CN108881333A - A kind of method and apparatus for predicting to enliven number of objects day - Google Patents
A kind of method and apparatus for predicting to enliven number of objects day Download PDFInfo
- Publication number
- CN108881333A CN108881333A CN201710324322.5A CN201710324322A CN108881333A CN 108881333 A CN108881333 A CN 108881333A CN 201710324322 A CN201710324322 A CN 201710324322A CN 108881333 A CN108881333 A CN 108881333A
- Authority
- CN
- China
- Prior art keywords
- day
- objects
- retention
- amount
- model
- 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
- 238000000034 method Methods 0.000 title claims abstract description 62
- 230000014759 maintenance of location Effects 0.000 claims abstract description 144
- 238000010992 reflux Methods 0.000 claims abstract description 65
- 238000007405 data analysis Methods 0.000 claims abstract description 21
- 238000004458 analytical method Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 4
- 230000002238 attenuated effect Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 206010067171 Regurgitation Diseases 0.000 claims 3
- 230000000717 retained effect Effects 0.000 abstract description 2
- 230000002354 daily effect Effects 0.000 description 74
- 238000010586 diagram Methods 0.000 description 16
- 230000006870 function Effects 0.000 description 10
- 230000002688 persistence Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 7
- 238000004891 communication Methods 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000005236 sound signal Effects 0.000 description 2
- 239000000470 constituent Substances 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002085 persistent effect Effects 0.000 description 1
- 230000007420 reactivation Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000004083 survival effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/535—Tracking the activity of the user
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/142—Network analysis or design using statistical or mathematical methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/147—Network analysis or design for predicting network behaviour
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Algebra (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Pure & Applied Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention discloses a kind of method and apparatus for predicting to enliven number of objects day.Present invention method includes：Pass through the first object reflux total amount in object reflux day model prediction (M+X), object reflux model is that modeling obtains after being analyzed by the statistics the M days reflux ratios for enlivening the first object in object day to the continuous day (N+1) in the day (M-N), and the first object is newly-increased object before the day (M-N)；The retention total amount of second object in the day model prediction (M+X) is retained by object, the retention amount for enlivening the second object in number of objects day according to object retention model in the day (N+1) carries out the model obtained after data analysis；Day number of objects is enlivened according to what the retention total amount for enlivening the second object of the reflux total amount and day (M+X) of first object in the day (M+X) in number of objects day in the day (M+X) determined the day (M+X).
Description
Technical Field
The invention relates to the field of computers, in particular to a method and equipment for predicting the number of daily active objects.
Background
The number of Active User terminals (Daily Active User, abbreviated as DAU) is usually counted within one day (counting day), and the number of User terminals (User terminals without repeated login) that log in or use a certain product is often used to reflect the operation condition of a website, an internet application or an internet game. The definition of activity may not be exactly the same for different products, for example, if the product is a client product with an account, the account login is usually used as an activity index, such as QQ, IM, online game. In the case of certain tool applications, which may be active by starting up, for example, a figure show, etc., the application needs to take at least one picture or use a cropping function to be active.
The DAU is an important index for measuring the number of the core user terminals of an application, so that the DAU prediction method has important significance for predicting the DAU, the number of the future user terminals of the application can be predicted by predicting the DAU, and reference data is provided for function improvement or function improvement of a product.
In the traditional method, the DAU in a period of time is decomposed into the number of newly-added user terminals and the number of old user terminals stored in the same day, the data of the number of the newly-added user terminals and the number of the old user terminals stored in the same day are analyzed to respectively obtain linear models of the retention rate of the newly-added user terminals and the retention rate of the old user terminals stored in the old user terminals, and the number of the newly-added user terminals and the number of the old user terminals stored in the old user terminals in a certain day in the future are predicted through the linear models to further predict the DAU in the day.
The new user terminal is the user terminal which uses the application for the first time every day, and the stock old user terminal is the reserve user terminal in the new user terminal, and after the application is started to be used, the user terminal of the application is continuously used after a period of time. The retention rate is obtained by dividing the number of the user terminals which remain active in the nth day after the first day by the number of the user terminals which are newly added in the first day after the first day.
Because the newly added user terminals are independent every day and are not coupled with each other, the prediction precision of the newly added retention rate is higher. However, since the old user terminals are not pure, in a relatively mature internet product, the old user terminals in stock may include periods of a release period, a shake period, a stabilization period, and the like, and the trends of the old user terminals in stock at each stage in the old user terminals are different, a linear model is established by obtaining the number of the old user terminals in stock in a period of time, the linear model is not accurate in predicting the old user terminals in stock, and for the mature product, the prediction accuracy of the entire DAU is very low because the old user terminals occupy a higher proportion in the DAU.
Disclosure of Invention
The embodiment of the invention provides a method and equipment for predicting the number of daily active objects, which are used for improving the accuracy of predicting the number of daily active objects.
In a first aspect, an embodiment of the present invention provides a method for predicting a number of daily active objects, including:
predicting the total reflux amount of a first object on the (M + X) th day through an object reflux model, wherein the object reflux model is obtained by modeling after counting the reflux rate of the first object in the daily active objects on the (N +1) th day from the M th day to the (M-N) th day through analysis, the first object is an object newly added before the (M-N) th day, X is a positive integer greater than or equal to 1, N is a positive integer greater than or equal to 1, and M is a positive integer greater than N;
predicting a total retention amount of a second object on the (M + X) th day through an object retention model, wherein the object retention model is obtained by performing data analysis according to the retention amount of a second object in the daily active object number in the (N +1) th day, and the second object is an object newly added after the (M-N) th day;
determining the number of daily active objects on the (M + X) th day according to the total reflow amount of the first object on the (M + X) th day and the total retention amount of the second object on the (M + X) th day in the number of daily active objects on the (M + X) th day.
In a second aspect, an embodiment of the present invention provides an apparatus for predicting a number of daily active objects, including:
a first determining module, configured to predict a total backflow amount of a first object on an (M + X) th day through an object backflow model, where the object backflow model is obtained by modeling after counting backflow rates of the first object in daily active objects on consecutive (N +1) th days from the (M) th day to the (M-N) th day are analyzed, the first object is an object newly added before the (M-N) th day, X is a positive integer greater than or equal to 1, N is a positive integer greater than or equal to 1, and M is a positive integer greater than or equal to N;
a second determining module, configured to predict a total retention amount of a second object on the (M + X) th day through an object retention model, where the object retention model is a model obtained by performing data analysis according to a retention amount of a second object in the number of daily active objects in the (N +1) th day, and the second object is an object newly added after the (M-N) th day;
a third determination module, for determining the number of active objects per day (M + X) according to the total reflow amount of the first object on the (M + X) th day and the total retention amount of the second object on the (M + X) th day determined by the second determination module.
In a third aspect, an embodiment of the present invention provides an apparatus for predicting a number of daily active objects, including:
a memory for storing computer executable program code;
an input/output interface, and
a processor coupled with the memory and the input/output interface;
wherein the program code comprises instructions which, when executed by the processor, cause the determining means to perform the method of the first aspect.
In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, which includes instructions that, when executed on a computer, cause the computer to perform the method of the first aspect.
In a fifth aspect, embodiments of the present invention provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of the first aspect.
According to the technical scheme, the embodiment of the invention has the following advantages:
in this embodiment, the total amount of backflow of the first object on the (M + X) th day may be predicted by an object backflow model, where the object backflow model is obtained by modeling after counting backflow rates of the first object in daily active objects on consecutive (N +1) th day from the (M) th day to the (M-N) th day through analysis, and the first object is an object newly added before the (M-N) th day; predicting the total retention amount of the second object on the (M + X) th day through an object retention model, wherein the object retention model is obtained by performing data analysis according to the retention amount of the second object in the daily active object number in the (N +1) th day; then, the number of active objects per day (M + X) is determined according to the total reflow amount of the first object on the (M + X) th day and the total retention amount of the second object on the (M + X) th day among the number of active objects per day (M + X) th day. In the embodiment of the invention, the newly added user terminal and the old user terminal in the DAU in the traditional mode are redefined, the newly added object before the (M-N) th day is a first object, and the newly added object after the (M-N) th day is a second object. In the embodiment, the future DAU is predicted by using the object backflow model and the object retention model, the model has high precision, the accuracy of predicting the number of daily active objects is improved, the resolvability is good, and the direct influences of the number of newly added objects, the retention rate of second objects and the backflow rate of first objects, which are concerned by internet products, on the number of daily active objects can be quantized.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings.
FIG. 1 is a schematic diagram illustrating steps of a method for predicting a number of daily active objects according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating steps of an embodiment of a method for predicting a number of daily active objects according to the present invention;
FIG. 3 is a schematic diagram of a curve fitted to the retention of a second subject in an embodiment of the invention;
FIG. 4 is a schematic illustration of a fitted curve of the reflux rate of the first object in an embodiment of the invention;
FIG. 5 is a diagram illustrating days in an object reflow model and an object persistence model in an embodiment of the invention;
FIG. 6 is a diagram illustrating steps of another embodiment of a method for predicting a number of daily active objects according to an embodiment of the present invention;
FIG. 7 is a schematic structural diagram of an embodiment of an apparatus for predicting a number of active daily objects according to an embodiment of the present invention;
FIG. 8 is a schematic structural diagram of another embodiment of an apparatus for predicting a number of active daily objects according to an embodiment of the present invention;
FIG. 9 is a schematic structural diagram of another embodiment of an apparatus for predicting a number of active daily objects according to an embodiment of the present invention;
fig. 10 is a schematic structural diagram of another embodiment of an apparatus for predicting the number of active daily objects according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides a method and equipment for predicting the number of daily active objects, which are used for improving the accuracy of predicting the number of daily active objects.
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments that can be derived from the embodiments of the present invention by a person of ordinary skill in the art are intended to fall within the scope of the present invention.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that the embodiments described herein may be practiced otherwise than as specifically illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
The embodiment of the invention provides a method for predicting the number of Daily Active objects, wherein the number of Daily Active objects in the embodiment of the invention can be understood as the number of Daily Active User terminals (DAU), and the method in the embodiment of the invention can predict the DAU under the condition of no manual intervention aiming at relatively mature Internet products (such as websites, Internet applications or network games and the like).
For convenience of understanding, words involved in the embodiments of the present invention are explained first:
number of daily active objects: it can be understood that the number of Active User terminals (Daily Active User, abbreviated as DAU) is usually counted within one day (counting day), and the number of User terminals that the User terminal logs in or uses a certain application (excluding the User terminal that logs in repeatedly) is often used to reflect the operation condition of the website, the internet application or the network game.
Object: a user terminal using or logging in to a certain application.
A first object: for the newly added object before the (M-N) th day, in the application scenario, the first object may also be understood as an old user terminal in this embodiment. Please refer to the data in table 1 below as an example, if M is day 110 and N is 4, all the newly added objects before day 106 are old ues.
A second object: new subjects before (M-N) days, please be understood in conjunction with table 1 below:
TABLE 1
Taking the data in table 1 above as an example, today is day 101, and table 1 above is the data of DAU which is about 10 days before today, the internet product is described by taking a game as an example, and as can be seen from the data in table 1 above, the game has been online for 110 days, and in table 1, the data in bold type represents the number of the user terminals which are newly added on the day. For example, the number of the user terminals newly added on the day 106 is 5000, and the user terminals newly added on the day can be understood as the number of the user terminals newly registered in the game on the day 106; the number of the newly added user terminals on the day 107 is 5300, and the number of the newly added user terminals on the day 106 is 2500 in the next day (namely, day 107); the number of the user terminals newly added on the day 108 is 5000, the number of the user terminals newly added on the day 106 to the number of the user terminals on the day 108 becomes 1800, the remaining amount of the user terminals newly added on the day 107 to the active user terminals on the next day (day 108) is 2700, and the number of the user terminals newly added on the day in the table is not repeated.
The new user terminal in the embodiment of the present invention refers to the new user terminal in the consecutive (N +1) days from the M th day to the (M-N) th day, or, taking table 1 as an example, when M is 101 and N is 4, the new user terminal is the new user terminal in the consecutive 5 days from the 101 th day to the 106 th day, that is, the data of the areas identified by the double-frame lines in table 1 are the number of the new user terminals, and the new user terminals before the (M-N) th day are all the old user terminals, that is, the data in the areas identified by the black frame lines of the thick user terminals in the upper table.
It should be noted that, the distinguishing point between the new user terminal and the old user terminal in the embodiment of the present invention is as follows: the newly added object before the (M-N) th day is an old user terminal, and the newly added object after the (M-N) th day is a second object, that is, a new user terminal.
To better understand the first object and the second object, taking an example of a scenario below, data of a DAU within two months before today is obtained, where the first object is an object newly added two months before, for example, today is 3 months and 2 days, and with today as a reference point, two months before is 1 months and 2 days, the terminal a downloads the application and registers on 1 month and 1 day, and the terminal a is the first object, that is, objects newly added (registered, used for the first time, etc.) before 1 month and 2 days are all the first objects; on day 1/2 month, terminal B downloads the application and registers it, and terminal B is the second object, that is, the object newly added (registered, first used, etc.) after day 2/1 month is the second object.
It should be noted that, in the embodiment of the present invention, the definition of the new ue and the old ue is not fixed, but determined according to the number of days included in the acquisition history DAU data, and the data in table 1 in the embodiment is an exemplary description of the new ue and the old ue, and does not make a limiting description of the present invention.
The method in the embodiment of the invention respectively establishes an object reflux model and an object retention model by acquiring the data of the historical DAU and analyzing the data of the historical DAU, wherein the object reflux model is established after training through the data of a first object (an old user terminal) in the DAU data; and the object persistence model is established after training through the data of the second object (new user terminal) in the DAU data.
Please refer to fig. 1 for understanding, fig. 1 is a schematic step diagram illustrating a method for predicting the number of daily active objects according to an embodiment of the present invention. 1) Predicting a total amount of reflow of the first subject on the (M + X) th day by a subject reflow model; 2) predicting a total amount of retention of the second subject on the (M + X) th day by the subject retention model; 3) the number of daily active objects on the (M + X) th day is determined from the total reflow amount and the total retention amount. The above steps 1) and 2) are not limited in terms of time sequence.
Referring to fig. 2, the method for predicting the number of active daily objects according to an embodiment of the present invention is described in detail below, and an embodiment of the method for predicting the number of active daily objects according to an embodiment of the present invention includes:
step 201, obtaining data of the number of daily active objects of continuous (N +1) days from the M th day to the (M-N) th day, wherein N is a positive integer greater than or equal to 60, and the number of daily active objects of each day comprises the number of second objects and the remaining amount of first objects except for newly added objects in the daily active objects.
Assuming that day M is today, data of the number of active objects (which may also be understood as DAUs) in day 60 consecutive days before today is obtained, taking the example in table 1 as an example for explanation, where M is 110, N is 4, and (M-N) is 106, that is, before establishing the model, historical data of DAUs needs to be obtained.
It should be noted that, in practical applications, N is a positive integer greater than or equal to 59, that is, in practical applications, DAU data of greater than or equal to 60 days is acquired. For example, DAU data for 60 days may be acquired, DAU data for 90 days may be acquired, and the value of N is not particularly limited in practical applications. It can be understood that, taking day 110 as an example, it is found that the new objects are added at day 60 (day 50), and after 59 days of silence, the proportion of the user terminal objects of the active user terminals at day 110 in the DAU at day 110 is less than 1%, and this part of the user terminals can be ignored for reducing complexity without affecting the overall prediction accuracy. From practical applications, this assumption is also reasonable, and the probability of reactivation of a user terminal that is silent for a long time is low. Therefore, in practical applications, N may be a positive integer greater than or equal to 59, and for convenience of description in this embodiment, N is 4 for example.
Step 202, calculating the retention rate of the second object according to the retention amount of the second object in the daily active objects of (N +1) days.
In this embodiment, S is any value from day 106 to day 110, and the retention rate of the second object is calculated according to the number of the second objects in the DAU data of day 60, and the retention rate is the probability of the newly added object being retained as the name implies, and the retention rate is often used for representing the stickiness of the user.
Determining a first quantity of second objects on the day S, wherein S takes any value from the day M to the day (M-N), and then determining a first reserved quantity after attenuation of the first quantity from the day S to the day Y; calculating a newly-increased retention rate of the number of newly-increased objects on the S day on the Y day according to the first number and the first retention amount, wherein Y is greater than S and is smaller than M;
taking the data in table 1 as an example, the number of objects newly added on the day 109 is 5000, the 5000 is the first number, and by the day 110, the user terminal is 2500, and the 2500 is the first remaining amount after 5000 decays. That is, in this embodiment, the S day may be 106 days, 107 days, 108 days, and the Y day is 109 days, and on the 109 days, the number of the newly registered user terminals is 5000, but only 2500 user terminals continue to log in the game on the second day (i.e., 110 days), that is, the retention rate of the next day is: the retention rate of the next day is 50% when the retention rate is divided by 5000.
In this embodiment, taking the newly added objects from day 106 to day 110 as the second object and taking the retention rate of day 110 as an example, the retention rate of the second object is calculated as shown in table 2 below:
TABLE 2
It should be noted that, in this embodiment, the retention rate of the second object is calculated by the DAU on one day, that is, on 110 th day, and the calculation method of the retention rate of the second object in the DAU on 109 th day, 108 th day, etc. may be understood by combining the retention rate of the DAU on 110 th day, the retention rate of the newly added object on 106 th day on 110 th day is 0.12, the retention rate of the newly added object on 107 th day on 110 th day is 0.19, the retention rate of the newly added object on 108 th day on 110 th day is 0.36, and the retention rate of the newly added object on 109 th day on 110 th day is 0.50.
And step 203, performing data analysis on the retention rate data of the second object to obtain an object retention model.
And performing curve fitting on the data of the retention rate of the second object calculated in the step 202 to obtain an object retention model.
The method for curve fitting of the data of the retention rate of the second object in this embodiment may be an R language, SigmaPlot, SPSS, SAS, and the like, the method for curve fitting in this embodiment may be described by taking the R language as an example, and the R language may provide some integrated statistical tools and provide various functions of mathematical computation and statistical computation, so that data analysis can be flexibly performed.
As will be understood from fig. 3, fig. 3 is a schematic diagram of a curve fitted to the retention rate of the second object, and fig. 3 is a schematic diagram of a curve fitted to the retention rate of the second object in the DAU for 60 days. The retention rate curve of the second object can be expressed by equation 1:
rni＝a·ln(i)+b，
wherein, a and b constant coefficients can be obtained according to the curve fitting result, and i in the expression represents the ith day.
The object persistence model can be derived from 1 above as follows:
wherein n isiDenotes the newly added object on day i, rniIndicating the retention rate of the newly added object on day i.
In this embodiment, the object persistence model only needs to be trained by using historical data (e.g., 60-day DAU data) in a recent period of time, and focuses on the DAU development trend in the recent period of time, so that the training is simple and fast.
And step 204, counting the reflux amount of the first object in the daily active objects in the (N +1) days.
Determining the number of reflowed first objects in day-active objects from day M to day y of (M-N), the reflowed first objects being objects that are active on day N but silent (y-N-1) after day N, the objects that are active on day y, the first objects being objects that are newly added on any day before day (M-N).
In practical application, the reflux amount of a first object in the daily active objects of each day of 60 days can be counted, wherein the first object is an object newly added before the (M-N) th day. In this embodiment, for the sake of illustration, the data size of the DAU in this embodiment is small, and also taking the example in table 1 as an example, in this embodiment, the first object includes: newly added objects on day 105, newly added objects on day 104; newly added subjects on day 103; objects newly added on day 102; new objects on day 101.
As an example of a scenario, terminal C is an object newly added on day 105, and is active on day 108 after being silenced for two days, day y is day 108, and day n is day 105, it can be understood that after terminal C registers the application on day 105, it is silenced (without re-registering or using the application) for (y-n-1) days, that is, after being silenced for 2 days, it is active on day y (day 108), and the first objects such as terminal C are all the first objects of the reflow. It is understood that the number of objects re-registered or used by the application on day 108 is the amount of reflow of the first object, and the amount of reflow of the first object on day 105 in the DAU on day 108 is 400.
As another example, the number of subjects who were silent for 3 days in day 104, active for 500 days on day 107, and the reflux for day 103 in the DAU for day 107 was 500.
It should be noted that the data of the reflux rate of the first object are only given as examples, and do not limit the present invention, but in practical applications, the reflux amount of the first object on the nth day in the DAU on each day may be counted. For example, the reflux amount of the first subject that was silent at day 102 and active at day 107 in the DAU on day 107 was counted; reflux of the first subject that was silent at day 103 in the DAU on day 107 for 3 days, active on day 107; day 104 in DAUs on day 107 was silent for 2 days, reflux for the first subject active on day 107, and so on.
And step 205, performing data analysis on the reflux quantity of the first object to obtain an object reflux model.
The reflux rate of the first object is then calculated from the amount of reflux of the first object, which is active on the day n, but which is silenced (y-n-1) days after the day n, the total number of all first objects and the number of first objects refluxed on the day y. The reflux rate of the first object is calculated by the following formula 2:
wherein, BnAmount of reflux as first object: (ii) the number of first subjects in the DAU on day y that were silenced (y-n-1) after day n, were active on day y; snRepresents the number of all first subjects that were silenced for (y-n-1) days after day n in the day y DAU.
This index (reflux rate of the first subject) reflects the probability that subjects who are silent for (y-n-1) days will continue to survive. After the reflux rate of the first object in the last (N +1) days (for example, 60 days, in this embodiment, 5 days) is calculated according to the above formula, the calculated reflux rate data is fitted using the R language, and the fitting curve adopts a power curve, the abscissa is the number of days, and the ordinate is the reflux rate of the old user terminal.
Fig. 4 is a schematic diagram of a fitted curve of the reflux rate of the first object, and fig. 4 is a schematic diagram of a fitted curve of the reflux rate of the first object in the DAU for 60 days, where it should be noted that the data in the fitted curve in fig. 4 is only an example for convenience of description, and does not limit the present invention. The expression formula 3 of the reflux rate curve of the first object may be:
roi＝a·ib，
wherein, roiIndicates the ith dayThe reflux rate of an object, a constant coefficient and a constant coefficient are obtained according to a curve fitting result.
The object reflow model of the first object is obtained according to the above equation 3 as follows:
wherein m isiDenotes the amount of all first objects remaining on day i, roiThe first object reflux rate on day i is indicated.
It should be noted that, step 201 to step 205 are steps of establishing an object persistence model and an object reflow model, and after the establishment of the object persistence model and the object reflow model is completed by acquiring data of the number of daily active objects of consecutive (N +1) days from the M-th day to the (M-N) -th day, the DAU can be predicted through the object persistence model and the object reflow model without repeatedly performing the steps of establishing the object persistence model and the object reflow model, and therefore, step 201 to step 205 are optional steps, and may not be performed, but directly perform step 206.
And step 206, predicting the total reflux amount of the first object on the (M + X) th day through an object reflux model, wherein the object reflux model is obtained by modeling after counting the reflux rate of the first object in the daily active objects on the (N +1) th day from the (M) th day to the (M-N) th day through analysis, the first object is an object newly added before the (M-N) th day, X is a positive integer greater than or equal to 1, N is a positive integer greater than or equal to 1, and M is a positive integer greater than N.
Predicting the total reflow amount of the first object through the object reflow model:
wherein m isiRepresents the amount of all first objects remaining before day i, roiRepresenting the first object reflux rate on day i; (N +1) represents the total number of days for which DAU data was acquired。
When i is (M + X), the total amount of backflow for the first subject on the (M + X) th day is predicted.
As will be understood in conjunction with fig. 5, fig. 5 is a schematic diagram showing the number of days in the object reflow model and the object persistence model. For example, the M th day is today, if today is taken as a reference day, the first day of the (M +1) th day in the future is taken as the 111 th day in terms of the online time of the application, and in this embodiment, the i th day is the (M +1) th day, that is, the total amount of backflow of the first object from the 101 th day to the 110 th day is calculated. By predicting the total reflow amount of the first object on the i-th day through the object reflow model, it can be understood that the following table 3 is an example of the time and actual time of the i-th day and the application line:
TABLE 3
| i | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
| Days of line | Day 101 | Day 102 | Day 103 | Day 104 | Day 105 | Day 106 | Day 107 | Day 108 | Day 109 | Day 110 |
| Date | 3 month and 1 day | 3 month and 2 days | 3 months and 3 days | 3 months and 3 days | 3 month and 4 days | 3 month and 5 days | 3 month and 6 days | 3 month and 7 days | 3 month and 8 days | 3 month and 9 days |
When it is necessary to predict day 111 (i ═ 11), it is necessary to sum the total amount of backflow of the first object on day 111 by calculating the sum of the amounts of backflow of the first object on days 101 to 110. That is, i is any value after the M day, and is not a constant value but a variable.
In addition, the number of days, i, and dates of the line on in table 3 are for convenience of understanding, and are not intended to limit the present invention.
And step 207, predicting the total retention amount of the second object on the (M + X) th day through an object retention model, wherein the object retention model is obtained by performing data analysis according to the first object number in the daily active object number in the (N +1) th day, and the first object is an object newly added in the (N +1) th day to the (M-N) th day.
Predicting the total amount of retention of the second subject by:
wherein n isiDenotes the retention of the newly added second object on day i, rniRepresenting the retention rate of the second object on day i;
when i ═ M + X, the total amount of retention of the second subject on day (M + X) was predicted.
Please refer to fig. 5, for example, the M th day is today, if today is taken as a reference day, the first future day of the (M +1) th day is the (M +1) th day 111 according to the online time of the application, in this embodiment, the i th day is the (M +1) th day, that is, the total amount of the second object remaining from the 101 th day to the 110 th day is calculated. Predicting the total amount of retention of the second object on day i by the object retention model.
Step 208, determining the number of active objects per day (M + X) according to the total reflow amount of the first object per day (M + X) and the total retention amount of the second object per day (M + X) in the number of active objects per day (M + X).
The number of daily active objects on the (M + X) th day is determined by the following formula 3.
DAU＝P+Q。
In this embodiment, newly added ues and old ues in the DAU in the conventional manner are redefined, where an object newly added before the (M-N) th day is a first object, and an object newly added after the (M-N) th day is a second object. In this embodiment, the first object is further structurally decomposed according to the number of silent days by using the acquired data of the historical DAU for the consecutive (N +1) days from the M-th day to the (M-N) -th day, and then the reflux rate of the first object is curve-fitted to obtain an object reflux model; and analyzing the retention rate of the second object to obtain an object retention model. The model not only has high precision, but also has good resolvability, and can quantify the number of newly added objects concerned by Internet products, the retention rate of second objects and the direct influence of the reflux rate of the first object on the DAU. The prediction accuracy of DAU tested in this example was within 4% over 90 days.
Based on the foregoing embodiments, please refer to fig. 6, in which another embodiment of the method for predicting the number of active daily objects according to the present invention includes:
in step 208, the (M + X) th DAU may be predicted by DAU ═ P + Q, where,however, if the (M + X +1) th DAU needs to be predicted, M in P needs to be predictediUpdate to predict the DAU one day after the starting day, since there were already ro on day i beforeiWhen the user comes back to the current day, the total number of the first objects on the ith day needs to be updated to be:
m′＝mi·(1-roi)，
it is understood that if the (M +1) th day is used as the prediction starting point, M in P needs to be predicted when predicting the DAU of the (M +2) th dayiUpdating is performed and then iterative prediction is performed.
Predicting a total amount of reflow for the first object by:
when i is (M + X +1), the total amount of backflow of the first subject on the (M + X +1) th day is predicted.
The above describes a method for predicting the number of active objects per day, and an apparatus applied by the method is described below, because the apparatus in this embodiment acquires relatively less DAU data, and thus has a low requirement on the processing capability of the apparatus, the apparatus may exist in the form of a PC or a server, please refer to fig. 7 for understanding, where fig. 7 is an embodiment of an apparatus 700 for predicting the number of active objects per day in the embodiment of the present invention, and the embodiment includes:
a first determining module 701, which predicts a total amount of backflow of a first object on an (M + X) th day through an object backflow model, wherein the object backflow model is obtained by modeling after counting backflow rates of the first object in daily active objects on consecutive (N +1) th days from the (M) th day to the (M-N) th day through analysis, the first object is an object newly added before the (M-N) th day, X is a positive integer greater than or equal to 1, N is a positive integer greater than or equal to 1, and M is a positive integer greater than or equal to N;
a second determining module 702, configured to predict a total retention amount of a second object on the (M + X) th day through an object retention model, where the object retention model is a model obtained by performing data analysis according to a retention amount of a second object in the number of daily active objects in the (N +1) th day, and the second object is an object newly added after the (M-N) th day;
the third determining module 703 determines the number of active objects per day on (M + X) according to the total reflow amount of the first object on (M + X) th day among the number of active objects per day on (M + X) th day of the first determining module 701 and the total retention amount of the second object on (M + X) th day determined by the second determining module 702.
Based on the embodiment shown in fig. 7, please refer to fig. 8, in which an embodiment of an apparatus 800 for predicting the number of active daily objects according to the present invention further includes:
an obtaining module 704, a calculating module 705, a first model establishing module 706, a statistical module 707 and a second model establishing module 708;
an obtaining module 704, configured to obtain the number of daily active objects for consecutive (N +1) days from the M-th day to the (M-N) th day, where N is a positive integer greater than or equal to 60, and the number of daily active objects per day includes the number of second objects and the remaining amount of first objects, except for newly added objects, in the daily active objects;
a calculating module 705, configured to calculate a retention rate of the second object according to the retention amount of the second object in the daily active objects of (N +1) days acquired by the acquiring module 704;
a first model establishing module 706, configured to perform data analysis on the data of the retention rate of the second object calculated by the calculating module 705 to obtain an object retention model;
a counting module 707 configured to count a reflux amount of a first object in daily active objects of (N +1) days acquired by the acquiring module 704;
the second model establishing module 708 is configured to perform data analysis on the reflow amount of the first object counted by the counting module 707, so as to obtain an object reflow model.
Optionally, the calculating module 705 is further configured to determine a first number of newly added objects on the day S, where S is taken over any value from the day M to the day (M-N); determining a first amount of the first quantity attenuated from the S day to the Y day; and calculating the retention rate of the number of the newly-added second objects on the day S on the day Y according to the first number and the first retention amount, wherein Y is greater than S and is less than M.
Optionally, the first model establishing module 706 is further configured to obtain an object retention model by performing curve fitting on data of a retention rate of the second object.
Optionally, the statistical module 707 is further configured to determine the number of reflowed first objects in the daily active objects from day M to day y of (M-N), wherein the reflowed first objects are objects that are active on day N but are silent (y-N-1) after day N and are active on day y.
Optionally, the second model building module 708 is further configured to calculate a reflux rate of the first object according to the number of all first objects that are active on the nth day but silenced for (y-n-1) day after the nth day and the number of first objects that reflux on the y day; and performing curve fitting on the data of the reflux rate to obtain an object retention model of the first object.
Optionally, the first determining module 701 is further configured to predict a total reflow amount of the first object through the following model:
wherein m isiDenotes the amount of retention, ro, of all first subjects on day iiRepresenting the first object reflux rate on day i;
when i is (M + X), the total amount of backflow for the first subject on the (M + X) th day is predicted.
Optionally, the second determining module 702 is further configured to predict the second object retention amount by the following model:
wherein n isiRepresenting a newly added second object, rn, on day iiRepresenting the retention rate of the second object on day i;
when i ═ M + X, the total amount of retention of the second subject on day (M + X) was predicted.
Optionally, the first determining module 701 is further configured to determine miIndicating that the retention amount of the first object on the ith day is updated, wherein the retention amount of the first object after updating is m' ═ mi·(1-roi)；
Predicting a total amount of reflow for the first object by:
when i is (M + X +1), the total amount of backflow of the first subject on the (M + X +1) th day is predicted.
Further, one of the devices of fig. 7 and 8 that predicts the number of daily active objects is presented in the form of a functional module. A "module" as used herein may refer to an application-specific integrated circuit (ASIC), an electronic circuit, a processor and memory that execute one or more software or firmware programs, an integrated logic circuit, and/or other devices that provide the described functionality. In a simple embodiment, one of the devices of fig. 7 and 8 for predicting the number of daily active objects may take the form of a terminal as shown in fig. 9.
The embodiment of the present invention further provides another device for predicting the number of daily active objects, as shown in fig. 9, the device in fig. 9 exists in a terminal form, and for convenience of description, only the part related to the embodiment of the present invention is shown, and details of the specific technology are not disclosed, please refer to the method part of the embodiment of the present invention. The terminal may be a PC for example.
Memory 920, input unit 930, display unit 940, audio circuit 960, processor 980, and power supply 990. Those skilled in the art will appreciate that the configuration shown in fig. 9 is not intended to be limiting of the apparatus and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components.
The following describes the respective constituent components of the apparatus in detail with reference to fig. 9:
the memory 920 may be used to store software programs and modules, and the processor 980 performs various functional applications of the device and data processing by operating the software programs and modules stored in the memory 920. The memory 920 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data (such as audio data, a phonebook, etc.) created according to the use of the device, and the like. Further, the memory 920 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.
The input unit 930 is operable to receive input numeric or character information and generate key signal inputs related to user settings and function control of the apparatus. Other input devices 932. In particular, other input devices 932 may include, but are not limited to, one or more of a physical keyboard, function keys (e.g., volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and the like.
The display unit 940 may be used to display information input by a user or information provided to the user and various menus of the device. The display unit 940 may include a display panel 941, and optionally, the display panel 941 may be configured in the form of a Liquid Crystal Display (LCD), an Organic Light-Emitting Diode (OLED), or the like.
Audio circuitry 960, speaker 961, microphone 962 may provide an audio interface between a user and the device. The audio circuit 960 may transmit the electrical signal converted from the received audio data to the speaker 961, and convert the electrical signal into a sound signal for output by the speaker 961; on the other hand, the microphone 962 converts the collected sound signal into an electric signal, converts the electric signal into audio data after being received by the audio circuit 960, outputs the audio data to the memory 920 after being processed by the audio data output processor 980, and further processes the audio data.
The processor 980 is a control center of the apparatus, connects various parts of the entire apparatus using various interfaces and lines, and performs various functions of the apparatus and processes data by running or executing software programs and/or modules stored in the memory 920 and calling data stored in the memory 920, thereby integrally monitoring the apparatus. Alternatively, processor 980 may include one or more processing units; preferably, the processor 980 may integrate an application processor, which primarily handles operating systems, user interfaces, applications, etc., and a modem processor, which primarily handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 980.
The processor 980 is further configured to perform the following steps:
predicting the total reflux amount of a first object on the (M + X) th day through an object reflux model, wherein the object reflux model is obtained by modeling after counting the reflux rate of the first object in a daily active object from the M th day to the (M-N) th day for continuous (N +1) th day, the first object is an object newly added before the (M-N) th day, X is a positive integer greater than or equal to 1, N is a positive integer greater than or equal to 1, and M is a positive integer greater than N;
predicting the total retention amount of a second object on the (M + X) th day through an object retention model, wherein the object retention model is a model obtained by performing data analysis according to the retention amount of the second object in the daily active object number in the (N +1) th day, and the second object is an object newly added after the (M-N) th day;
the number of active objects per day (M + X) is determined according to the total reflow amount of the first object per day (M + X) and the total survival amount of the second object per day (M + X) among the number of active objects per day (M + X).
Optionally, the processor 980 is further configured to obtain the number of daily active objects for consecutive (N +1) days from the M th day to the (M-N) th day, where N is a positive integer greater than or equal to 60, and the number of daily active objects per day includes the second number of objects and the remaining amount of the first objects, excluding the newly added objects, in the daily active objects; calculating a retention rate of a second object from retention amounts of the second object among the daily active objects for (N +1) days; performing data analysis on the retention rate data of the second object to obtain an object retention model; counting the backflow amount of the first object in the daily active objects of each day in the (N +1) days; and carrying out data analysis on the backflow amount of the first object to obtain an object backflow model.
Optionally, the processor 980 is further configured to determine a first number of newly added objects on the day S, where S is taken over any value from the day M to the (M-N); determining a first amount of the first quantity attenuated from the S day to the Y day; and calculating the retention rate of the number of the newly-added second objects on the day S on the day Y according to the first number and the first retention amount, wherein Y is greater than S and is less than M.
Optionally, the processor 980 is further configured to obtain an object retention model by performing curve fitting on the data of the retention rate of the second object.
Optionally, the processor 980 is further configured to determine a number of reflowed first objects in the daily active objects on day M to day y of (M-N), the reflowed first objects being objects that are active on day N but silent (y-N-1) after day N, and are active on day y.
Optionally, the processor 980 calculates a reflow rate of the first objects based on the number of all first objects that are active on day n but silenced for (y-n-1) day (y-n-1) after day n and the number of first objects reflowed on day y; and performing curve fitting on the data of the reflux rate to obtain an object retention model of the first object.
Optionally, the processor 980 is further configured to predict a total reflow amount of the first object by the following model:
wherein m isiDenotes the amount of retention, ro, of all first subjects on day iiRepresenting the first object reflux rate on day i; when i is (M + X), the total amount of backflow for the first subject on the (M + X) th day is predicted.
Optionally, the processor 980 is further configured to predict a second object retention amount by:
wherein,nirepresenting a newly added second object, rn, on day iiRepresenting the retention rate of the second object on day i; when i ═ M + X, the total amount of retention of the second subject on day (M + X) was predicted.
Optionally, processor 980 is also configured to process miIndicating that the retention amount of the first object on the ith day is updated, wherein the retention amount of the first object after updating is m' ═ mi·(1-roi)；
Predicting a total amount of reflow for the first object by:
when i is (M + X +1), the total amount of backflow of the first subject on the (M + X +1) th day is predicted.
Embodiments of the present invention further provide a computer readable storage medium for storing computer software instructions for the apparatus for predicting the number of active daily objects shown in fig. 9, which contains a program designed to execute the method embodiments. By executing the stored program, a prediction of the number of daily active objects may be achieved.
Embodiments of the present invention also provide a computer program product containing instructions, which when run on a computer, cause the computer to perform the method in the above method embodiments.
Another apparatus 1000 for predicting the number of active daily objects is provided in the embodiments of the present invention, as shown in fig. 10, the apparatus in fig. 10 exists in the form of a server, and fig. 10 is a schematic structural diagram of a server provided in the embodiments of the present invention, where the server may generate a relatively large difference due to different configurations or performances, and may include one or more processors 1022 and a memory 1032, and one or more storage media 1030 (e.g., one or more mass storage devices) for storing an application 1042 or data 1044. Memory 1032 and storage medium 1030 may be, among other things, transient or persistent storage. The program stored on the storage medium 1030 may include one or more modules (not shown), each of which may include a series of instruction operations for the server. Still further, the processor 1022 may be arranged in communication with the storage medium 1030 to execute a series of instruction operations in the storage medium 1030 on a server.
The Server may also include one or more power supplies 1026, one or more wired or wireless network interfaces 1050, one or more input-output interfaces 1058, and/or one or more operating systems 1041, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc.
The steps performed by the apparatus for predicting the number of daily active objects in the above embodiment may be based on the server structure shown in fig. 10.
The processor 1022 is configured to execute the method performed by the apparatus for predicting the number of the daily active objects in the above method embodiments.
Embodiments of the present invention further provide a computer readable storage medium for storing computer software instructions for the apparatus for predicting the number of active daily objects shown in fig. 10, which contains a program designed to execute the method embodiments. By executing the stored program, a prediction of the number of daily active objects may be achieved.
Embodiments of the present invention also provide a computer program product containing instructions, which when run on a computer, cause the computer to perform the method in the above method embodiments.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (14)
1. A method of predicting the number of daily active objects, comprising:
predicting the total reflux amount of a first object on the (M + X) th day through an object reflux model, wherein the object reflux model is obtained by modeling after counting the reflux rate of the first object in the daily active objects on the (N +1) th day from the M th day to the (M-N) th day through analysis, the first object is an object newly added before the (M-N) th day, X is a positive integer greater than or equal to 1, N is a positive integer greater than or equal to 1, and M is a positive integer greater than N;
predicting a total retention amount of a second object on the (M + X) th day through an object retention model, wherein the object retention model is obtained by performing data analysis according to the retention amount of a second object in the daily active object number in the (N +1) th day, and the second object is an object newly added after the (M-N) th day;
determining the number of daily active objects on the (M + X) th day according to the total reflow amount of the first object on the (M + X) th day and the total retention amount of the second object on the (M + X) th day in the number of daily active objects on the (M + X) th day.
2. The method of claim 1, wherein prior to predicting the total amount of reflow of the first object on the (M + X) th day by the object reflow model, the method further comprises:
acquiring the number of daily active objects of continuous (N +1) days from the M day to the (M-N) day, wherein N is a positive integer greater than or equal to 60, and the number of daily active objects of each day comprises the number of second objects and the remaining amount of first objects except the newly added objects in the daily active objects;
calculating a retention rate of a second object from the retention amount of the second object in the (N +1) -day active objects;
performing data analysis on the retention rate data of the second object to obtain an object retention model;
counting the backflow amount of the first object in the daily active objects of each day in the (N +1) days;
and carrying out data analysis on the reflux quantity of the first object to obtain an object reflux model.
3. The method of claim 2, wherein calculating a retention rate for a second object based on a second number of objects in the (N +1) day's active-on-day objects comprises:
determining a first number of newly added objects on the day S, wherein S takes any value from the day M to the day (M-N);
determining a first amount of said first quantity attenuated from said S to Y days;
calculating a retention rate of the number of newly added second objects on the S day on the Y day according to the first number and the first retention amount, wherein Y is larger than S, and Y is smaller than M.
4. The method of claim 2, wherein the performing data analysis on the data of the retention rate of the second object to obtain an object retention model comprises:
and performing curve fitting on the data of the retention rate of the second object to obtain the object retention model.
5. The method of claim 2, wherein said counting the amount of reflux of the first one of the daily active objects for each of the (N +1) days comprises:
determining the number of reflowed first objects in day-active objects from day M to day y of (M-N), the reflowed first objects being objects that are active on day N but silenced for day (y-N-1) after day N, active on day y.
6. The method of claim 2, wherein the analyzing the data of the reflow quantity of the first object to obtain an object reflow model comprises:
calculating a reflux rate for the first subject based on the number of all first subjects that were active on day n but silenced for (y-n-1) day after day n and the number of first subjects that refluxed on day y;
and performing curve fitting on the data of the reflux rate to obtain an object retention model of the first object.
7. The method of claim 1, wherein predicting the total amount of regurgitation of the first subject on (M + X) th day by the subject regurgitation model comprises:
predicting a total amount of reflow for the first object by:
wherein m isiDenotes the amount of retention, ro, of all first subjects on day iiRepresenting the first object reflux rate on day i;
when i is (M + X), the total amount of backflow for the first subject on the (M + X) th day is predicted.
8. The method of claim 1, wherein predicting a second total object retention amount for a (M + X) th day by an object retention model comprises:
predicting a second object retention total by:
wherein n isiRepresenting a newly added second object, rn, on day iiRepresenting the retention rate of the second object on day i;
when i ═ M + X, the total amount of retention of the second subject on day (M + X) was predicted.
9. The method of claim 8, wherein when predicting the total amount of regurgitation of the first subject on day (M + X +1), the method further comprises:
to miThe retention amount of the first object on the ith day is updated, and the retention amount of the first object after the update is m' ═ mi·(1-roi)；
Predicting a total amount of reflow for the first object by:
when i is (M + X +1), the total amount of backflow of the first subject on the (M + X +1) th day is predicted.
10. An apparatus for predicting a number of daily active objects, comprising:
a first determining module, configured to predict a total backflow amount of a first object on an (M + X) th day through an object backflow model, where the object backflow model is obtained by modeling after counting backflow rates of the first object in daily active objects on consecutive (N +1) th days from the (M) th day to the (M-N) th day are analyzed, the first object is an object newly added before the (M-N) th day, X is a positive integer greater than or equal to 1, N is a positive integer greater than or equal to 1, and M is a positive integer greater than or equal to N;
a second determining module, configured to predict a total retention amount of a second object on the (M + X) th day through an object retention model, where the object retention model is a model obtained by performing data analysis according to a retention amount of a second object in the number of daily active objects in the (N +1) th day, and the second object is an object newly added after the (M-N) th day;
a third determination module, for determining the number of active objects per day (M + X) according to the total reflow amount of the first object on the (M + X) th day and the total retention amount of the second object on the (M + X) th day determined by the second determination module.
11. The apparatus of claim 10, further comprising: the device comprises an acquisition module, a calculation module, a first model building module, a statistic module and a second model building module;
the acquiring module is used for acquiring the number of daily active objects from the M th day to the (M-N) th day for continuous (N +1) days, wherein N is a positive integer greater than or equal to 60, and the number of daily active objects per day comprises the number of second objects and the remaining amount of first objects except the newly added objects in the daily active objects;
the calculation module is used for calculating the retention rate of a second object according to the retention amount of the second object in the (N +1) -day active objects acquired by the acquisition module;
the first model establishing module is used for carrying out data analysis on the retention rate data of the second object calculated by the calculating module to obtain an object retention model;
the statistic module is used for counting the reflux quantity of a first object in daily active objects in the (N +1) days;
and the second model establishing module is used for carrying out data analysis on the backflow amount of the first object counted by the counting module to obtain an object backflow model.
12. An apparatus for predicting a number of daily active objects, comprising:
a memory for storing computer executable program code;
an input/output interface, and
a processor coupled with the memory and the input/output interface;
wherein the program code comprises instructions which, when executed by the processor, cause the determining means to carry out the method of any one of claims 1 to 9.
13. A computer-readable storage medium comprising instructions which, when executed on a computer, cause the computer to perform the method of any of claims 1 to 9.
14. A computer program product comprising instructions which, when run on a computer, cause the computer to perform the method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710324322.5A CN108881333A (en) | 2017-05-09 | 2017-05-09 | A kind of method and apparatus for predicting to enliven number of objects day |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710324322.5A CN108881333A (en) | 2017-05-09 | 2017-05-09 | A kind of method and apparatus for predicting to enliven number of objects day |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108881333A true CN108881333A (en) | 2018-11-23 |
Family
ID=64287591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710324322.5A Pending CN108881333A (en) | 2017-05-09 | 2017-05-09 | A kind of method and apparatus for predicting to enliven number of objects day |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108881333A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109684594A (en) * | 2018-11-26 | 2019-04-26 | 长安通信科技有限责任公司 | Enliven the metering method and device, readable storage medium storing program for executing of object |
| CN109711897A (en) * | 2018-12-29 | 2019-05-03 | 贵州创鑫旅程网络技术有限公司 | Day any active ues quantity prediction technique and device |
| CN109858694A (en) * | 2019-01-28 | 2019-06-07 | 上海连尚网络科技有限公司 | A kind of method and apparatus of day active users prediction |
| CN112116381A (en) * | 2020-08-31 | 2020-12-22 | 北京基调网络股份有限公司 | Moon life prediction method based on LSTM neural network, storage medium and computer equipment |
| CN114385700A (en) * | 2022-01-10 | 2022-04-22 | 腾讯科技（深圳）有限公司 | Method, device and equipment for determining seed object and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914743A (en) * | 2014-04-21 | 2014-07-09 | 中国科学技术大学先进技术研究院 | On-line serial content popularity prediction method based on autoregressive model |
| CN105208534A (en) * | 2014-06-25 | 2015-12-30 | 腾讯科技（深圳）有限公司 | Event notification method and system |
| CN105869022A (en) * | 2016-04-07 | 2016-08-17 | 腾讯科技（深圳）有限公司 | Application popularity prediction method and apparatus |
| CN106056400A (en) * | 2016-05-19 | 2016-10-26 | 政和科技股份有限公司 | Method, apparatus and system of predicting number of new users |
| US20160357708A1 (en) * | 2015-06-05 | 2016-12-08 | Panasonic Intellectual Property Corporation Of America | Data analysis method, data analysis apparatus, and recording medium having recorded program |
| CN106251017A (en) * | 2016-08-02 | 2016-12-21 | 北京百度网讯科技有限公司 | Data predication method and device |
| CN106600344A (en) * | 2016-12-30 | 2017-04-26 | 广州虎牙信息科技有限公司 | Method and apparatus for obtaining active user data of target product |
-
2017
- 2017-05-09 CN CN201710324322.5A patent/CN108881333A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103914743A (en) * | 2014-04-21 | 2014-07-09 | 中国科学技术大学先进技术研究院 | On-line serial content popularity prediction method based on autoregressive model |
| CN105208534A (en) * | 2014-06-25 | 2015-12-30 | 腾讯科技（深圳）有限公司 | Event notification method and system |
| US20160357708A1 (en) * | 2015-06-05 | 2016-12-08 | Panasonic Intellectual Property Corporation Of America | Data analysis method, data analysis apparatus, and recording medium having recorded program |
| CN105869022A (en) * | 2016-04-07 | 2016-08-17 | 腾讯科技（深圳）有限公司 | Application popularity prediction method and apparatus |
| CN106056400A (en) * | 2016-05-19 | 2016-10-26 | 政和科技股份有限公司 | Method, apparatus and system of predicting number of new users |
| CN106251017A (en) * | 2016-08-02 | 2016-12-21 | 北京百度网讯科技有限公司 | Data predication method and device |
| CN106600344A (en) * | 2016-12-30 | 2017-04-26 | 广州虎牙信息科技有限公司 | Method and apparatus for obtaining active user data of target product |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109684594A (en) * | 2018-11-26 | 2019-04-26 | 长安通信科技有限责任公司 | Enliven the metering method and device, readable storage medium storing program for executing of object |
| CN109711897A (en) * | 2018-12-29 | 2019-05-03 | 贵州创鑫旅程网络技术有限公司 | Day any active ues quantity prediction technique and device |
| CN109858694A (en) * | 2019-01-28 | 2019-06-07 | 上海连尚网络科技有限公司 | A kind of method and apparatus of day active users prediction |
| CN112116381A (en) * | 2020-08-31 | 2020-12-22 | 北京基调网络股份有限公司 | Moon life prediction method based on LSTM neural network, storage medium and computer equipment |
| CN114385700A (en) * | 2022-01-10 | 2022-04-22 | 腾讯科技（深圳）有限公司 | Method, device and equipment for determining seed object and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108881333A (en) | A kind of method and apparatus for predicting to enliven number of objects day | |
| CN109697522B (en) | Data prediction method and device | |
| US9846574B2 (en) | Representing result data streams based on execution of data stream language programs | |
| CN109274987B (en) | Video set ordering method, server and readable storage medium | |
| US20200219016A1 (en) | Method for Cleaning Up Background Application, Storage Medium, and Electronic Device | |
| USRE47933E1 (en) | Reliability estimator for ad hoc applications | |
| US10496392B2 (en) | Staged application rollout | |
| WO2015158149A1 (en) | Method and device for forecasting changes of feature information | |
| US8606905B1 (en) | Automated determination of system scalability and scalability constraint factors | |
| CN104504084A (en) | Method and device for determining user retention rate | |
| Choi et al. | Inference for discretely observed stochastic kinetic networks with applications to epidemic modeling | |
| CN108021673A (en) | A kind of user interest model generation method, position recommend method and computing device | |
| US9292336B1 (en) | Systems and methods providing optimization data | |
| CN109272348B (en) | Method and device for determining number of active users, storage medium and electronic equipment | |
| CN107943677A (en) | Application performance monitoring method, device, readable storage medium storing program for executing and electronic equipment | |
| KR20210154831A (en) | Business Agnostic Consultation Center Estimated Wait Times Prediction Using Deep Neural Networks | |
| Jain et al. | Unreliable server M/G/1 queue with multi-optional services and multi-optional vacations | |
| US20170300819A1 (en) | Time-Series Prediction Apparatus and Time-Series Prediction Method | |
| JP2019508774A (en) | Database system optimization method, system, electronic device and storage medium | |
| CN105574032A (en) | Rule matching operation method and device | |
| CN110955438A (en) | Method, device and equipment for monitoring performance of small program and storage medium | |
| CN111311014B (en) | Service data processing method, device, computer equipment and storage medium | |
| CN112669091A (en) | Data processing method, device and storage medium | |
| CN110378543A (en) | Leaving office Risk Forecast Method, device, computer equipment and storage medium | |
| JP2020504375A (en) | Computer program, computer system, and computer implemented method for facilitating processing in a computing environment |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181123 |