CN107463660A - Product any active ues data measuring method and computer equipment - Google Patents
Product any active ues data measuring method and computer equipment Download PDFInfo
- Publication number
- CN107463660A CN107463660A CN201710640388.5A CN201710640388A CN107463660A CN 107463660 A CN107463660 A CN 107463660A CN 201710640388 A CN201710640388 A CN 201710640388A CN 107463660 A CN107463660 A CN 107463660A
- Authority
- CN
- China
- Prior art keywords
- data
- product
- amplification
- day
- active ues
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of product any active ues data measuring method.The product any active ues data measuring method includes obtaining the first training sample set;It is trained using first the first mathematical modeling of training sample set pair, obtain the first mathematical modeling characterizes day any active ues data and the parameter of ranking sequence relation;Obtain the second training sample set;It is trained using second the second mathematical modeling of training sample set pair, obtains the parameter for characterizing day any active ues data amplification, product keyword search exponent data amplification and website visiting data on flows amplification relation of the second mathematical modeling;Using target product day any active ues data amplification, target product keyword search exponent data amplification and website visiting data on flows amplification input the second mathematical modeling as parameter, calculate the day any active ues data of target product.The unknown day any active ues data of product can be deduced by the above method.Present invention also offers a kind of computer equipment.
Description
Technical field
The present invention relates to Internet technical field, and specifically, the present invention relates to a kind of measuring and calculating of product any active ues data
Method and computer equipment.
Background technology
With the development of internet, all kinds of application products based on internet gradually increase, the competition of like product
Intensity is also increasing.If the day any active ues data of like product can be obtained, the hair of competing product in overall market can be known clearly
Show shape, trend and rule, sharp can just hold the opportunity to develop of own product, the competing of own product is improved in keen competition
Degree of striving.
Belong to the internal confidential of company however, the day any active ues situation of like product belongs to it and constantly change, it is outside
Personnel can not obtain detailed day any active ues data situation.
The content of the invention
The purpose of the present invention aims to provide a kind of product any active ues data measuring method and computer equipment, realizes and obtains
The day any active ues data of product.
To achieve these goals, the present invention provides following technical scheme:
A kind of product any active ues data measuring method, including:
Obtain the first training sample set;First training sample set includes the day any active ues data and ranking sequence of preset product
Row;Establish on day any active ues data and the first mathematical modeling of ranking serial correlation, use the first training sample set pair
First mathematical modeling is trained, and obtain the first mathematical modeling characterizes day any active ues data and the ginseng of ranking sequence relation
Number;
Obtain the second training sample set;Second training sample set include preset product day any active ues data amplification, production
Product keyword search exponent data amplification and website visiting data on flows amplification;Establish on day any active ues data amplification, production
Second mathematical modeling of product keyword search exponent data amplification and website visiting data on flows amplification correlation, use the second instruction
Practice sample set the second mathematical modeling is trained, obtain the second mathematical modeling sign day any active ues data amplification, product
The parameter of keyword search exponent data amplification and website visiting data on flows amplification relation;
Obtain ranking sequence, target product keyword search exponent data amplification and the website visiting flow number of target product
According to amplification;The ranking sequence of target product is inputted the first mathematical modeling as parameter, obtains the day any active ues of target product
Data, the day any active ues data amplification of target product is obtained according to this day any active ues data;Enlivening the day of target product
User data increment, target product keyword search exponent data amplification and website visiting data on flows amplification input as parameter
Second mathematical modeling, calculate the day any active ues data of target product.
Wherein, the ranking sequence is the ranking sequence in IOS application market ranking lists.
Wherein, the preset product day any active ues data and ranking sequence include:Day on the day of the preset product
Any active ues data, the ranking on preset the product same day and latter two days;
The ranking sequence of the target product includes the ranking on the target product same day and latter two days.
Wherein, the preset product and the target product day any active ues data amplification, product keyword search
Exponent data amplification and website visiting data on flows amplification be respectively day any active ues data moon amplification, product keyword search refers to
Number data moon amplification and website visiting data on flows moon amplification.
Wherein, first mathematical modeling is multilayer perceptron model.
Wherein, described the step of being trained using first the first mathematical modeling of training sample set pair, including:Will be described pre-
The ranking sequence of product is put as independent variable, day any active ues data multilayer perceptron model is trained as dependent variable.
Wherein, the product keyword search exponent data amplification of the preset product and the target product includes:Production
The integral loop of Baidu's exponent data of product keyword than amplification, Baidu's exponent data of product keyword shift(ing) ring than amplification,
Or the PC rings of Baidu's index of product keyword compare amplification.
Wherein, the website visiting data on flows amplification of the preset product and the target product includes:The Alexa of product
The ring of website visiting traffic ranking accesses the ring of the number at the station than amplification, every million netizen in Alexa websites than amplification, Alexa
The ring of every million interviewed webpage of the website station webpage number compares amplification than amplification, or the ring of Alexa websites IP quantity.
Wherein, second mathematical modeling is generalized linear model.
A kind of computer equipment, including memory, processor and storage can be run on a memory and on a processor
Computer program, the product any active ues data measuring and calculating side of any of the above-described is realized during the computing device described program
Method.
Compared with prior art, the solution of the present invention has advantages below:
Product any active ues data measuring method provided by the invention, according to preset product day any active ues data and ranking
Sequence, which obtains, characterizes day any active ues data and the parameter of ranking sequence relation in the first mathematical modeling, further according to preset product
Day any active ues data amplification, product keyword search exponent data amplification and website visiting data on flows amplification obtain the second number
Learn in model, characterize day any active ues data amplification, product keyword search exponent data amplification and website visiting data on flows
The parameter of amplification relation.It is contemplated that the parameter of the ranking sequence and day any active ues data correlation that characterize product is obtained,
And obtain and characterize day any active ues data amplification, product keyword search exponent data amplification and the increasing of website visiting data on flows
The parameter of width.When by the ranking sequence inputting of target product, into corresponding relevant parameter, the day that can obtain target product enlivens
User data.Again by target product day any active ues data amplification, target product keyword search exponent data amplification and net
Flowing of access data amplification of standing is input in corresponding relevant parameter, you can calculates the day any active ues data of target product.
Brief description of the drawings
Of the invention above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments
Substantially and it is readily appreciated that, wherein:
Fig. 1 is the method flow diagram of the product any active ues data measuring method of one embodiment of the invention;
Fig. 2 is the IOS application products DAU and the form of the correlation of its ranking sequence of one embodiment of the invention
Figure;
Fig. 3 is the structure that DAU is speculated according to preset product ranking sequence and internal DAU of one embodiment of the invention
Figure;
Fig. 4 is the product DAU amplification and Baidu index, the table of Alexa website traffic correlations of one embodiment of the invention
Trrellis diagram;
Fig. 5 is the final measuring and calculating product DAU of one embodiment of the invention supposition structure chart;
Fig. 6 is the schematic diagram of the terminal part structure of one embodiment of the invention.
Embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, is only used for explaining the present invention, and is not construed as limiting the claims.
Any active ues scale:In timing statisticses, the number of users of application product has been used.
DAU:Day any active ues, abbreviation day are lived, and the user of application product has been used in a natural day.
Baidu's index:Baidu search temperature is also referred to as in this programme, based on Baidu's magnanimity netizen's behavioral data, performance
Search degree of concern and lasting situation of change of the netizen to keyword, and then reflect in certain degree the development of things
Trend.
Alexa web publishing data:Website traffic data, including website ranking, visitor's quantity, page are also referred to as in this programme
The flow indicators such as face pageview, the flow and pouplarity of website can be reflected in certain degree.
Ranking sequence:The ranking index set of the free list ranking compositions of IOS in one section of date is refered in particular in this programme.
The present invention provides a kind of product any active ues data measuring method.In this programme, product represents application program production
Product.Specifically include the live APP (Application, application program) of game.As shown in figure 1, a kind of product of the present invention is actively used
User data measuring method, including step:
S10, obtain the first training sample set;First training sample set includes the day any active ues data and row of preset product
Name sequence;Establish on day any active ues data and the first mathematical modeling of ranking serial correlation, use the first training sample
The mathematical modeling of set pair first is trained, and obtain the first mathematical modeling characterizes day any active ues data and ranking sequence relation
Parameter.
First training sample is concentrated, and preset product is application product known to day any active ues data.In the present invention
In, preset product is own product.The ranking sequence of own product is the ranking sequence in IOS application market ranking lists.It is own
The ranking sequence of product can obtain in third party's monitoring data platform.Third party's monitoring data platform include App Annie,
SensorTower etc..
In this programme, the day any active ues data of preset product include the day any active ues data on the same day.Preset product
Ranking sequence include the same day and the ranking of latter two days.
Specifically, the method for obtaining the ranking sequence of preset product is:
By third party's monitor supervision platform (such as App Annie, SensorTower) of IOS application markets, preset production is gathered
The ranking of the daily free list of product.Therefore, the tuple of ranking two (date, rank) of preset product can be obtained.Wherein, date is to work as
Its date, rank are same day ranking.Because same day DAU (Daily Active User, day any active ues quantity) may be by nearly one
All rankings are influenceed, and the tuple of ranking two is expanded into ranking sequence (date, rank [order]).Wherein order takes oneself less than 7
So number, 0 represents the same day, and 1 represents one day after, by that analogy.
In the present embodiment, the first mathematical modeling on day any active ues data and ranking serial correlation is established.Its
In, the first mathematical modeling can be the multiform expression for characterizing day any active ues data and ranking serial correlation.Herein no longer
Repeat.It is trained by using first the first mathematical modeling of training sample set pair, can obtain the sign day of the first mathematical modeling
The parameter of any active ues data and ranking sequence relation.
Specifically, according to the ranking sequence obtained from third party's monitoring data platform, the day being had by oneself with reference to preset product lives
Jump user data, analysis obtain preset product day any active ues data with the correlation of its ranking sequence.According to both phases
Closing property establishes the first mathematical modeling, and is trained using first the first mathematical modeling of training sample set pair, and then obtains first
Day any active ues data and the parameter of ranking sequence relation are characterized in mathematical modeling.In this programme, using the day of preset product
Any active ues data and the ranking on the preset product same day and latter two days are trained to the first mathematical modeling.Specifically, first
Mathematical modeling is multilayer perceptron model.Using the ranking sequence of preset product as independent variable, day any active ues data as because
Variable is trained to multilayer perceptron model, and day any active ues data and ranking are characterized in multilayer perceptron model so as to obtain
The parameter of sequence relation.
Specifically, the ranking sequence of preset product (it is live live with YY that this programme includes protruding canine teeth) 3 months, analysis are gathered
The inside DAU of preset product and the correlation of ranking sequence, find the DAU on the same day and the ranking sequence on the same day and latter two days
Negative correlation it is larger.As shown in Figure 2.Wherein both coefficient correlations are located at [- 1,1] section, are negative correlation close to -1, approach
1 is positive correlation, is uncorrelated close to 0.Therefore, can be by the sets of relational data in the DAU of preset product and ranking sequence
(date, DAU, rank [0], rank [1], rank [2]), as training sample set, on day any active ues data and ranking
The multilayer perceptron model of serial correlation is trained, and day any active ues data are characterized in multilayer perceptron model so as to obtain
With the parameter of ranking serial correlation.Specifically, make the sets of relational data of ranking sequence in sample set as independent variable, DAU
For dependent variable, multilayer perceptron model is created, and multilayer perceptron model is trained using ranking sequence and DAU.Such as Fig. 3
It is shown.
Wherein, the first mathematical modeling (multilayer perceptron model is referred in this programme) can be existing mathematical modeling.This
Scheme main purpose be obtain product day any active ues data and ranking sequence correlation, and obtained according to the correlation existing
Day any active ues data and the parameter of ranking sequence relation are characterized in the first mathematical modeling having.By the ginseng for characterizing both sides relation
Number, the day any active ues data according to corresponding to can obtaining ranking sequence.
S20, obtain the second training sample set;The day any active ues data that second training sample set includes preset product increase
Width, product keyword search exponent data amplification and website visiting data on flows amplification;Establish and increase on day any active ues data
Second mathematical modeling of width, product keyword search exponent data amplification and website visiting data on flows amplification correlation, use
Second the second mathematical modeling of training sample set pair is trained, and the sign day any active ues data for obtaining the second mathematical modeling increase
The parameter of width, product keyword search exponent data amplification and website visiting data on flows amplification relation.
In the present embodiment, establish on day any active ues data amplification, product keyword search exponent data amplification and
Second mathematical modeling of website visiting data on flows amplification correlation.Second mathematical modeling can be to characterize day any active ues data to increase
The multiform expression of width, product keyword search exponent data amplification and website visiting data on flows amplification correlation.Herein
Repeat no more.It is trained by using second the second mathematical modeling of training sample set pair, can obtain the table of the second mathematical modeling
Levy day any active ues data amplification, product keyword search exponent data amplification and website visiting data on flows amplification correlation
Parameter.
Specifically, the day any active ues data amplification for the preset product that the second training sample is concentrated is day any active ues data
Month amplification.The product keyword search exponent data amplification of preset product is product keyword search exponent data moon amplification.In advance
The website visiting data on flows amplification for putting product is website visiting data on flows moon amplification.Certainly, in other embodiments,
Preset product day any active ues data amplification, product keyword search exponent data amplification and website visiting data on flows amplification
Also other representations such as day amplification or all amplification can be used.Second mathematical modeling is generalized linear model.
In this programme, the day any active ues data amplification of preset product can be enlivened day according to preset product a couple of days is gathered
User data is to obtain its amplification data.The product keyword search exponent data of preset product refers to using the Baidu of product
Number.The product keyword search exponent data amplification of preset product includes the integral loop ratio of Baidu's exponent data of product keyword
Amplification, product keyword Baidu's exponent data shift(ing) ring than amplification, or Baidu's index of product keyword PC rings than increasing
Width.Website visiting data on flows is using Alexa website visiting datas on flows.The website visiting data on flows of preset product increases
Width includes:The ring of the Alexa website visiting traffic rankings of product accesses the people at the station than amplification, every million netizen in Alexa websites
Several rings than amplification, every million interviewed webpage of the Alexa websites station webpage number ring than amplification, or Alexa websites IP quantity
Ring compares amplification.
In this programme, in order to improve the DAU calculated of product accuracy, Baidu's index of product is further gathered
With Alexa web publishing data.(refer to being applied to be trained the first mathematical modeling in this case with reference to the DAU of preliminary survey
Day any active ues data), screen and create index system, create across data source, various dimensions DAU presumption models, changed
The DAU of final products after entering.
Specifically, Baidu's index and Alexa website traffics reflect all from different aspect and to a certain extent monitoring
APP (such as live platform) DAU change, analyse in depth preliminary survey DAU, Baidu's index, the multidimensional index numerical value of Alexa website traffics
Distribution and change contact between index amplification on a month-on-month basis, it is found that the DAU amplification for monitoring APP refers to Baidu as a whole
There is change uniformity in several amplification, Alexa website traffics amplification, as shown in Figure 4.
As shown in figure 4, the IOS first lunar months (preliminary survey) DAU and Baidu search temperature and website traffic performance are not consistent, so
And the equal DAU rings of the IOS first lunar months refer to (Baidu's index) than amplification with hundred, the ring of Alexa website visiting number of pages IP amounts shows one than amplification
Cause.Hundred refer to overall (PC) distinguishes negatively correlated and positive correlation with Alexa rankings and website traffic, and it is high to indicate search temperature
Website, ranking is more forward, and flow is also higher.
Therefore, using preset product monthly day any active ues ring than amplification data, the moon of Baidu's index of keyword
Average daily ring than amplification data and Alexa website traffics the moon average daily ring than amplification data correlation, to the second mathematical modeling
(that is to say generalized linear model) is trained, and obtains and day any active ues data amplification is characterized in generalized linear model, product closes
The parameter of keyword searchable index data amplification and website visiting data on flows amplification relation.Referring specifically to shown in accompanying drawing 5.
Wherein, the second mathematical modeling can be existing mathematical modeling.This programme be intended to obtain day any active ues data amplification,
The correlation of product keyword search exponent data amplification and website visiting data on flows amplification, and according to the correlation to second
Mathematical modeling is trained, with obtain the second mathematical modeling sign day any active ues data amplification, product keyword search refers to
The parameter of number data amplification and website visiting data on flows amplification relation.By characterizing the parameter of triadic relation, according to step S10
The day any active ues data of the preliminary survey of acquisition, final day any active ues data can be obtained.
S30, obtain the ranking sequence, target product keyword search exponent data amplification and website visiting stream of target product
Measure data amplification;The ranking sequence of target product is inputted the first mathematical modeling as parameter, enlivened the day for obtaining target product
User data, the day any active ues data amplification of target product is obtained according to this day any active ues data;The day of target product
Any active ues data amplification, target product keyword search exponent data amplification and website visiting data on flows amplification are as parameter
The second mathematical modeling is inputted, calculates the day any active ues data of target product.
In this programme, the ranking sequence of target product includes the same day and the ranking of latter two days.It is specific to obtain target production
The method of the ranking sequence of product is:By third party's monitor supervision platform (such as App Annie, SensorTower of IOS application markets
Deng), gather the daily free list ranking of target product.Therefore, the tuple of ranking two (date, rank) of target product can be obtained.
Wherein, date is the date on the same day, and rank is same day ranking.Due to same day DAU (Daily Active User, day active users
Amount) it may be influenceed by nearly one week ranking, the tuple of ranking two is expanded into ranking sequence (date, rank [order]).Wherein
Order takes the natural number less than 7, and 0 represents the same day, and 1 represents one day after, by that analogy.
In this programme, the day any active ues data amplification of target product can be enlivened day according to collection target product a couple of days
User data is to obtain its amplification data.The product keyword search exponent data of target product refers to using the Baidu of product
Number.The product keyword search exponent data amplification of target product includes the integral loop ratio of Baidu's exponent data of product keyword
Amplification, product keyword Baidu's exponent data shift(ing) ring than amplification, or Baidu's index of product keyword PC rings than increasing
Width.Website visiting data on flows is using Alexa website visiting datas on flows.The website visiting data on flows of target product increases
Width includes:The ring of the Alexa website visiting traffic rankings of product accesses the people at the station than amplification, every million netizen in Alexa websites
Several rings than amplification, every million interviewed webpage of the Alexa websites station webpage number ring than amplification, or Alexa websites IP quantity
Ring compares amplification.
In this programme, the day any active ues data amplification for the target product that the second training sample is concentrated is day any active ues
The data moon amplification.The product keyword search exponent data amplification of target product increases for the product keyword search exponent data moon
Width.The website visiting data on flows amplification of target product is website visiting data on flows moon amplification.Certainly, in other embodiment party
In case, target product day any active ues data amplification, product keyword search exponent data amplification and website visiting flow number
Other representations such as day amplification or all amplification can be also used according to amplification.
In this programme, by preset product day any active ues data and ranking sequence pair multilayer perceptron model instructed
Practice, day any active ues data and the parameter of ranking sequence relation are characterized in the first mathematical modeling to obtain.Therefore, by target product
Ranking sequence inputting into multilayer perceptron, to obtain the day any active ues data of target product.By gathering target product
The day any active ues data of a couple of days, the day any active ues data amplification of target product can be obtained.Specifically, this programme collection is
Day any active ues data in target product one month, obtain the monthly day any active ues data amplification of target product.
The target product of acquisition day any active ues data amplification, target product keyword search exponent data amplification and
Website visiting data on flows amplification inputs the second mathematical modeling as parameter, you can calculates the day active users of target product
According to.
Product any active ues data measuring method provided by the invention, according to preset product day any active ues data and ranking
Sequence, which obtains, characterizes day any active ues data and the parameter of ranking sequence relation in the first mathematical modeling, further according to preset product
Day any active ues data amplification, product keyword search exponent data amplification and website visiting data on flows amplification obtain the second number
Learn in model, characterize day any active ues data amplification, product keyword search exponent data amplification and website visiting data on flows
The parameter of amplification relation.It is contemplated that the parameter of the ranking sequence and day any active ues data correlation that characterize product is obtained,
And obtain and characterize day any active ues data amplification, product keyword search exponent data amplification and the increasing of website visiting data on flows
The parameter of width.When by the ranking sequence inputting of target product, into corresponding relevant parameter, the day that can obtain target product enlivens
User data.Again by target product day any active ues data amplification, target product keyword search exponent data amplification and net
Flowing of access data amplification of standing is input in corresponding relevant parameter, you can calculates the day any active ues data of target product.
The present invention also provides a kind of computer equipment.A kind of computer equipment includes memory, processor and is stored in
On memory and the computer program that can run on a processor.Realize that any one of such scheme produces during computing device program
Product any active ues data measuring method.
Referring to Fig. 6.Fig. 6 is the block diagram of the relevant portion structure of computer equipment provided in an embodiment of the present invention.This implementation
In example, computer equipment is specially mobile phone.The mobile phone includes:Radio frequency (Radio Frequency, RF) circuit 610, memory
620th, input block 630, display unit 640, sensor 650, voicefrequency circuit 660, Wireless Fidelity (wireless fidelity,
WiFi) the part such as module 670 (namely WiFi chip module), processor 680 and power supply 690.The work(of the associated components of mobile phone
Energy and interaction relationship will not be described here.
It will be understood by those skilled in the art that the handset structure shown in Fig. 6 does not form the restriction to mobile phone, can wrap
Include than illustrating more or less parts, either combine some parts or different parts arrangement.
Described above is only some embodiments 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 (10)
- A kind of 1. product any active ues data measuring method, it is characterised in that including:Obtain the first training sample set;First training sample set includes the day any active ues data and ranking sequence of preset product; Establish on day any active ues data and the first mathematical modeling of ranking serial correlation, use the first training sample set pair first Mathematical modeling is trained, and obtain the first mathematical modeling characterizes day any active ues data and the parameter of ranking sequence relation;Obtain the second training sample set;Second training sample set include preset product day any active ues data amplification, product close Keyword searchable index data amplification and website visiting data on flows amplification;Establish and closed on day any active ues data amplification, product Second mathematical modeling of keyword searchable index data amplification and website visiting data on flows amplification correlation, use the second training sample The mathematical modeling of this set pair second is trained, obtain the second mathematical modeling sign day any active ues data amplification, product it is crucial The parameter of word searchable index data amplification and website visiting data on flows amplification relation;The ranking sequence, target product keyword search exponent data amplification and website visiting data on flows for obtaining target product increase Width;The ranking sequence of target product is inputted the first mathematical modeling as parameter, obtains the day any active ues data of target product, The day any active ues data amplification of target product is obtained according to this day any active ues data;The day active users of target product According to amplification, target product keyword search exponent data amplification and website visiting data on flows amplification as the number of parameter input second Model is learned, calculates the day any active ues data of target product.
- 2. product any active ues data measuring method according to claim 1, it is characterised in that the ranking sequence is Ranking sequence in IOS application market ranking lists.
- 3. product any active ues data measuring method according to claim 2, it is characterised in thatThe preset product day any active ues data and ranking sequence include:Day active users on the day of the preset product According to the ranking on preset the product same day and latter two days;The ranking sequence of the target product includes the ranking on the target product same day and latter two days.
- 4. product any active ues data measuring method according to claim 1, it is characterised in that the preset product and The target product day any active ues data amplification, product keyword search exponent data amplification and website visiting data on flows Amplification be respectively day any active ues data moon amplification, product keyword search exponent data moon amplification and website visiting data on flows Month amplification.
- 5. product any active ues data measuring method according to claim 1, it is characterised in that first mathematical modeling For multilayer perceptron model.
- 6. product any active ues data measuring method according to claim 5, it is characterised in that described to use the first training The step of sample set is trained to the first mathematical modeling, including:Using the ranking sequence of the preset product as independent variable, day Any active ues data are trained as dependent variable to multilayer perceptron model.
- 7. product any active ues data measuring method according to claim 1, it is characterised in that the preset product and The product keyword search exponent data amplification of the target product includes:The integral loop of Baidu's exponent data of product keyword Than amplification, Baidu's exponent data of product keyword shift(ing) ring than amplification, or Baidu's index of product keyword PC rings ratio Amplification.
- 8. product any active ues data measuring method according to claim 1, it is characterised in that the preset product and institute Stating the website visiting data on flows amplification of target product includes:The ring of the Alexa website visiting traffic rankings of product than amplification, Every million netizen in Alexa websites accesses the ring of the number at the station than amplification, every million interviewed webpage of the Alexa websites station webpage number Ring compare amplification than amplification, or the ring of Alexa websites IP quantity.
- 9. product any active ues data measuring method according to claim 1, it is characterised in that second mathematical modeling For generalized linear model.
- 10. a kind of computer equipment, including memory, processor and storage can be run on a memory and on a processor Computer program, it is characterised in that realize that any one of claim 1-9 product is lived during the computing device described program Jump user data measuring method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710640388.5A CN107463660B (en) | 2017-07-31 | 2017-07-31 | Method for measuring and calculating data of active users of products and computer equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710640388.5A CN107463660B (en) | 2017-07-31 | 2017-07-31 | Method for measuring and calculating data of active users of products and computer equipment |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107463660A true CN107463660A (en) | 2017-12-12 |
CN107463660B CN107463660B (en) | 2020-10-16 |
Family
ID=60547981
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710640388.5A Active CN107463660B (en) | 2017-07-31 | 2017-07-31 | Method for measuring and calculating data of active users of products and computer equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107463660B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108322783A (en) * | 2018-01-25 | 2018-07-24 | 广州虎牙信息科技有限公司 | Video website userbase estimation method, storage medium and terminal |
CN109711652A (en) * | 2017-10-26 | 2019-05-03 | 厦门一品威客网络科技股份有限公司 | A kind of Chuan Ke team potential methods of marking |
CN110458360A (en) * | 2019-08-13 | 2019-11-15 | 腾讯科技(深圳)有限公司 | Prediction technique, device, equipment and the storage medium of hot resource |
CN111475718A (en) * | 2020-03-30 | 2020-07-31 | 清华大学 | User activity model construction method and system based on preference diversity |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101266610A (en) * | 2008-04-25 | 2008-09-17 | 浙江大学 | Web active user website accessing mode on-line excavation method |
US20150186535A1 (en) * | 2013-12-27 | 2015-07-02 | Quixey, Inc. | Determining an Active Persona of a User Device |
CN105787039A (en) * | 2016-02-25 | 2016-07-20 | 四川长虹电器股份有限公司 | Method for multi-period calculation and display of user retention rate |
CN105869022A (en) * | 2016-04-07 | 2016-08-17 | 腾讯科技(深圳)有限公司 | Application popularity prediction method and apparatus |
CN106126641A (en) * | 2016-06-24 | 2016-11-16 | 中国科学技术大学 | A kind of real-time recommendation system and method based on Spark |
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
- 2017-07-31 CN CN201710640388.5A patent/CN107463660B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101266610A (en) * | 2008-04-25 | 2008-09-17 | 浙江大学 | Web active user website accessing mode on-line excavation method |
US20150186535A1 (en) * | 2013-12-27 | 2015-07-02 | Quixey, Inc. | Determining an Active Persona of a User Device |
CN105787039A (en) * | 2016-02-25 | 2016-07-20 | 四川长虹电器股份有限公司 | Method for multi-period calculation and display of user retention rate |
CN105869022A (en) * | 2016-04-07 | 2016-08-17 | 腾讯科技(深圳)有限公司 | Application popularity prediction method and apparatus |
CN106126641A (en) * | 2016-06-24 | 2016-11-16 | 中国科学技术大学 | A kind of real-time recommendation system and method based on Spark |
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 |
Non-Patent Citations (1)
Title |
---|
陈江宁 等: "基于时间线数据的微博活跃用户数估计", 《计算机应用与软件》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109711652A (en) * | 2017-10-26 | 2019-05-03 | 厦门一品威客网络科技股份有限公司 | A kind of Chuan Ke team potential methods of marking |
CN108322783A (en) * | 2018-01-25 | 2018-07-24 | 广州虎牙信息科技有限公司 | Video website userbase estimation method, storage medium and terminal |
CN108322783B (en) * | 2018-01-25 | 2021-08-17 | 广州虎牙信息科技有限公司 | Video website user scale presumption method, storage medium and terminal |
CN110458360A (en) * | 2019-08-13 | 2019-11-15 | 腾讯科技(深圳)有限公司 | Prediction technique, device, equipment and the storage medium of hot resource |
CN110458360B (en) * | 2019-08-13 | 2023-07-18 | 腾讯科技(深圳)有限公司 | Method, device, equipment and storage medium for predicting hot resources |
CN111475718A (en) * | 2020-03-30 | 2020-07-31 | 清华大学 | User activity model construction method and system based on preference diversity |
Also Published As
Publication number | Publication date |
---|---|
CN107463660B (en) | 2020-10-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107463660A (en) | Product any active ues data measuring method and computer equipment | |
CN103106259B (en) | A kind of mobile webpage content recommendation method based on situation | |
WO2016107523A1 (en) | Access path analysis method and apparatus for website | |
Gleich et al. | Tracking the random surfer: empirically measured teleportation parameters in PageRank | |
CN103631957B (en) | Visitor's behavioral data statistical method and device | |
CN103353872B (en) | A kind of teaching resource personalized recommendation method based on neutral net | |
TWI567572B (en) | Data acquisition method, device and system | |
CN106445971A (en) | Application recommendation method and system | |
CA3062119A1 (en) | Method and device for setting sample weight, and electronic apparatus | |
CN103177086A (en) | Method and device for ranking order through using searching result correspondingly | |
Cortés et al. | The productivity of top researchers: a semi-nonparametric approach | |
CN104572757A (en) | Microblog group processing method and device | |
CN104281619A (en) | System and method for ordering search results | |
CN106156123A (en) | Active value computational methods and device | |
CN103530392B (en) | Determine the method and apparatus of crawl flow | |
CN106933885A (en) | The acquisition methods and device of website propagating influence | |
Sha et al. | Bayesian inference for Birnbaum–Saunders distribution and its generalization | |
CN110020105A (en) | The determination method and device of site search performance | |
Pramanik et al. | An empirical approach towards an efficient “whom to mention?” Twitter app | |
CN103425643A (en) | Recommending method and system for relevant search request strings | |
Wang et al. | Degree of user attention to a webpage based on Baidu Index: an alternative to page view | |
CN103995893A (en) | Quick link recommendation method, device and system in browser | |
Roshan et al. | An analysis of global warming effect on the temperature variations in the northwest of Iran | |
Yurtseven et al. | Estimation of interception with artificial neural networks in oak-beech mixed stand. | |
Peng Wei et al. | Individual tree diameter growth model for Larix olgensis plantation in Heilongjiang Province, China. |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |