CN108564404A - A kind of method and device of prediction investment in advertising return rate - Google Patents
A kind of method and device of prediction investment in advertising return rate Download PDFInfo
- Publication number
- CN108564404A CN108564404A CN201810310740.3A CN201810310740A CN108564404A CN 108564404 A CN108564404 A CN 108564404A CN 201810310740 A CN201810310740 A CN 201810310740A CN 108564404 A CN108564404 A CN 108564404A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- stock
- data
- advertisement quantity
- target date
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0242—Determining effectiveness of advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0242—Determining effectiveness of advertisements
- G06Q30/0245—Surveys
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Human Resources & Organizations (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention proposes a kind of method of prediction investment in advertising return rate, including:Advertisement quantity in stock week mean value statistical data after obtaining the spending on ads data of the target date of target channel and launching advertisement from the target date in setting time section;According to the advertisement quantity in stock week mean value statistical data, the advertisement quantity in stock anticipation function after the target date launches advertisement is calculated;According to the advertisement quantity in stock anticipation function, determine that income is launched in the advertisement of the target date;The rate of return on investment that the advertisement of the target date is launched is calculated in the spending on ads data that income and the target date are launched according to the advertisement of the target date.Individual data fluctuation can be avoided to influence the fluctuation of short term statistics data using above-mentioned technical proposal, achieved the purpose that through short term statistics data Accurate Prediction investment in advertising return rate.
Description
Technical field
The present invention relates to rate of return on investment electric powder prediction more particularly to a kind of methods of prediction investment in advertising return rate
And device.
Background technology
In advertisement promotion, it would be desirable to the activation cost of each advertisement promotion channel is examined, it is final to determine which canal chosen
Road carries out advertisement dispensing.And activate the following producible advertisement value of user be us more concerned about the problem of, rate of return on investment
Prediction is exactly accurate by existing advertisement inventory variation tendency (rate of decay, inflection point, mean variance after dispensing etc.) training
Calculation formula, for the channel this time launch advertisement future understand how many by user reading estimate.
Advertisement dispensing side is usually to be counted to advertisement quantity in stock in the period with a certain longer period (such as 30 days), profit
It is fitted to obtain the matched curve of advertisement quantity in stock variation with statistical data, be launched in following inventory by the curve prediction advertisement
Situation of change.It is longer with measurement period that longer period (such as 30 days) is the period, to the prediction of rate of return on investment not enough and
When, advertisement dispensing side wants to predict investment in advertising return rate in time by short-term data statistics.
But due to advertising user negligible amounts, the individual data fluctuation in short term statistics data can all influence entirety
The variation tendency of statistical data influences the curve matching to statistical data, and then causes forecasting inaccuracy true.Therefore, how to avoid
Influence of the individual data fluctuation to short term statistics data is had become by short term statistics data Accurate Prediction investment in advertising return rate
For the active demand of advertisement dispensing side.
Invention content
Based on above-mentioned state of the art, the present invention proposes a kind of method and device of prediction investment in advertising return rate, energy
Enough go out investment in advertising return rate using short-term advertisement quantity in stock statistical data Accurate Prediction.
In order to achieve the above object, the following technical solutions are proposed by the present invention:
A method of prediction investment in advertising return rate, including:
It obtains the spending on ads data of the target date of target channel and is set after launching advertisement from the target date
Advertisement quantity in stock week mean value statistical data in time interval;Wherein, the N in the advertisement quantity in stock week mean value statistical data
It advertisement quantity in stock is to launch the N days advertisement quantitys in stock after advertisement daily in one week since the target date
Average value, N are the integer more than zero;
According to the advertisement quantity in stock week mean value statistical data, it is calculated wide after the target date launches advertisement
Accuse quantity in stock anticipation function;
According to the advertisement quantity in stock anticipation function, determine that income is launched in the advertisement of the target date;
The spending on ads data that income and the target date are launched according to the advertisement of the target date, are calculated
The rate of return on investment that the advertisement of the target date is launched.
Optionally, it according to the advertisement quantity in stock week mean value statistical data, is being calculated in target date dispensing
After advertisement quantity in stock anticipation function after advertisement, this method further includes:
Advertisement quantity in stock historical data of the target channel in the set period of time before the target date is obtained,
And the advertisement quantity in stock week Data-Statistics set of metadata of similar data with the advertisement quantity in stock week mean value statistical data difference minimum;
According to the advertisement quantity in stock historical data and advertisement quantity in stock week Data-Statistics set of metadata of similar data, to described wide
It accuses quantity in stock anticipation function and is modified processing.
Optionally, described according to the advertisement quantity in stock historical data and advertisement quantity in stock week Data-Statistics similarity number
According to, processing is modified to the advertisement quantity in stock anticipation function, including:
According to the advertisement quantity in stock anticipation function and the advertisement quantity in stock historical data, the target date is determined
Advertisement quantity in stock stage of stable development starting point;
According to the data after the stage of stable development starting point in the advertisement quantity in stock week mean value statistical data, it is calculated
The first advertisement quantity in stock anticipation function after the target date launches advertisement;
According to the advertisement quantity in stock historical data and advertisement quantity in stock week Data-Statistics set of metadata of similar data, it is calculated
Correct slope;
According to the advertisement quantity in stock week mean value statistical data, amendment intercept is calculated;
Place is modified to the first advertisement quantity in stock anticipation function using the amendment slope and the amendment intercept
Reason, obtains revised advertisement quantity in stock anticipation function.
Optionally, described according to the advertisement quantity in stock historical data and advertisement quantity in stock week Data-Statistics similarity number
According to, amendment slope is calculated, including:
According to the advertisement quantity in stock historical data, the second advertisement quantity in stock anticipation function is calculated, and according to institute
Advertisement quantity in stock week Data-Statistics set of metadata of similar data is stated, third advertisement quantity in stock anticipation function is calculated;
According to the slope of the second advertisement quantity in stock anticipation function and the third advertisement quantity in stock anticipation function, calculate
It obtains correcting slope.
Optionally, described according to the advertisement quantity in stock anticipation function, determine that income is launched in the advertisement of the target date,
Including:
According to the advertisement quantity in stock anticipation function, advertisement inventory daily after the target date launches advertisement is determined
Amount;Wherein, the daily advertisement quantity in stock is to the data after daily advertisement quantity in stock rounding;
Calculate all advertisement quantitys in stock for being reduced to for zero period after launching advertisement from the target date to advertisement quantity in stock
Income is launched in summation, the advertisement as the target date.
A kind of device of prediction investment in advertising return rate, including:
First data capture unit, the spending on ads data of the target date for obtaining target channel and from the mesh
Advertisement quantity in stock week mean value statistical data after the dispensing advertisement of mark date in setting time section;Wherein, the advertisement quantity in stock
The N days advertisement quantitys in stock in all mean value statistical data are after launching advertisement daily in one week since the target date
The N days advertisement quantitys in stock average value, N is integer more than zero;
Function calculating unit, for according to the advertisement quantity in stock week mean value statistical data, being calculated in the target
Date launches the advertisement quantity in stock anticipation function after advertisement;
Income determination unit, for according to the advertisement quantity in stock anticipation function, determining that the advertisement of the target date is thrown
Put income;
Rate of return on investment computing unit, for launching income and the target date according to the advertisement of the target date
The rate of return on investment that the advertisement of the target date is launched is calculated in spending on ads data.
Optionally, described device further includes:
Second data capture unit, for obtaining the target channel in the set period of time before the target date
Advertisement quantity in stock historical data, and advertisement quantity in stock week with the advertisement quantity in stock week mean value statistical data difference minimum
Equal Data-Statistics set of metadata of similar data;
Function amending unit, for according to the advertisement quantity in stock historical data and advertisement quantity in stock week equal Data-Statistics
Set of metadata of similar data is modified processing to the advertisement quantity in stock anticipation function.
Optionally, the function amending unit, including:
Stage of stable development determination unit, for according to the advertisement quantity in stock anticipation function and the advertisement quantity in stock history number
According to determining the advertisement quantity in stock stage of stable development starting point of the target date;
First computing unit, for according to the stage of stable development starting point in the advertisement quantity in stock week mean value statistical data it
The first advertisement quantity in stock anticipation function after the target date launches advertisement is calculated in data afterwards;
Second computing unit, for according to the advertisement quantity in stock historical data and advertisement quantity in stock week equal Data-Statistics
Amendment slope is calculated in set of metadata of similar data;
Third computing unit, for according to the advertisement quantity in stock week mean value statistical data, amendment intercept to be calculated;
Amending unit, for predicting letter to the first advertisement quantity in stock using the amendment slope and the amendment intercept
Number is modified processing, obtains revised advertisement quantity in stock anticipation function.
Optionally, second computing unit is according to the advertisement quantity in stock historical data and the advertisement quantity in stock Zhou Jun
Data-Statistics set of metadata of similar data is specifically used for when amendment slope is calculated:
According to the advertisement quantity in stock historical data, the second advertisement quantity in stock anticipation function is calculated, and according to institute
Advertisement quantity in stock week Data-Statistics set of metadata of similar data is stated, third advertisement quantity in stock anticipation function is calculated;It is wide according to described second
The slope for accusing quantity in stock anticipation function and the third advertisement quantity in stock anticipation function, is calculated amendment slope.
Optionally, the income determination unit determines the target date according to the advertisement quantity in stock anticipation function
When income is launched in advertisement, it is specifically used for:
According to the advertisement quantity in stock anticipation function, advertisement inventory daily after the target date launches advertisement is determined
Amount;Wherein, the daily advertisement quantity in stock is to the data after daily advertisement quantity in stock rounding;It calculates from the target day
Phase launches all advertisement quantity in stock summations for being reduced to for zero period after advertisement to advertisement quantity in stock, wide as the target date
It accuses and launches income.
Using technical solution of the present invention, by obtaining the spending on ads data of the target date of target channel and from described
Advertisement quantity in stock week mean value statistical data after target date dispensing advertisement in setting time section, is calculated in the target
Date launches the advertisement quantity in stock anticipation function after advertisement, determines that income is launched in the advertisement of target date according to the anticipation function,
The spending on ads data that income and the target date are finally launched according to the advertisement of the target date are calculated described
The rate of return on investment that the advertisement of target date is launched.The above process only passes through advertisement inventory short-term after target date dispensing advertisement
The all mean value statistical data of amount just calculate the rate of return on investment that the advertisement of target date is launched.Week mean value statistical data therein
Avoiding individual data fluctuation influences the fluctuation of short term statistics data, has reached wide by short term statistics data Accurate Prediction
Accuse the purpose of rate of return on investment.
Description of the drawings
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow diagram of the method for prediction investment in advertising return rate provided in an embodiment of the present invention;
Fig. 2 is advertisement quantity in stock statistics table schematic diagram provided in an embodiment of the present invention;
Fig. 3 is the flow diagram of the method for another prediction investment in advertising return rate provided in an embodiment of the present invention;
Fig. 4 is a kind of structural schematic diagram of the device of prediction investment in advertising return rate provided in an embodiment of the present invention;
Fig. 5 is the structural schematic diagram of the device of another prediction investment in advertising return rate provided in an embodiment of the present invention.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation describes, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a kind of method of prediction investment in advertising return rate, shown in Figure 1, this method packets
It includes:
S101, obtain target channel target date spending on ads data and from the target date launch advertisement after
Advertisement quantity in stock week mean value statistical data in setting time section;Wherein, in the advertisement quantity in stock week mean value statistical data
The N days advertisement quantitys in stock be to launch the N days advertisement bases after advertisement daily in one week since the target date
The average value of storage, N are the integer more than zero;
Refer to a certain media outlets that advertiser launches advertisement specifically, above-mentioned target channel, for example, network, TV,
The one of which media outlets such as newspapers and periodicals.
Above-mentioned target date refers to the date of its investment in advertising return rate to be predicted of the embodiment of the present invention.For example, it is assumed that ×
××× year × moon 11, advertiser to a certain target channel launched advertisement, if pre- using technical solution of the embodiment of the present invention
The investment in advertising return rate on above-mentioned ×××× year × moon 11 is surveyed, then was used as target day by above-mentioned ×××× year × moon 11
Phase.
The embodiment of the present invention is set, and for above-mentioned target channel, for the advertisement that some day launches, is all thrown from advertisement
The advertisement quantity in stock of dispensing is started to track and record from putting, physical record from launch advertisement after every day the channel it is wide
Accuse quantity in stock.And regular average smooth processing is carried out to the data of record.
Specific average smooth processing, which is batch, forms the data sum-average arithmetic on corresponding date and its rear some cycles pair
The period statistical data answered.The embodiment of the present invention is specifically using the advertisement in setting time section after launching advertisement from target date
Quantity in stock week, mean value statistical data was as advertisement quantity in stock anticipation function calculating basis.Also, advertisement quantity in stock week equal Data-Statistics
The N days advertisement quantitys in stock in data are to be launched daily the N days after advertisement in one week since the target date
The average value of advertisement quantity in stock, N are the integer more than zero.
According to above-mentioned computational methods, the advertisement quantity in stock week mean value statistical data since every day can be calculated,
Or month statistical data or season mean value statistical data etc..After these mean value statistical data are calculated, the present invention is implemented
Example stores result of calculation in sql databases together with original statistical data, when needing the investment in advertising to some day
When return rate is predicted, then corresponding data are read from sql databases.It, can be with when reading data from sql databases
It is read out by json strings, ' time ' field of json strings indicates to read number required for target date, ' name ' field indicate
According to time interval.
S102, according to the advertisement quantity in stock week mean value statistical data, be calculated the target date launch advertisement
Advertisement quantity in stock anticipation function afterwards;
Specifically, in the advertisement quantity in stock week for acquiring target date after mean value statistical data, by the wide of data i-th day
Accuse quantity in stock yiAs prediction target, to advertisement quantity in stock yiLogarithm carry out linear fit with statistics number of days i, using minimum two
The slope and intercept of multiplication digital simulation function, so that it is determined that advertisement quantity in stock yiLogarithm and statistics number of days i functional relation,
Also advertisement quantity in stock y has been determined thatiWith the functional relation of statistics number of days i, that is, it is calculated and launches advertisement in above-mentioned target date
Advertisement quantity in stock anticipation function afterwards.
The form of the anticipation function of the logarithm for the advertisement quantity in stock being calculated is:
ln(yi)=α * i+ β
Further, formal argument is carried out to the anticipation function of the logarithm of above-mentioned advertisement quantity in stock, can finally determined wide
Accuse quantity in stock anticipation function:
yi=eα*i+β
S103, according to the advertisement quantity in stock anticipation function, determine that income is launched in the advertisement of the target date;
Specifically, after determining advertisement quantity in stock anticipation function, you can calculated according to the anticipation function and thrown in target date
After putting advertisement, daily advertisement quantity in stock.Since the embodiment of the present invention is merely with all mean value statistical data, it will be understood that thing
The statistical data since target date in one week has only been counted in reality, and for data after a week, then it needs by above-mentioned
Advertisement quantity in stock anticipation function is predicted.
The process that advertisement quantity in stock decays to 0 is calculated by above-mentioned advertisement quantity in stock anticipation function for the embodiment of the present invention
In daily advertisement quantity in stock mesh can be calculated in conjunction with the advertisement inventory data in short-term (one week) counted
It marks after the date launches advertisement, until advertisement quantity in stock decays to advertisement quantity in stock daily during 0, then to above-mentioned daily
Advertisement quantity in stock is summed, and income is launched in the advertisement for obtaining above-mentioned target date.
S104, the spending on ads data that income and the target date are launched according to the advertisement of the target date, meter
It calculates and obtains the rate of return on investment that the advertisement of the target date is launched.
Specifically, launching income divided by the spending on ads data of the target date with the advertisement of target date, you can
The rate of return on investment launched to the advertisement of above-mentioned target date.
By above-mentioned introduction as it can be seen that using technical solution of the embodiment of the present invention, by the target date for obtaining target channel
Spending on ads data and the advertisement quantity in stock week equal primary system after launching advertisement from the target date in setting time section
It counts, the advertisement quantity in stock anticipation function after the target date launches advertisement is calculated, it is true according to the anticipation function
Income is launched in the advertisement on date of setting the goal, and finally launches the wide of income and the target date according to the advertisement of the target date
Injected volume data are accused, the rate of return on investment that the advertisement of the target date is launched is calculated.The above process only passes through target day
Short-term advertisement quantity in stock week mean value statistical data after phase dispensing advertisement, just calculates the investment time that the advertisement of target date is launched
Report rate.Week mean value statistical data therein, which avoids individual data fluctuation, influences the fluctuation of short term statistics data, reaches
Pass through the purpose of short term statistics data Accurate Prediction investment in advertising return rate.
Below for shown in Fig. 3, illustrate the method for above-mentioned prediction investment in advertising return rate disclosed by the embodiments of the present invention
Concrete processing procedure.
Shown in Figure 3, the method for prediction investment in advertising return rate disclosed by the embodiments of the present invention specifically includes:
S301, obtain target channel target date spending on ads data and from the target date launch advertisement after
Advertisement quantity in stock week mean value statistical data in setting time section;Wherein, in the advertisement quantity in stock week mean value statistical data
The N days advertisement quantitys in stock be to launch the N days advertisement bases after advertisement daily in one week since the target date
The average value of storage, N are the integer more than zero;
Refer to a certain media outlets that advertiser launches advertisement specifically, above-mentioned target channel, for example, network, TV,
The one of which media outlets such as newspapers and periodicals.
Above-mentioned target date refers to the date of its investment in advertising return rate to be predicted of the embodiment of the present invention.For example, it is assumed that ×
××× year × moon 11, advertiser to a certain target channel launched advertisement, if pre- using technical solution of the embodiment of the present invention
The investment in advertising return rate on above-mentioned ×××× year × moon 11 is surveyed, then was used as target day by above-mentioned ×××× year × moon 11
Phase.
The embodiment of the present invention is set, and for above-mentioned target channel, for the advertisement that some day launches, is all thrown from advertisement
The advertisement quantity in stock of dispensing is started to track and record from putting, the advertisement of physical record channel every day after launching advertisement
Quantity in stock.
Such as shown in Fig. 2, the clutter newly launched daily from 11 to 31 ×××× year × moon for a certain channel with
The tracking statistical result of daily advertisement quantity in stock, wherein indicate every day after same day dispensing advertisement in table per a line
Advertisement quantity in stock.It has been tracked 20 as can be seen that data are launched in the advertisement of × moon 11, during which data variation has not been
The full statistical law for following monotonic decreasing, but there is fluctuation upwards the 6th day, the 13rd day, this counts institute on the 11st for us
Variation tendency is very unfavorable in stock.In addition, in order to understand daily advertising input (activation amount) is finally obtained how many output
(how many advertisement base deposits the user's reading that is activated in total), it is intended that complete to count the channel institute that some day, input generated
Changing in stock, but since the part high-quality channel time-to-live is very long, tracking, which is more than one hundred days, still the inventory change that can not ignore,
Simple counting statistics data time consumption is too big.But directly initial data is predicted using mathematical method, odd-numbered day activation
The inventory change fluctuation of generation is again than more significant, so to carry out data by the way of regular average smooth pre- for the embodiment of the present invention
Processing.
Specific pretreatment is that the data sum-average arithmetic on corresponding date and its rear some cycles is formed corresponding week by batch
Phase statistical data.The embodiment of the present invention is specifically using the advertisement quantity in stock in setting time section after launching advertisement from target date
All mean value statistical data calculate basis as advertisement quantity in stock anticipation function.It is appreciated that advertisement quantity in stock week mean value statistical number
The N days advertisement quantitys in stock in are launch the N days after advertisement in one week since the target date daily wide
The average value of quantity in stock is accused, N is the integer more than zero.
Specifically, as shown in Fig. 2, to count the advertisement quantity in stock week mean value statistical data since the × moon 11,
Then at least need advertisement inventory data daily in 7 days since the × moon 11.It is appreciated that above-mentioned setting time area
Between included number of days, at least more than since target date, need calculate mean value statistical data number of days.Below in Fig. 2
Data illustrated for mean value statistical data to calculate the advertisement quantity in stock week since the × moon 11.
Above-mentioned set period of time is 20 days since the × moon 11 shown in Fig. 2.It is calculating since × the moon 11
When advertisement quantity in stock week mean value statistical data, by the N days from the × moon 11 to the × moon 17 in Fig. 2 in rectangular box region
Advertisement quantity in stock statistical data carry out sum-average arithmetic by row, obtained data are the advertisement quantity in stock since the × moon 11
The N days advertisement quantitys in stock of all mean value statistical data.For example, it is assumed that the advertisement inventory of first day later calculating × moon 11
Amount, then by Fig. 2 × moon 11 after the data from the × moon 11 to the × moon 17 of a row where first day sum
It is average, obtain since × the moon 11 advertisement quantity in stock week mean value statistical data first day advertisement quantity in stock;Assuming that meter
Second day advertisement quantity in stock after calculation × moon 11, then by Fig. 2 × moon 11 after a row where second day from ×
Months 11 days data to the × moon 17 carry out sum-average arithmetic, obtain the advertisement quantity in stock week mean value statistical number since the × moon 11
According to second day advertisement quantity in stock;And so on, you can the advertisement quantity in stock week mean value since the × moon 11 is calculated
The advertisement quantity in stock of every day in statistical data.
It should be noted that for the 25th day advertisement quantity in stock week equal Data-Statistics in Fig. 2 since the × moon 11
The calculating of data is only counted on since above-mentioned set period of time has only taken the data on the 20th since the × moon 11
× the moon 31, the 25th day data of the statistical data less than seven days since the × moon 11 at this time, therefore its week cannot be calculated
Mean value statistical data, only when first day advertisement inventory data for counting next month again, that is, count from × opened the moon 17
After the advertisement quantity in stock on the 25th to begin, all equal primary systems of the 25th day advertisement quantity in stock started since × the moon 11 can be just calculated
It counts.
According to above-mentioned computational methods, the advertisement quantity in stock week mean value statistical data since every day can be calculated,
Or month statistical data or season mean value statistical data etc..After these mean value statistical data are calculated, the present invention is implemented
Example stores result of calculation in sql databases together with original statistical data, when needing the investment in advertising to some day
When return rate is predicted, then corresponding data are read from sql databases.For example, it is assumed that using technology of the embodiment of the present invention
Scheme calculates the rate of return on investment that the advertisement of above-mentioned ×××× year × moon 11 is launched, then from the sql databases being previously stored
It the middle spending on ads data for reading above-mentioned ×××× year × moon 11 and was launched from above-mentioned ×××× year × moon 11 wide
Advertisement quantity in stock week mean value statistical data after announcement in setting time section, such as read from ×××× year × moon 11 and launch
Advertisement quantity in stock week mean value statistical data in 20 days after advertisement.
Specifically, when reading data from sql databases, it can be gone here and there and be read out by json, json strings
' time ' field indicates that target date, ' name ' field indicate the required time interval for reading data.
S302, according to the advertisement quantity in stock week mean value statistical data, be calculated the target date launch advertisement
Advertisement quantity in stock anticipation function afterwards;
Specifically, in the advertisement quantity in stock week for acquiring target date after mean value statistical data, by the wide of data i-th day
Accuse quantity in stock yiAs prediction target, to advertisement quantity in stock yiLogarithm carry out linear fit with statistics number of days i, using minimum two
The slope and intercept of multiplication digital simulation function, so that it is determined that advertisement quantity in stock yiLogarithm and statistics number of days i functional relation,
Also advertisement quantity in stock y has been determined thatiWith the functional relation of statistics number of days i, that is, it is calculated and launches advertisement in above-mentioned target date
Advertisement quantity in stock anticipation function afterwards.
The form of the anticipation function of the logarithm for the advertisement quantity in stock being calculated is:
ln(yi)=α * i+ β
Further, formal argument is carried out to the anticipation function of the logarithm of above-mentioned advertisement quantity in stock, can finally determined wide
Accuse quantity in stock anticipation function:
yi=eα*i+β
S303, advertisement quantity in stock history of the target channel in the set period of time before the target date is obtained
Data, and the advertisement quantity in stock week Data-Statistics similarity number with the advertisement quantity in stock week mean value statistical data difference minimum
According to;
Specifically, advertisement quantity in stock historical data of the above-mentioned channel in the set period of time before the target date,
It refer to the daily advertisement quantity in stock statistical data of above-mentioned channel in the set period of time before above-mentioned target date.On for example,
State the advertisement quantity in stock historical data in three months before channel ×××× year × moon 11, refer to from ×××× year ×
Started within 90th to the daily advertisement quantity in stock statistical number of the above-mentioned channel counted between ×××× year × moon 11 before the moon 11
According to.
The acquisition of historical data can directly be read according to time interval from database.
It is above-mentioned similar to the advertisement quantity in stock week Data-Statistics of the advertisement quantity in stock week mean value statistical data difference minimum
Data refer to all prediction data yiTo advertisement quantity in stock y on the firstth1After being normalized, in advertisement quantity in stock week equal primary system
Count the data that all durations are more than the wherein preceding 7 days and prediction data difference minimum chosen in data on the 50th.
In embodiments of the present invention, the difference of statistical data and prediction data is represented by:
Wherein, yiIndicate the advertisement quantity in stock on the i-thth predicted using the anticipation function obtained in step S202;Yi
Indicate advertisement quantity in stock on the i-thth of statistical data;H is the set (H={ 1,2,3,4,5,6,7 }) of the first seven day, | H | it represents
Number wherein to the data that should be greater than 0.
According to the difference of above-mentioned formula counting statistics data and prediction data, you can select first 7 days it is poor with prediction data
The data of different minimum, the advertisement quantity in stock week Data-Statistics set of metadata of similar data as above-mentioned advertisement quantity in stock week mean value statistical data.
S304, according to the advertisement quantity in stock anticipation function and the advertisement quantity in stock historical data, determine the target
The advertisement quantity in stock stage of stable development starting point on date;
Specifically, the advertisement quantity in stock anticipation function being calculated by step S302 is close to logarithmic curve form, still
The advertisement quantity in stock development that the later stage is launched in advertisement is more nearly with power exponent curve, it is believed that the early period after advertisement dispensing
The variation of advertisement quantity in stock is jiggly.Therefore, the embodiment of the present invention thinks, after the variation of advertisement quantity in stock enters the stage of stable development,
Accurate Prediction could be carried out to it, also only utilize advertisement quantity in stock stage of stable development data, can just obtain accurate advertisement quantity in stock
Anticipation function.
Therefore, the embodiment of the present invention is after step S302 tentatively obtains advertisement quantity in stock anticipation function, further according to just
Obtained advertisement quantity in stock anticipation function and advertisement quantity in stock historical data are walked, it is wide to determine that the channel is launched in above-mentioned target date
Advertisement quantity in stock stage of stable development starting point after announcement.
Specifically, the embodiment of the present invention calculates the prediction of the different stage of stable development starting points of advertisement quantity in stock historical data by enumerating
The difference of data and truthful data therefrom finds out the stage of stable development starting point of difference minimum, the i.e. advertisement base as above-mentioned target date
Storage stage of stable development starting point.
The prediction data of the different stage of stable development starting points of advertisement quantity in stock historical data and the difference of truthful data can be expressed as:
Wherein, f (i) is indicated i-th day be calculated using the advertisement quantity in stock anticipation function obtained in step S302
Advertisement quantity in stock;yiIndicate real ad quantity in stock on the i-thth;T indicate the later stage stage of stable development, in embodiments of the present invention for 50,
60,70,80,90,100 }, | T | represent y in the later stageiThe number of data more than 0.
According to above-mentioned formula, minimum D is calculatedT, corresponding T is advertisement quantity in stock stage of stable development starting point.
It should be noted that the embodiment of the present invention goes through the data in 90 days before above-mentioned target date as advertisement quantity in stock
History data calculate advertisement quantity in stock stage of stable development starting point and can flexibly be set according to data volume when the embodiments of the present invention are specifically implemented
Determine the time interval of advertisement quantity in stock historical data.Theoretically, the time interval of advertisement quantity in stock historical data cannot be less than 60
Day, if the time interval of the advertisement quantity in stock historical data of setting channel was less than 60 days, then it is assumed that the channel is not believable
Historical data can be at this time stage of stable development starting point assignment, such as can set stage of stable development starting point to 5, i.e., from the 5th day after advertisement dispensing
It plays advertisement quantity in stock and enters the stage of stable development.
S305, according to the data after the stage of stable development starting point in the advertisement quantity in stock week mean value statistical data, meter
It calculates and obtains the first advertisement quantity in stock anticipation function after the target date launches advertisement;
Specifically, after determining advertisement quantity in stock stage of stable development starting point, for the advertisement quantity in stock week obtained in step S301
The data after above-mentioned stage of stable development starting point in mean value statistical data, re-start the calculating of advertisement quantity in stock anticipation function.Tool
The calculating process of body is identical as the calculating process of step S302, and details are not described herein again.In order to be calculated in step S302
Advertisement quantity in stock anticipation function is distinguished, and the advertisement quantity in stock anticipation function that this step is calculated is named as the first advertisement quantity in stock
Anticipation function.
In the form such as step S302 of first advertisement quantity in stock anticipation function shown in formula:
yi=eα*i+β
S306, according to the advertisement quantity in stock historical data, the second advertisement quantity in stock anticipation function, Yi Jigen is calculated
According to advertisement quantity in stock week Data-Statistics set of metadata of similar data, third advertisement quantity in stock anticipation function is calculated;
Specifically, after getting advertisement quantity in stock historical data and all Data-Statistics set of metadata of similar data of advertisement quantity in stock, by
The long long term data more than 60 days in the presence of the two is all, it is possible to directly first be sought pair by what is introduced in step S302
The mode that number carries out least square fitting again obtains its corresponding fitting result, i.e., corresponding advertisement quantity in stock anticipation function.
For the ease of distinguishing, the advertisement quantity in stock anticipation function that the embodiment of the present invention will be calculated according to advertisement quantity in stock historical data
It is named as the second advertisement quantity in stock anticipation function, the advertisement that will be calculated according to advertisement quantity in stock week equal Data-Statistics set of metadata of similar data
Quantity in stock anticipation function is named as third advertisement quantity in stock anticipation function.
S307, according to the oblique of the second advertisement quantity in stock anticipation function and the third advertisement quantity in stock anticipation function
Amendment slope is calculated in rate;
Specifically, the embodiment of the present invention is pre- to above-mentioned second advertisement quantity in stock anticipation function and above-mentioned third advertisement quantity in stock
The slope for surveying function is weighted to obtain amendment slope.
Shown in the following formula of specific weighting scheme:
Wherein, αtarFor the amendment slope being calculated;αhistAnd αsimRespectively the second advertisement quantity in stock anticipation function and
The slope of third advertisement quantity in stock anticipation function;WithRespectively the second advertisement quantity in stock anticipation function and third advertisement base
The data difference of the stage of stable development starting point of storage anticipation function.
S308, according to the advertisement quantity in stock week mean value statistical data, amendment intercept is calculated;
Specifically, amendment slope is calculated in step S307tarAfterwards, it will correct and be calculated in slope substitution step S305
In obtained advertisement quantity in stock anticipation function, advertisement quantity in stock anticipation function is obtained:
Then any one day advertisement inventory data by the advertisement quantity in stock week of acquisition in mean value statistical data, and
This day is which day (i.e. the value of i) after target date substitutes into above-mentioned advertisement quantity in stock anticipation function, and amendment is calculated and cuts
Away from βtar。
S309, the first advertisement quantity in stock anticipation function is repaiied using the amendment slope and the amendment intercept
Positive processing, obtains revised advertisement quantity in stock anticipation function;
Specifically, the amendment slope that will be calculated in above-mentioned steps S307 and S308tarWith amendment intercept βtarIt replaces respectively
Change the advertisement quantity in stock anticipation function y obtained in step S305i=eα*i+βIn slope and intercept β, obtain revised advertisement
Quantity in stock anticipation function:
S310, according to the advertisement quantity in stock anticipation function, determine advertisement daily after the target date launches advertisement
Quantity in stock;Wherein, the daily advertisement quantity in stock is to the data after daily advertisement quantity in stock rounding;
Specifically, according to the revised advertisement quantity in stock anticipation function determined in step S309It can
To substitute into different date i, it is calculated after target date launches advertisement, daily advertisement quantity in stock.Due to referring to function
Value is never to decay to 0, this is not inconsistent with reality, therefore the embodiment of the present invention is to the above-mentioned daily advertisement being calculated
Quantity in stock carries out rounding, when the advertisement quantity in stock being calculated is 0, that is, thinks that the advertisement dispensing of the target date no longer generates
Income.
It should be noted that since the embodiment of the present invention is the advertisement quantity in stock system by short-term (one week) interior target date
Then the advertisement quantity in stock anticipation function for counting to determine utilizes the anticipation function to predict the advertisement base in the later stage not yet counted
Therefore data on stock can directly utilize the advertisement inventory data after target date for having counted, and for still
The advertisement inventory data in nonevent later stage, then can be predicted by above-mentioned formula.
S311, all advertisement bases being reduced to advertisement quantity in stock after launching advertisement from the target date during zero are calculated
Income is launched in storage summation, the advertisement as the target date;
Specifically, daily advertisement quantity in stock after target date dispensing advertisement is determined by step S310, until advertisement
Quantity in stock is kept to 0.Then by (including the advertisement inventory data of statistics and pre- of daily advertisement quantity in stock summation after target date
The advertisement inventory data of survey), obtained advertisement quantity in stock total amount launches income as the advertisement of the target date.
S312, the spending on ads data that income and the target date are launched according to the advertisement of the target date, meter
It calculates and obtains the rate of return on investment that the advertisement of the target date is launched.
Specifically, launching income divided by the spending on ads data of the target date with the advertisement of target date, you can
The rate of return on investment launched to the advertisement of above-mentioned target date.
The embodiment of the invention also discloses a kind of device of prediction investment in advertising return rate, shown in Figure 4, the device packets
It includes:
First data capture unit 100, the spending on ads data of the target date for obtaining target channel and from institute
It states target date and launches advertisement quantity in stock week mean value statistical data after advertisement in setting time section;Wherein, the advertisement base
The N days advertisement quantitys in stock in storage week mean value statistical data be launched daily in one week since the target date it is wide
The average value of the N days advertisement quantitys in stock after announcement, N are the integer more than zero;
Function calculating unit 110, for according to the advertisement quantity in stock week mean value statistical data, being calculated in the mesh
It marks the date and launches the advertisement quantity in stock anticipation function after advertisement;
Income determination unit 120, for according to the advertisement quantity in stock anticipation function, determining the advertisement of the target date
Launch income;
Rate of return on investment computing unit 130, for launching income and the target day according to the advertisement of the target date
The rate of return on investment that the advertisement of the target date is launched is calculated in the spending on ads data of phase.
Specifically, the specific works content of each unit in the present embodiment, refers to the content of above method embodiment,
Details are not described herein again.
Shown in Figure 5 in another embodiment of invention, described device further includes:
Second data capture unit 140, for obtaining setting time of the target channel before the target date
Advertisement quantity in stock historical data in section, and the advertisement inventory with the advertisement quantity in stock week mean value statistical data difference minimum
The all Data-Statistics set of metadata of similar data of amount;
Function amending unit 150, for according to the advertisement quantity in stock historical data and the advertisement quantity in stock week mean value
Set of metadata of similar data is counted, processing is modified to the advertisement quantity in stock anticipation function.
Wherein, the function amending unit 150, including:
Stage of stable development determination unit, for according to the advertisement quantity in stock anticipation function and the advertisement quantity in stock history number
According to determining the advertisement quantity in stock stage of stable development starting point of the target date;
First computing unit, for according to the stage of stable development starting point in the advertisement quantity in stock week mean value statistical data it
The first advertisement quantity in stock anticipation function after the target date launches advertisement is calculated in data afterwards;
Second computing unit, for according to the advertisement quantity in stock historical data and advertisement quantity in stock week equal Data-Statistics
Amendment slope is calculated in set of metadata of similar data;
Third computing unit, for according to the advertisement quantity in stock week mean value statistical data, amendment intercept to be calculated;
Amending unit, for predicting letter to the first advertisement quantity in stock using the amendment slope and the amendment intercept
Number is modified processing, obtains revised advertisement quantity in stock anticipation function.
Second computing unit is according to the advertisement quantity in stock historical data and advertisement quantity in stock week equal Data-Statistics
Set of metadata of similar data is specifically used for when amendment slope is calculated:
According to the advertisement quantity in stock historical data, the second advertisement quantity in stock anticipation function is calculated, and according to institute
Advertisement quantity in stock week Data-Statistics set of metadata of similar data is stated, third advertisement quantity in stock anticipation function is calculated;It is wide according to described second
The slope for accusing quantity in stock anticipation function and the third advertisement quantity in stock anticipation function, is calculated amendment slope.
Specifically, the specific works content of each unit in above-described embodiment, refers to the interior of above method embodiment
Hold, details are not described herein again.
It discloses in another embodiment of the present invention, above-mentioned income determination unit 120 is according to the advertisement quantity in stock
Anticipation function is specifically used for when determining that income is launched in the advertisement of the target date:
According to the advertisement quantity in stock anticipation function, advertisement inventory daily after the target date launches advertisement is determined
Amount;Wherein, the daily advertisement quantity in stock is to the data after daily advertisement quantity in stock rounding;It calculates from the target day
Phase launches all advertisement quantity in stock summations for being reduced to for zero period after advertisement to advertisement quantity in stock, wide as the target date
It accuses and launches income.
Specifically, the specific works content of the income determination unit 120 in the present embodiment, refers to above method embodiment
Content, details are not described herein again.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only that
A little elements, but also include other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Each embodiment is described by the way of progressive in this specification, the highlights of each of the examples are with other
The difference of embodiment, just to refer each other for identical similar portion between each embodiment.
The foregoing description of the disclosed embodiments enables those skilled in the art to implement or use the present invention.
Various modifications to these embodiments will be apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, of the invention
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest range caused.
Claims (10)
1. a kind of method of prediction investment in advertising return rate, which is characterized in that including:
Setting time after obtaining the spending on ads data of the target date of target channel and launching advertisement from the target date
Advertisement quantity in stock week mean value statistical data in section;Wherein, the N days in the advertisement quantity in stock week mean value statistical data
Advertisement quantity in stock is to launch being averaged for the N days advertisement quantitys in stock after advertisement daily in one week since the target date
Value, N are the integer more than zero;
According to the advertisement quantity in stock week mean value statistical data, the advertisement base after the target date launches advertisement is calculated
Storage anticipation function;
According to the advertisement quantity in stock anticipation function, determine that income is launched in the advertisement of the target date;
The spending on ads data that income and the target date are launched according to the advertisement of the target date are calculated described
The rate of return on investment that the advertisement of target date is launched.
2. according to the method described in claim 1, it is characterized in that, according to the advertisement quantity in stock week mean value statistical data,
It is calculated after the advertisement quantity in stock anticipation function that the target date launches after advertisement, this method further includes:
Advertisement quantity in stock historical data of the target channel in the set period of time before the target date is obtained, and
With the advertisement quantity in stock week Data-Statistics set of metadata of similar data of the advertisement quantity in stock week mean value statistical data difference minimum;
According to the advertisement quantity in stock historical data and advertisement quantity in stock week Data-Statistics set of metadata of similar data, to the advertisement base
Storage anticipation function is modified processing.
3. according to the method described in claim 2, it is characterized in that, described according to the advertisement quantity in stock historical data and described
The all Data-Statistics set of metadata of similar data of advertisement quantity in stock, processing is modified to the advertisement quantity in stock anticipation function, including:
According to the advertisement quantity in stock anticipation function and the advertisement quantity in stock historical data, the advertisement of the target date is determined
Quantity in stock stage of stable development starting point;
According to the data after the stage of stable development starting point in the advertisement quantity in stock week mean value statistical data, it is calculated in institute
It states target date and launches the first advertisement quantity in stock anticipation function after advertisement;
According to the advertisement quantity in stock historical data and advertisement quantity in stock week Data-Statistics set of metadata of similar data, amendment is calculated
Slope;
According to the advertisement quantity in stock week mean value statistical data, amendment intercept is calculated;
Processing is modified to the first advertisement quantity in stock anticipation function using the amendment slope and the amendment intercept, is obtained
To revised advertisement quantity in stock anticipation function.
4. according to the method described in claim 3, it is characterized in that, described according to the advertisement quantity in stock historical data and described
The all Data-Statistics set of metadata of similar data of advertisement quantity in stock, are calculated amendment slope, including:
According to the advertisement quantity in stock historical data, the second advertisement quantity in stock anticipation function is calculated, and according to described wide
Quantity in stock week Data-Statistics set of metadata of similar data is accused, third advertisement quantity in stock anticipation function is calculated;
According to the slope of the second advertisement quantity in stock anticipation function and the third advertisement quantity in stock anticipation function, it is calculated
Correct slope.
5. according to the method described in claim 1, it is characterized in that, described according to the advertisement quantity in stock anticipation function, determination
Income is launched in the advertisement of the target date, including:
According to the advertisement quantity in stock anticipation function, advertisement quantity in stock daily after the target date launches advertisement is determined;Its
In, the daily advertisement quantity in stock is to the data after daily advertisement quantity in stock rounding;
All advertisement quantity in stock summations for being reduced to for zero period after launching advertisement from the target date to advertisement quantity in stock are calculated,
Income is launched in advertisement as the target date.
6. a kind of device of prediction investment in advertising return rate, which is characterized in that including:
First data capture unit, the spending on ads data of the target date for obtaining target channel and from the target day
Advertisement quantity in stock week mean value statistical data after phase dispensing advertisement in setting time section;Wherein, the advertisement quantity in stock Zhou Jun
Primary system count in the N days advertisement quantitys in stock be after advertisement is launched in one week since the target date daily the
The average value of N days advertisement quantitys in stock, N are the integer more than zero;
Function calculating unit, for according to the advertisement quantity in stock week mean value statistical data, being calculated in the target date
Launch the advertisement quantity in stock anticipation function after advertisement;
Income determination unit, for according to the advertisement quantity in stock anticipation function, determining that the advertisement of the target date is launched and receiving
Benefit;
Rate of return on investment computing unit, the advertisement for launching income and the target date according to the advertisement of the target date
The rate of return on investment that the advertisement of the target date is launched is calculated in injected volume data.
7. device according to claim 6, which is characterized in that described device further includes:
Second data capture unit, it is wide in the set period of time before the target date for obtaining the target channel
Accuse quantity in stock historical data, and the advertisement quantity in stock week mean value with the advertisement quantity in stock week mean value statistical data difference minimum
Count set of metadata of similar data;
Function amending unit, for similar with advertisement quantity in stock week Data-Statistics according to the advertisement quantity in stock historical data
Data are modified processing to the advertisement quantity in stock anticipation function.
8. device according to claim 7, which is characterized in that the function amending unit, including:
Stage of stable development determination unit is used for according to the advertisement quantity in stock anticipation function and the advertisement quantity in stock historical data, really
The advertisement quantity in stock stage of stable development starting point of the fixed target date;
First computing unit, after according to the stage of stable development starting point in the advertisement quantity in stock week mean value statistical data
The first advertisement quantity in stock anticipation function after the target date launches advertisement is calculated in data;
Second computing unit, for similar with advertisement quantity in stock week Data-Statistics according to the advertisement quantity in stock historical data
Amendment slope is calculated in data;
Third computing unit, for according to the advertisement quantity in stock week mean value statistical data, amendment intercept to be calculated;
Amending unit, for using the amendment slope and the amendment intercept to the first advertisement quantity in stock anticipation function into
Row correcting process obtains revised advertisement quantity in stock anticipation function.
9. device according to claim 8, which is characterized in that second computing unit is gone through according to the advertisement quantity in stock
History data and advertisement quantity in stock week Data-Statistics set of metadata of similar data are specifically used for when amendment slope is calculated:
According to the advertisement quantity in stock historical data, the second advertisement quantity in stock anticipation function is calculated, and according to described wide
Quantity in stock week Data-Statistics set of metadata of similar data is accused, third advertisement quantity in stock anticipation function is calculated;According to second advertisement base
The slope of storage anticipation function and the third advertisement quantity in stock anticipation function, is calculated amendment slope.
10. device according to claim 6, which is characterized in that the income determination unit is according to the advertisement quantity in stock
Anticipation function is specifically used for when determining that income is launched in the advertisement of the target date:
According to the advertisement quantity in stock anticipation function, advertisement quantity in stock daily after the target date launches advertisement is determined;Its
In, the daily advertisement quantity in stock is to the data after daily advertisement quantity in stock rounding;It calculates and is thrown from the target date
It puts and is reduced to all advertisement quantity in stock summations during zero to advertisement quantity in stock after advertisement, the advertisement as the target date is thrown
Put income.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810310740.3A CN108564404B (en) | 2018-04-09 | 2018-04-09 | Method and device for predicting return on investment of advertisement |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810310740.3A CN108564404B (en) | 2018-04-09 | 2018-04-09 | Method and device for predicting return on investment of advertisement |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108564404A true CN108564404A (en) | 2018-09-21 |
CN108564404B CN108564404B (en) | 2021-10-15 |
Family
ID=63534392
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810310740.3A Active CN108564404B (en) | 2018-04-09 | 2018-04-09 | Method and device for predicting return on investment of advertisement |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108564404B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109697635A (en) * | 2018-12-27 | 2019-04-30 | 拉扎斯网络科技(上海)有限公司 | One kind is bidded acquisition methods, device, electronic equipment and medium |
CN111222915A (en) * | 2019-12-31 | 2020-06-02 | 上海昌投网络科技有限公司 | Public number ROI (region of interest) estimation method and device based on linear regression model |
CN112232862A (en) * | 2020-10-12 | 2021-01-15 | 上海移卓网络科技有限公司 | Method, device, equipment and storage medium for predicting advertisement business channel ROI |
CN112258245A (en) * | 2020-11-10 | 2021-01-22 | 深圳市欢太科技有限公司 | Control method, device and storage medium for advertisement putting |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105894136A (en) * | 2016-05-26 | 2016-08-24 | 北京京东尚博广益投资管理有限公司 | Category inventory prediction method and prediction device |
CN106056239A (en) * | 2016-05-24 | 2016-10-26 | 北京京东尚博广益投资管理有限公司 | Product inventory predicting method and product inventory predicting device |
CN106803190A (en) * | 2017-01-03 | 2017-06-06 | 北京掌阔移动传媒科技有限公司 | A kind of ad personalization supplying system and method |
CN107871247A (en) * | 2017-02-08 | 2018-04-03 | 湖南快乐阳光互动娱乐传媒有限公司 | Signposted before one kind stock's predictor method |
-
2018
- 2018-04-09 CN CN201810310740.3A patent/CN108564404B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106056239A (en) * | 2016-05-24 | 2016-10-26 | 北京京东尚博广益投资管理有限公司 | Product inventory predicting method and product inventory predicting device |
CN105894136A (en) * | 2016-05-26 | 2016-08-24 | 北京京东尚博广益投资管理有限公司 | Category inventory prediction method and prediction device |
CN106803190A (en) * | 2017-01-03 | 2017-06-06 | 北京掌阔移动传媒科技有限公司 | A kind of ad personalization supplying system and method |
CN107871247A (en) * | 2017-02-08 | 2018-04-03 | 湖南快乐阳光互动娱乐传媒有限公司 | Signposted before one kind stock's predictor method |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109697635A (en) * | 2018-12-27 | 2019-04-30 | 拉扎斯网络科技(上海)有限公司 | One kind is bidded acquisition methods, device, electronic equipment and medium |
CN111222915A (en) * | 2019-12-31 | 2020-06-02 | 上海昌投网络科技有限公司 | Public number ROI (region of interest) estimation method and device based on linear regression model |
CN112232862A (en) * | 2020-10-12 | 2021-01-15 | 上海移卓网络科技有限公司 | Method, device, equipment and storage medium for predicting advertisement business channel ROI |
CN112258245A (en) * | 2020-11-10 | 2021-01-22 | 深圳市欢太科技有限公司 | Control method, device and storage medium for advertisement putting |
Also Published As
Publication number | Publication date |
---|---|
CN108564404B (en) | 2021-10-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108564404A (en) | A kind of method and device of prediction investment in advertising return rate | |
US20230325887A1 (en) | Systems and methods for determining bids for placing advertisements | |
CN109657138A (en) | A kind of video recommendation method, device, electronic equipment and storage medium | |
CN111143738B (en) | Resource display method and device, electronic equipment and storage medium | |
US8473339B1 (en) | Automatically switching between pricing models for services | |
CN111651722B (en) | Risk assessment method and platform for advertisement putting effect of advertiser | |
CN101673385A (en) | Consumption preliminary estimate method and device thereof | |
CN103942236A (en) | System and method for serving electronic content | |
US20100241597A1 (en) | Dynamic estimation of the popularity of web content | |
CN103136280A (en) | System and method of automatically searching hot point video | |
CN108363730B (en) | Content recommendation method, system and terminal equipment | |
US20090094095A1 (en) | Recommendations based on an adoption curve | |
US20140058793A1 (en) | Forecasting a number of impressions of a prospective advertisement listing | |
CN106096047A (en) | User partition preference computational methods based on Information Entropy and system | |
CN110737859A (en) | UP main matching method and device | |
CN107368573A (en) | Video quality evaluation method and device | |
CN111899041A (en) | Information delivery processing method, information delivery device, information delivery equipment and storage medium | |
US20130080247A1 (en) | Ad Placement | |
CN107862556A (en) | A kind of put-on method and system of VIP advertisements | |
CN108846525A (en) | Dealing amount of foreign exchange prediction technique and device | |
CN106296247A (en) | The online sort method of network information resource and device | |
CN103309894A (en) | User attribute-based search realization method and system | |
CN106296287A (en) | For the method and apparatus adjusting the bid of promotion message | |
CN112465549A (en) | System and method for identifying channel cheating | |
CN111861542B (en) | Method, device, terminal and non-transitory computer readable storage medium for advertising |
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 |