obtain
Take the average same period rate of change of j-th of element;
Described j-th in first line duration section in the M element is obtained according to the average same period rate of change
The target online data of element.
7. according to the method described in claim 1, it is characterised in that the basis matches with the particular type period
Acquisition strategy, the history online data obtained in the second line duration section corresponding with first line duration section includes:
In the case of the type of time of the first line duration Duan Wei tri- section, determine that the acquisition strategy is and the described 3rd
The 3rd strategy that type of time section matches, wherein, the particular type period includes the 3rd type of time section, described
3rd type of time section repeats according to the period 3;
According to the described 3rd strategy obtained out of the 3rd time interval in the period for belonging to the 3rd type of time section
Line number evidence, as the history online data in the second line duration section corresponding with first line duration section,
Wherein, the 3rd time interval be the 5th line duration section and the request time between time interval, the described 5th
The line period occurs before the request time, and the time interval between first line duration section is one or many
The individual period 3.
8. the method stated according to claim 7, it is characterised in that the history using in second line duration section
The target online data that online data is obtained in the first line duration section includes:
In the case where first line duration section is the 3rd type of time section, the first line duration section includes
In the case of P element, obtained and the P element respectively from each described period 3 in the 3rd time interval
In the corresponding history online data of k-th of element, wherein, it is corresponding with k-th of element in the P element
The history online data includes the 3rd same period rate of change of k-th of element, the 3rd same period of k-th of element
Rate of change was k-th of element in the current period 3 relative to described in a upper period 3
The online data rate of change of k-th of element, k is the integer more than or equal to 1 and less than or equal to P;
According to the history online data corresponding with k-th of element in the P element in the 3rd time interval, obtain
Take the average same period rate of change of k-th of element;
Described k-th in first line duration section in the P element is obtained according to the average same period rate of change
The target online data of element.
9. according to the method described in claim 1, it is characterised in that after the reception obtains request,
Also include:
In the case where first line duration section is the non specified type period, according to the intended application in the request
Characteristic before time, obtains the target online data in the first line duration section, wherein, the characteristic
According to including following at least one:The registration amount of the intended application, the login amount of the intended application, the intended application are stayed
The capacity of returns of storage, the number of dropouts of the intended application and the intended application.
10. method according to claim 9, it is characterised in that it is described according to the intended application in the request time
Characteristic before, the target online data obtained in the first line duration section includes:
Determined and first line duration section according to the characteristic of the intended application before the request time
The acquisition model matched;
The target online data in the first line duration section is obtained using the acquisition model determined, wherein,
In the case where the characteristic is designated as linear relationship, described first is obtained using the linear regression model (LRM) determined online
The target online data in period;Or, in the case where the characteristic is designated as non-linear relation, using true
The Random Forest model made obtains the target online data in the first line duration section;Or, it is described in customization
In the case of obtaining model, the target obtained using the acquisition model of customization in the first line duration section is online
Data.
11. a kind of device for obtaining online data, it is characterised in that including:
Receiving unit, request is obtained for receiving, wherein, it is online first that the acquisition request is used for acquisition request intended application
Target online data in period, the first line duration section, which is later than, generates the request time for obtaining request;
First acquisition unit, for first line duration section be the particular type period in the case of, according to it is described
The acquisition strategy that the particular type period matches, is obtained in the second line duration section corresponding with first line duration section
History online data, wherein, second line duration section is earlier than the request time;
Second acquisition unit, for being existed using the history online data acquisition described first in second line duration section
The target online data in the line period.
12. device according to claim 11, it is characterised in that the first acquisition unit includes:
First determining module, in the case of being the first kind period in first line duration section, it is determined that described obtain
It is the first strategy matched with the first kind period to take strategy, wherein, the particular type period includes described
First kind period, online data rate of change of the intended application in the first kind period is higher than the first predetermined threshold
Value;
First acquisition module, belongs to the first kind time for being obtained according to the described first strategy out of the very first time interval
Online data in the period of section, is used as the institute in the second line duration section corresponding with first line duration section
History online data is stated, wherein, the very first time is at intervals of the time between the 3rd line duration section and the request time
Interval, the 3rd line duration section occurs before the request time, and between first line duration section when
Between at intervals of one or more period 1;
The first acquisition unit, in addition to:Second determining module, for when first line duration section is Second Type
Between in the case of section, it is the second strategy matched with the Second Type period to determine the acquisition strategy, wherein, it is described
The particular type period includes the Second Type period, and the Second Type period includes:During with the first kind
Between section is adjacent and the first sub- period before the first kind period, it is adjacent with the first kind period and
The second sub- period after the first kind period;
Second acquisition module, belongs to the Second Type time for being obtained according to the described second strategy out of second time interval
Online data in the period of section, is used as the institute in the second line duration section corresponding with first line duration section
History online data is stated, wherein, second time interval is the time between the 4th line duration section and the request time
Interval, the 4th line duration section occurs before the request time, and between first line duration section when
Between at intervals of one or more second rounds, the second round is less than the period 1.
13. device according to claim 11, it is characterised in that the first acquisition unit includes:
3rd determining module, in the case of the type of time of the first line duration Duan Wei tri- section, it is determined that described obtain
It is the 3rd strategy matched with the 3rd type of time section to take strategy, wherein, the particular type period includes described
3rd type of time section, the 3rd type of time section repeats according to the period 3;
3rd acquisition module, belongs to the 3rd type of time for being obtained according to the described 3rd strategy out of the 3rd time interval
Online data in the period of section, is used as the institute in the second line duration section corresponding with first line duration section
History online data is stated, wherein, the 3rd time interval is the time between the 5th line duration section and the request time
Interval, the 5th line duration section occurs before the request time, and between first line duration section when
Between at intervals of one or more period 3.
14. a kind of storage medium, it is characterised in that the storage medium includes the program of storage, wherein, when described program is run
Perform the method described in any one of described claim 1 to 10.
15. a kind of electronic installation, including memory, processor and it is stored on the memory and can transports on the processor
Capable computer program, it is characterised in that the processor is performed in the claim 1 to 10 by the computer program
Any one described in method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710439790.7A CN107329882A (en) | 2017-06-12 | 2017-06-12 | Obtain method and apparatus, storage medium and the electronic installation of online data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710439790.7A CN107329882A (en) | 2017-06-12 | 2017-06-12 | Obtain method and apparatus, storage medium and the electronic installation of online data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107329882A true CN107329882A (en) | 2017-11-07 |
Family
ID=60195323
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710439790.7A Pending CN107329882A (en) | 2017-06-12 | 2017-06-12 | Obtain method and apparatus, storage medium and the electronic installation of online data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107329882A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108989802A (en) * | 2018-08-14 | 2018-12-11 | 华中科技大学 | A kind of quality estimation method and system of the HEVC video flowing using inter-frame relation |
CN111399746A (en) * | 2020-04-17 | 2020-07-10 | Oppo广东移动通信有限公司 | Split screen display method and device, mobile terminal and computer readable storage medium |
CN112494935A (en) * | 2020-12-14 | 2021-03-16 | 咪咕互动娱乐有限公司 | Cloud game platform pooling method, electronic equipment and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103685347A (en) * | 2012-09-03 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Method and device for allocating network resources |
CN104462270A (en) * | 2014-11-24 | 2015-03-25 | 华为软件技术有限公司 | Information recommendation method and device |
CN106492459A (en) * | 2016-10-17 | 2017-03-15 | 腾讯科技(深圳)有限公司 | A kind of data handling system, data processing method and data processing equipment |
-
2017
- 2017-06-12 CN CN201710439790.7A patent/CN107329882A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103685347A (en) * | 2012-09-03 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Method and device for allocating network resources |
CN104462270A (en) * | 2014-11-24 | 2015-03-25 | 华为软件技术有限公司 | Information recommendation method and device |
CN106492459A (en) * | 2016-10-17 | 2017-03-15 | 腾讯科技(深圳)有限公司 | A kind of data handling system, data processing method and data processing equipment |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108989802A (en) * | 2018-08-14 | 2018-12-11 | 华中科技大学 | A kind of quality estimation method and system of the HEVC video flowing using inter-frame relation |
CN111399746A (en) * | 2020-04-17 | 2020-07-10 | Oppo广东移动通信有限公司 | Split screen display method and device, mobile terminal and computer readable storage medium |
CN111399746B (en) * | 2020-04-17 | 2021-09-24 | Oppo广东移动通信有限公司 | Split screen display method and device, mobile terminal and computer readable storage medium |
CN112494935A (en) * | 2020-12-14 | 2021-03-16 | 咪咕互动娱乐有限公司 | Cloud game platform pooling method, electronic equipment and storage medium |
CN112494935B (en) * | 2020-12-14 | 2023-10-17 | 咪咕互动娱乐有限公司 | Cloud game platform pooling method, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109902849B (en) | User behavior prediction method and device, and behavior prediction model training method and device | |
Sun et al. | Contextual intent tracking for personal assistants | |
Ghosh et al. | Predicting influential users in online social networks | |
CN107818344A (en) | The method and system that user behavior is classified and predicted | |
CN108874823A (en) | The implementation method and device of intelligent customer service | |
CN108399564B (en) | Credit scoring method and device | |
CN109829818A (en) | Cash demand amount prediction technique, device, electronic equipment and readable storage medium storing program for executing | |
CN107329882A (en) | Obtain method and apparatus, storage medium and the electronic installation of online data | |
CN111144468A (en) | Power consumer information labeling method and device, electronic equipment and storage medium | |
CN109214730A (en) | Information-pushing method and device | |
CN110880127A (en) | Consumption level prediction method and device, electronic equipment and storage medium | |
CN109255000A (en) | A kind of the dimension management method and device of label data | |
CN110191183A (en) | Accurate intelligent method for pushing, system, device and computer readable storage medium | |
CN103530304A (en) | On-line recommendation method, system and mobile terminal based on self-adaption distributed computation | |
CN108228684A (en) | Training method, device, electronic equipment and the computer storage media of Clustering Model | |
CN114663198A (en) | Product recommendation method, device and equipment based on user portrait and storage medium | |
CN106909560A (en) | Point of interest sort method | |
CN109978575A (en) | A kind of method and device excavated customer flow and manage scene | |
CN109948773A (en) | The method and apparatus for generating information | |
CN116680090A (en) | Edge computing network management method and platform based on big data | |
CN114782062A (en) | Commodity recall optimization method and device, equipment, medium and product thereof | |
CN114429362A (en) | Advertisement product delivery method and device, electronic device and readable storage medium | |
CN112637793A (en) | Scene charging method, system, electronic equipment and storage medium based on 5G | |
CN113822734A (en) | Method and apparatus for generating information | |
Kim et al. | Revisit Prediction by Deep Survival Analysis |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171107 |
|
RJ01 | Rejection of invention patent application after publication |