CN104636954A - Data mining method and data mining device for advertising media putting quantity - Google Patents

Abstract

The invention discloses a data mining method for advertising media putting quantity. The data mining method includes: step 1), acquiring historical statistic data of advertising media putting, wherein the historical statistic data include clicking quantity ctr and putting quantity, on advertising media of an advertisement; step 2), acquiring a preset optimal object model; step 3), acquiring a preset limiting condition of the clicking quantity, on the media, of the advertisement and the optimal object model in the step 2), and creating a data mining model of the advertising media putting quantity; step 4), calculating the data mining model created in the step 3) to acquire optimal advertising media putting quantity data.

Description

The data digging method of a kind of advertising media injected volume and device

Technical field

The application is based on moving advertising industry, and the exposure of throwing in the media advertisement is main or manually by the prioritization scheme of experience setting.

Background technology

To advertisement injected volume in the media, prior art is mainly manually, by experience setting.Shortcoming is a lot: 1. pair advertisement clicking rate change is in the media insensitive,

2. the ctr memory capability of human brain to advertisement or media is limited,

3. setup parameter is to the business need of people, and memory requires all very high, and new hand's follow-up is comparatively slow,

4. the subjective understanding by people is larger with memory people deviation.

Summary of the invention

Technical matters to be solved by this invention is to provide a kind of optimization method, i.e. the data digging method of a kind of advertising media injected volume and device.

It is as follows that the present invention solves the problems of the technologies described above taked technical scheme:

A data digging method for advertising media's injected volume, comprising:

Step 1) obtain the historical statistical data that advertising media throws in, this historical statistical data comprises: advertisement clicking rate ctr in the media and injected volume;

Step 2) obtain the optimization aim model preset;

Step 3) obtain advertisement in the default qualifications of media click volume and step 2) in optimization aim model, set up the data mining model of advertising media's injected volume;

Step 4) to step 3) data mining model set up calculates, and obtains best advertising media's injected volume data.

Further, preferably, step 2) in, described in the optimization aim model that presets be profit margin, wherein, choose following profit margin model:

z＝ctr ₁₁*price ₁*x ₁₁+…ctr _ij*price _j*x _ij+…+ctr _nm*price _m*x _nm

Wherein, n is media number, and m is advertisement number, x _ijcorresponding i-th media, the injected volume of a jth advertisement, ctr _ijit is the historic click-through rate of i-th media, a jth advertisement.

Further, preferably, advertisement, at the default qualifications of media click volume, comprising:

A data mining device for advertising media's injected volume, comprising:

Historical data acquiring unit, for obtaining the historical statistical data that advertising media throws in, this historical statistical data comprises: advertisement clicking rate ctr in the media and injected volume;

Objective optimization unit, for obtaining the optimization aim model preset;

Obtain advertisement at the default qualifications of media click volume and optimization aim model, set up the data mining model of advertising media's injected volume;

Computing unit, calculates for the data mining model set up, and obtains best advertising media's injected volume data.

Further, preferably, described in the optimization aim model that presets be profit margin, wherein, choose following profit margin model:

z＝ctr ₁₁*price ₁*x ₁₁+…ctr _ij*price _j*x _ij+…+ctr _nm*price _m*x _nm

Wherein, n is media number, and m is advertisement number, x _ijcorresponding i-th media, the injected volume of a jth advertisement, ctr _ijit is the historic click-through rate of i-th media, a jth advertisement.

Further, preferably, advertisement, at the default qualifications of media click volume, comprising:

After this invention takes such scheme, can according to history advertisement ctr in the media, injected volume optimizes the distribution of up-to-date injected volume, can according to maximization corporate profit, maximize ctr, optimal anchor direction is determined in one or several the combination maximized in clicks etc., notifies the result calculated, instructs or revise and artificially throw in data.

Other features and advantages of the present invention will be set forth in the following description, and, partly become apparent from instructions, or understand by implementing the present invention.Object of the present invention and other advantages realize by structure specifically noted in write instructions, claims and accompanying drawing and obtain.

Accompanying drawing explanation

Below in conjunction with accompanying drawing, the present invention is described in detail, to make above-mentioned advantage of the present invention definitely.Wherein,

The schematic flow sheet of the data digging method of Tu1Shi advertising media of the present invention injected volume;

The structural representation of the data mining device of Tu2Shi advertising media of the present invention injected volume.

Embodiment

Describe embodiments of the present invention in detail below with reference to drawings and Examples, to the present invention, how application technology means solve technical matters whereby, and the implementation procedure reaching technique effect can fully understand and implement according to this.It should be noted that, only otherwise form conflict, each embodiment in the present invention and each feature in each embodiment can be combined with each other, and the technical scheme formed is all within protection scope of the present invention.

In addition, can perform in the computer system of such as one group of computer executable instructions in the step shown in the process flow diagram of accompanying drawing, and, although show logical order in flow charts, but in some cases, can be different from the step shown or described by order execution herein.

Embodiment one:

As shown in Figure 1, the data digging method of a kind of advertising media injected volume, comprising:

Step 1) obtain the historical statistical data that advertising media throws in, this historical statistical data comprises: advertisement clicking rate ctr in the media and injected volume;

Step 2) obtain the optimization aim model preset;

Step 3) obtain advertisement in the default qualifications of media click volume and step 2) in optimization aim model, set up the data mining model of advertising media's injected volume;

Step 4) to step 3) data mining model set up calculates, and obtains best advertising media's injected volume data.

Further, preferably, step 2) in, described in the optimization aim model that presets be profit margin, wherein, choose following profit margin model:

z＝ctr ₁₁*price ₁*x ₁₁+…ctr _ij*price _j*x _ij+…+ctr _nm*price _m*x _nm

Wherein, n is media number, and m is advertisement number, x _ijcorresponding i-th media, the injected volume of a jth advertisement, ctr _ijit is the historic click-through rate of i-th media, a jth advertisement.

Further, preferably, advertisement, at the default qualifications of media click volume, comprising:

Wherein, after this invention takes such scheme, can according to history advertisement ctr in the media, injected volume optimizes the distribution of up-to-date injected volume, according to maximization corporate profit, can maximize ctr, optimal anchor direction is determined in one or several the combination maximized in clicks etc., notify the result calculated, instruct or revise and artificially throw in data.

Embodiment two:

Two above embodiment one is described in detail in conjunction with the embodiments, specifically, the application mainly can according to these last few days, advertisement ctr in the media, injected volume optimizes the distribution of up-to-date injected volume, can according to maximization corporate profit, maximize ctr, optimal anchor direction is determined in one or several the combination maximized in clicks etc., notifies the result calculated, instructs or revise and artificially throw in data

Wherein, suppose that optimization aim optimizes total profit:

First profit margin model (i.e. optimization aim model) is set up:

z＝ctr ₁₁*price ₁*x ₁₁+…ctr _ij*price _j*x _ij+…+ctr _nm

*price _m*x _nm

(n is media number, and m is advertisement number,

X _ijcorresponding i-th media, the injected volume of a jth advertisement,

Ctr _ijit is the historic click-through rate of i-th media, a jth advertisement

z＝ctr ₁₁*price ₁*x ₁₁+…ctr _ij*price _j*x _ij+…+ctr _nm*price _m*x _nm

(n is media number, and m is advertisement number,

X _ijcorresponding i-th media, the injected volume of a jth advertisement,

Ctr _ijthe historic click-through rate of i-th media, a jth advertisement)

Then set up some restrictive conditions of profit, such as company's total exposure amount is certain, condition that the ad click amount of advertiser request oneself is certain etc., and then the value of each exposure (i.e. variable x) can not be less than 0, as follows:

Finally by simplicial method, this model is solved, obtain optimization exposure x.

Wherein, being explained as follows of this formula: Article 1 formula represents: the click volume of the ctr of the first advertisement on all media (clicking rate) * exposure x>=advertiser request

Article 2 to m article of formula roughly the same.

M+1 formula represents the history average exposure (impNum) of all click volume x on first media and these media of <=

M+2 is the same to m+n formula

The last item formula represents, all exposures with <=corporate history average exposure (totalImpNum).

After this invention takes above scheme, there is following beneficial effect:

1. very fast for new hand's left-hand seat ratio, without the need to more to historical data cognition, as long as know current demand

2. reaction that can be more instant for the change of historical data

3. the breadth and depth for data understanding is also better

4. can tackle complicated demand, and the result specific aim obtained is very strong

Wherein, simplicial method: a kind of method being the polynary inequality of a kind of iterative, specifically, according to described simplicial method, it mainly performs an action as follows:

The first step, suppose that newly-increased several variablees (coefficient is all 1) are basic variables, other is nonbasic variable

Second step, the value of setting nonbasic variable is 0, calculates the value of basic variable

3rd step, substitutes into optimization function, calculates optimized results (this step can not have)

4th step, calculates the variable xmaxNoBase of nonbasic variable Optimum Profit Rate

5th step, calculates and contributes minimum basic variable xminBase (negative not very) relative to xmaxNoBase

6th step, replace xminBase to become basic variable with xmaxNoBase, xminBase becomes nonbasic variable

7th step, coefficient normalizing

8th step, until calculate so terminate time rate of profit is all less than 0, otherwise repeat above a few step

Simplicial method example is as follows:

Z=5x ₁+ 2x ₂+ 3x ₃-x ₄+ x ₅reach maximum

Constraint condition: x ₁+ 2x ₂+ 2x ₃+ x ₄=8

3x ₁+4x ₂+x ₃+x ₅＝7

x ₁，x ₂，x ₃，x ₄，x ₅≥0

With matrix representation be:

$A=\left(\begin{array}{ccccc}1& 2& 2& 1& 0\\ 3& 4& 1& 0& 1\end{array}\right)$ Constraint condition coefficient

$b=\left(\begin{array}{c}8\\ 7\end{array}\right)$ Constraint condition constant

$C_1=\left(\begin{array}{ccccc}5& 2& 3& -1& 1\end{array}\right)$ Objective function coefhcient

$C_B=\left(\begin{array}{cc}-1,& 1\end{array}\right)$ Basic variable objective function coefhcient

Tabulation, puts form into above-mentioned matrix of coefficients

X ₄=8, x ₅=7; And nonbasic variable x ₁=x ₂=x ₃=0, target function value is by C _bthe inner product vectorial with constant term two is tried to achieve, that is:

$Z=C_{B}b=\left(\begin{array}{cc}-1,& 1\end{array}\right)\left(\begin{array}{c}8\\ 7\end{array}\right)=-8+7=-1$

In order to check above-mentioned basic feasible solution whether to reach optimum, the relative profit of all nonbasic variables must be calculated.Its calculating has been come by matrix operation.As variable x ₁relative profit coefficient use represent, then:

p ₁for corresponding x ₁column vector

Example: $\begin{array}{c}{\stackrel{\&OverBar;}{C}}_1=C_1-(C_4,C_5)P_1\\ =5-(-\mathrm{1,1})\left(\begin{array}{c}1\\ 3\end{array}\right)\\ =5-(-1+3)=3\end{array}$

$\begin{array}{c}{\stackrel{\&OverBar;}{C}}_2=C_2-(C_4,C_5)P_2\\ =2-(-\mathrm{1,1})\left(\begin{array}{c}2\\ 4\end{array}\right)\\ =2-(-2+4)=0\end{array}$

$\begin{array}{c}{\stackrel{\&OverBar;}{C}}_3=C_3-(C_4,C_5)P_3\\ =3-(-\mathrm{1,1})\left(\begin{array}{c}2\\ 1\end{array}\right)\\ =3-(-2+1)=4\end{array}$

Ask a profit coefficient again, obtain maximum profit coefficient, calculate the fundamental element of the minimum contribution degree of relative maximum profit figure parameters.

Until terminate time the capable rate of profit of C is all less than or equal to 0, Z=81/5 is exactly optimum solution

embodiment three:

A data mining device for advertising media's injected volume, comprising:

Historical data acquiring unit, for obtaining the historical statistical data that advertising media throws in, this historical statistical data comprises: advertisement clicking rate ctr in the media and injected volume;

Objective optimization unit, for obtaining the optimization aim model preset;

Obtain advertisement at the default qualifications of media click volume and optimization aim model, set up the data mining model of advertising media's injected volume;

Computing unit, calculates for the data mining model set up, and obtains best advertising media's injected volume data.

Further, preferably, described in the optimization aim model that presets be profit margin, wherein, choose following profit margin model:

z＝ctr ₁₁*price ₁*x ₁₁+…ctr _ij*price _j*x _ij+…+ctr _nm*price _m*x _nm

Wherein, n is media number, and m is advertisement number, x _ijcorresponding i-th media, the injected volume of a jth advertisement, ctr _ijit is the historic click-through rate of i-th media, a jth advertisement.

Further, preferably, advertisement, at the default qualifications of media click volume, comprising:

Apparatus of the present invention have the identical technique effect of embodiment of the method, namely after this invention takes such scheme, can according to history advertisement ctr in the media, injected volume optimizes the distribution of up-to-date injected volume, according to maximization corporate profit, can maximize ctr, optimal anchor direction is determined in one or several the combination maximized in clicks etc., notify the result calculated, instruct or revise and artificially throw in data.

It should be noted that, for said method embodiment, in order to simple description, therefore it is all expressed as a series of combination of actions, but those skilled in the art should know, the application is not by the restriction of described sequence of movement, because according to the application, some step can adopt other orders or carry out simultaneously.Secondly, those skilled in the art also should know, the embodiment described in instructions all belongs to preferred embodiment, and involved action and module might not be that the application is necessary.

Those skilled in the art should understand, the embodiment of the application can be provided as method, system or computer program.Therefore, the application can adopt the form of complete hardware embodiment, completely software implementation or the embodiment in conjunction with software and hardware aspect.

And the application can adopt in one or more form wherein including the upper computer program implemented of computer-usable storage medium (including but not limited to magnetic disk memory, CD-ROM, optical memory etc.) of computer usable program code.

Last it is noted that the foregoing is only the preferred embodiments of the present invention, be not limited to the present invention, although with reference to previous embodiment to invention has been detailed description, for a person skilled in the art, it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein portion of techniques feature.Within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (6)

1. a data digging method for advertising media's injected volume, is characterized in that, comprising:

Step 1) obtain the historical statistical data that advertising media throws in, this historical statistical data comprises: advertisement clicking rate ctr in the media and injected volume;

Step 2) obtain the optimization aim model preset;

Step 3) obtain advertisement in the default qualifications of media click volume and step 2) in optimization aim model, set up the data mining model of advertising media's injected volume;

Step 4) to step 3) data mining model set up calculates, and obtains best advertising media's injected volume data.

2. the data digging method of advertising media according to claim 1 injected volume, is characterized in that, step 2) in, described in the optimization aim model that presets be profit margin, wherein, choose following profit margin model:

z＝ctr ₁₁*price ₁*x ₁₁+…ctr _ij*price _j*x _ij+…+ctr _nm*price _m*x _nm

Wherein, n is media number, and m is advertisement number, x _ijcorresponding i-th media, the injected volume of a jth advertisement, ctr _ijit is the historic click-through rate of i-th media, a jth advertisement.

3. the data digging method of advertising media according to claim 1 and 2 injected volume, is characterized in that, advertisement, at the default qualifications of media click volume, comprising:

4. a data mining device for advertising media's injected volume, is characterized in that, comprising:

Historical data acquiring unit, for obtaining the historical statistical data that advertising media throws in, this historical statistical data comprises: advertisement clicking rate ctr in the media and injected volume;

Objective optimization unit, for obtaining the optimization aim model preset;

Obtain advertisement at the default qualifications of media click volume and optimization aim model, set up the data mining model of advertising media's injected volume;

Computing unit, calculates for the data mining model set up, and obtains best advertising media's injected volume data.

5. the data mining device of advertising media according to claim 4 injected volume, is characterized in that, described in the optimization aim model that presets be profit margin, wherein, choose following profit margin model:

z＝ctr ₁₁*price ₁*x ₁₁+…ctr _ij*price _j*x _ij+…+ctr _nm*price _m*x _nm

Wherein, n is media number, and m is advertisement number, x _ijcorresponding i-th media, the injected volume of a jth advertisement, ctr _ijit is the historic click-through rate of i-th media, a jth advertisement.

6. the data mining device of the advertising media's injected volume according to claim 4 or 5, is characterized in that, advertisement, at the default qualifications of media click volume, comprising: