CN108564404B - Method and device for predicting return on investment of advertisement - Google Patents
Method and device for predicting return on investment of advertisement Download PDFInfo
- Publication number
- CN108564404B CN108564404B CN201810310740.3A CN201810310740A CN108564404B CN 108564404 B CN108564404 B CN 108564404B CN 201810310740 A CN201810310740 A CN 201810310740A CN 108564404 B CN108564404 B CN 108564404B
- Authority
- CN
- China
- Prior art keywords
- advertisement
- data
- inventory
- advertisement inventory
- 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.)
- Active
Links
Images
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)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer 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 invention provides a method for predicting the return on investment of advertisements, which comprises the following steps: acquiring advertisement putting quantity data of a target date of a target channel and advertisement inventory weekly average statistical data in a set time interval after advertisements are put from the target date; calculating to obtain an advertisement inventory prediction function after the advertisement is put on the target date according to the weekly average statistical data of the advertisement inventory; determining the advertisement putting income of the target date according to the advertisement inventory prediction function; and calculating to obtain the return on investment of the advertisement delivery on the target date according to the advertisement delivery income on the target date and the advertisement delivery amount data on the target date. By adopting the technical scheme, the influence of individual data fluctuation on the fluctuation of short-term statistical data can be avoided, and the purpose of accurately predicting the return on investment of the advertisement through the short-term statistical data is achieved.
Description
Technical Field
The invention relates to the technical field of investment return rate prediction, in particular to a method and a device for predicting advertisement investment return rate.
Background
In advertisement promotion, the activation cost of each advertisement promotion channel needs to be examined, and finally, which channels are selected for advertisement putting is determined. The method is characterized in that the value of the advertisements which can be generated by the user in the future is activated, and the prediction of the return on investment rate is to predict how many advertisements put in the channel in the future can be read by the user through an existing accurate calculation formula trained by the change trend of the advertisement inventory (attenuation rate, inflection point, mean variance and the like after putting).
The advertisement delivery party usually counts the advertisement inventory at a certain period of time (for example, 30 days), and fits the statistical data to obtain a fitted curve of the advertisement inventory change, and predicts the inventory change of the advertisement delivery in the future through the curve. The statistical period with a long period of time (e.g., 30 days) as a period is long, and the prediction of the return on investment is not timely enough, and the advertisement delivery party hopes to predict the return on investment of the advertisement timely through short-term data statistics.
However, since the number of the advertisement users is small, individual data fluctuation in the short-term statistical data affects the variation trend of the overall statistical data, affects curve fitting to the statistical data, and further causes inaccurate prediction. Therefore, how to avoid the influence of individual data fluctuation on short-term statistical data and accurately predict the return on investment of the advertisement through the short-term statistical data becomes an urgent need of an advertisement delivery party.
Disclosure of Invention
Based on the current state of the prior art, the invention provides a method and a device for predicting the return on investment of advertisements, which can accurately predict the return on investment of advertisements by utilizing the statistical data of short-term advertisement inventory.
In order to achieve the purpose, the invention provides the following technical scheme:
a method of predicting return on investment for an advertisement, comprising:
acquiring advertisement putting quantity data of a target date of a target channel and advertisement inventory weekly average statistical data in a set time interval after advertisements are put from the target date; the advertisement inventory on the Nth day in the advertisement inventory weekly average statistical data is the average value of the advertisement inventory on the Nth day after advertisements are put every day in a week starting from the target date, and N is an integer greater than zero;
calculating to obtain an advertisement inventory prediction function after the advertisement is put on the target date according to the weekly average statistical data of the advertisement inventory;
determining the advertisement putting income of the target date according to the advertisement inventory prediction function;
and calculating to obtain the return on investment of the advertisement delivery on the target date according to the advertisement delivery income on the target date and the advertisement delivery amount data on the target date.
Optionally, after calculating a prediction function of the advertisement inventory after the advertisement is delivered on the target date according to the weekly average statistical data of the advertisement inventory, the method further includes:
acquiring historical advertisement inventory data of the target channel in a set time period before the target date and advertisement inventory weekly average statistical similarity data with the minimum difference with the advertisement inventory weekly average statistical data;
and correcting the advertisement stock prediction function according to the historical advertisement stock data and the similar advertisement stock weekly average statistical data.
Optionally, the modifying the advertisement inventory prediction function according to the advertisement inventory historical data and the advertisement inventory weekly average statistical similarity data includes:
determining the starting point of the stable period of the advertisement inventory on the target date according to the advertisement inventory prediction function and the historical data of the advertisement inventory;
calculating to obtain a first advertisement inventory prediction function after the advertisement is put on the target date according to the data after the stationary period starting point in the advertisement inventory weekly average statistical data;
counting similar data according to the historical data of the advertisement inventory and the weekly average value of the advertisement inventory, and calculating to obtain a correction slope;
calculating to obtain a correction intercept according to the week average statistical data of the advertisement inventory;
and correcting the first advertisement inventory prediction function by using the correction slope and the correction intercept to obtain a corrected advertisement inventory prediction function.
Optionally, the calculating to obtain a correction slope according to the advertisement inventory historical data and the advertisement inventory weekly average statistical similarity data includes:
calculating to obtain a second advertisement inventory prediction function according to the historical advertisement inventory data, and calculating to obtain a third advertisement inventory prediction function according to the weekly average statistical similarity data of the advertisement inventory;
and calculating to obtain a correction slope according to the slopes of the second advertisement inventory prediction function and the third advertisement inventory prediction function.
Optionally, the determining the advertisement delivery revenue of the target date according to the advertisement inventory prediction function includes:
determining the daily advertisement inventory after the advertisements are put on the target date according to the advertisement inventory prediction function; the daily advertisement inventory is data obtained by rounding the daily advertisement inventory;
and calculating the total inventory of all advertisements from the time when the advertisements are released on the target date to the time when the inventory of the advertisements is reduced to zero, and taking the total inventory of all advertisements as the advertisement release yield of the target date.
An apparatus for predicting return on investment for advertising, comprising:
the system comprises a first data acquisition unit, a second data acquisition unit and a third data acquisition unit, wherein the first data acquisition unit is used for acquiring advertisement putting quantity data of a target date of a target channel and advertisement inventory weekly average statistical data in a set time interval after advertisements are put on the target date; the advertisement inventory on the Nth day in the advertisement inventory weekly average statistical data is the average value of the advertisement inventory on the Nth day after advertisements are put every day in a week starting from the target date, and N is an integer greater than zero;
the function calculation unit is used for calculating and obtaining an advertisement inventory prediction function after the advertisement is put on the target date according to the week-average statistical data of the advertisement inventory;
the profit determining unit is used for determining the advertisement putting profit of the target date according to the advertisement inventory prediction function;
and the return on investment rate calculating unit is used for calculating the return on investment rate of the advertisement delivery on the target date according to the advertisement delivery income on the target date and the advertisement delivery amount data on the target date.
Optionally, the apparatus further comprises:
the second data acquisition unit is used for acquiring historical advertisement stock data of the target channel in a set time period before the target date and similar advertisement stock weekly average statistical data with the minimum difference with the advertisement stock weekly average statistical data;
and the function correction unit is used for correcting the advertisement stock prediction function according to the historical advertisement stock data and the similar advertisement stock weekly average statistical data.
Optionally, the function modifying unit includes:
the stationary period determining unit is used for determining the starting point of the stationary period of the advertisement inventory quantity of the target date according to the advertisement inventory quantity prediction function and the historical data of the advertisement inventory quantity;
the first calculation unit is used for calculating and obtaining a first advertisement inventory prediction function after the advertisement is delivered on the target date according to the data after the stationary period starting point in the advertisement inventory weekly average statistical data;
the second calculation unit is used for calculating to obtain a correction slope according to the historical advertisement stock data and the weekly average statistical similarity data of the advertisement stock;
the third calculating unit is used for calculating to obtain a correction intercept according to the week average statistical data of the advertisement stock;
and the correction unit is used for correcting the first advertisement stock prediction function by using the correction slope and the correction intercept to obtain a corrected advertisement stock prediction function.
Optionally, when the second calculating unit calculates to obtain the correction slope according to the historical advertisement inventory data and the weekly average statistical similarity of advertisement inventory, the second calculating unit is specifically configured to:
calculating to obtain a second advertisement inventory prediction function according to the historical advertisement inventory data, and calculating to obtain a third advertisement inventory prediction function according to the weekly average statistical similarity data of the advertisement inventory; and calculating to obtain a correction slope according to the slopes of the second advertisement inventory prediction function and the third advertisement inventory prediction function.
Optionally, when the revenue determining unit determines the revenue of advertisement delivery on the target date according to the advertisement inventory prediction function, the revenue determining unit is specifically configured to:
determining the daily advertisement inventory after the advertisements are put on the target date according to the advertisement inventory prediction function; the daily advertisement inventory is data obtained by rounding the daily advertisement inventory; and calculating the total inventory of all advertisements from the time when the advertisements are released on the target date to the time when the inventory of the advertisements is reduced to zero, and taking the total inventory of all advertisements as the advertisement release yield of the target date.
By adopting the technical scheme of the invention, the advertisement inventory prediction function after the advertisement is put on the target date is calculated and obtained by acquiring the advertisement putting quantity data of the target date of the target channel and the advertisement inventory weekly average statistical data in the set time interval after the advertisement is put on the target date, the advertisement putting income of the target date is determined according to the prediction function, and finally the investment return rate of the advertisement putting on the target date is calculated and obtained according to the advertisement putting income of the target date and the advertisement putting quantity data of the target date. In the process, the investment return rate of the advertisement delivery on the target date is calculated only through the short-term advertisement inventory weekly average statistical data after the advertisement delivery on the target date. The cycle-average statistical data avoids the fluctuation influence of individual data fluctuation on short-term statistical data, and the purpose of accurately predicting the return rate of advertisement investment through the short-term statistical data is achieved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a flow chart illustrating a method for predicting return on investment for advertisement according to an embodiment of the present invention;
FIG. 2 is a table of advertisement inventory statistics provided by an embodiment of the present invention;
FIG. 3 is a flow chart illustrating another method for predicting return on investment for advertisements according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an apparatus for predicting return on investment for advertisement according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of another apparatus for predicting return on investment for advertisement according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention discloses a method for predicting return on investment of advertisements, which is shown in figure 1 and comprises the following steps:
s101, obtaining advertisement putting quantity data of a target date of a target channel and advertisement inventory weekly average statistical data in a set time interval after advertisements are put on the target date; the advertisement inventory on the Nth day in the advertisement inventory weekly average statistical data is the average value of the advertisement inventory on the Nth day after advertisements are put every day in a week starting from the target date, and N is an integer greater than zero;
specifically, the target channel refers to a certain media channel for advertisers to deliver advertisements, such as one of a network, a television, a newspaper, a magazine, and the like.
The target date refers to a date for which the return on investment of the advertisement is to be predicted according to the embodiment of the present invention. For example, assuming that an advertiser places an advertisement on a certain target channel on day xxxxxxyextyextyextyext11, if the return on investment for the advertisement on day xxxxxxyextyextyextyextyextyextye11 is predicted by using the technical solution of the embodiment of the present invention, day xxxxxxxxyextyextyextyextyextyextyextyextyextye11 is used as a target date.
In the embodiment of the present invention, for the target channel, for the advertisement delivered on a certain day, the inventory quantity of the delivered advertisement is tracked from the date of advertisement delivery, and the inventory quantity of the advertisement in the channel is specifically recorded from each day after the advertisement is delivered. And the recorded data is subjected to a periodic average smoothing process.
The specific average smoothing processing is to add and average data of a corresponding date and a certain period after the date in batch to form corresponding period statistical data. The embodiment of the invention specifically adopts the statistical data of the weekly average value of the advertisement stock in a set time interval after the advertisement is put in from a target date as the calculation basis of the prediction function of the advertisement stock. And the advertisement inventory amount of the nth day in the advertisement inventory amount weekly average statistical data is an average value of the advertisement inventory amounts of the nth day after the advertisement is delivered every day in the week from the target date, and N is an integer greater than zero.
According to the calculation method, the statistical data of the weekly average value, the monthly average value, the quarterly average value and the like of the advertisement inventory stock from each day can be calculated. After the mean statistical data are calculated, the calculation result and the original statistical data are stored in the sql database, and when the return on investment of the advertisement in a certain day needs to be predicted, corresponding data are read from the sql database. When reading data from the sql database, the data can be read through a json string, wherein the 'time' field of the json string represents the target date and the 'name' field represents the time interval in which the data needs to be read.
S102, calculating to obtain an advertisement inventory prediction function after the advertisement is put on the target date according to the weekly average statistical data of the advertisement inventory;
specifically, after the week-average statistical data of the advertisement inventory quantity of the target date is obtained, the advertisement inventory quantity y of the ith date is obtainediAs a prediction target, for the advertisement inventory yiLinear fitting is carried out on the logarithm of the number of the advertisement inventory and the statistical number of days i, and the slope and the intercept of a fitting function are calculated by adopting a least square method, so that the advertisement inventory y is determinediThe logarithm of (d) and the number of statistical days i, the advertisement inventory y is determinediAnd calculating to obtain the function of predicting the stock quantity of the advertisements after the advertisements are placed on the target date.
The form of the prediction function of the logarithm of the calculated advertisement inventory quantity is as follows:
ln(yi)＝α*i+β
further, the prediction function of the logarithm of the advertisement stock quantity is subjected to form transformation, so that the prediction function of the advertisement stock quantity can be finally determined:
yi＝eα*i+β
s103, determining the advertisement release income of the target date according to the advertisement inventory prediction function;
specifically, after the advertisement inventory prediction function is determined, the daily advertisement inventory after the advertisement is delivered on the target date can be calculated according to the prediction function. Since the embodiment of the present invention only uses the week-average statistical data, it can be understood that only the statistical data in one week from the target date is actually counted, and for the data after one week, the prediction needs to be performed by the above-mentioned advertisement inventory prediction function.
According to the embodiment of the invention, the daily advertisement inventory in the process that the advertisement inventory is attenuated to 0 is calculated through the advertisement inventory prediction function, the daily advertisement inventory in the process that the advertisement inventory is attenuated to 0 after the advertisement is delivered on the target date can be calculated by combining the counted advertisement inventory data in a short period (one week), and then the daily advertisement inventory is summed to obtain the advertisement delivery income on the target date.
And S104, calculating to obtain the return on investment of the advertisement putting on the target date according to the advertisement putting income on the target date and the advertisement putting amount data on the target date.
Specifically, the return on investment of advertisement delivery on the target date can be obtained by dividing the advertisement delivery income on the target date by the advertisement delivery amount data on the target date.
It can be seen from the above description that, by adopting the technical scheme of the embodiment of the present invention, the advertisement inventory prediction function after the advertisement is delivered on the target date is calculated by obtaining the advertisement delivery amount data on the target date of the target channel and the advertisement inventory weekly average statistical data in the set time interval after the advertisement is delivered on the target date, the advertisement delivery revenue on the target date is determined according to the prediction function, and finally, the return on investment for the advertisement delivery on the target date is calculated according to the advertisement delivery revenue on the target date and the advertisement delivery amount data on the target date. In the process, the investment return rate of the advertisement delivery on the target date is calculated only through the short-term advertisement inventory weekly average statistical data after the advertisement delivery on the target date. The cycle-average statistical data avoids the fluctuation influence of individual data fluctuation on short-term statistical data, and the purpose of accurately predicting the return rate of advertisement investment through the short-term statistical data is achieved.
The following describes a specific processing procedure of the method for predicting return on investment of advertisements, which is disclosed by the embodiment of the present invention, by taking fig. 3 as an example.
Referring to fig. 3, a method for predicting return on investment of an advertisement disclosed in the embodiment of the present invention specifically includes:
s301, acquiring advertisement putting quantity data of a target date of a target channel and advertisement inventory weekly average statistical data in a set time interval after advertisements are put on the target date; the advertisement inventory on the Nth day in the advertisement inventory weekly average statistical data is the average value of the advertisement inventory on the Nth day after advertisements are put every day in a week starting from the target date, and N is an integer greater than zero;
specifically, the target channel refers to a certain media channel for advertisers to deliver advertisements, such as one of a network, a television, a newspaper, a magazine, and the like.
The target date refers to a date for which the return on investment of the advertisement is to be predicted according to the embodiment of the present invention. For example, assuming that an advertiser places an advertisement on a certain target channel on day xxxxxxyextyextyextyext11, if the return on investment for the advertisement on day xxxxxxyextyextyextyextyextyextye11 is predicted by using the technical solution of the embodiment of the present invention, day xxxxxxxxyextyextyextyextyextyextyextyextyextye11 is used as a target date.
In the embodiment of the invention, for the target channel, for the advertisements released in a certain day, the inventory quantity of the released advertisements is tracked and recorded from the date of the advertisement release, and the inventory quantity of the channel after the advertisement release is recorded in detail.
For example, as shown in fig. 2, the tracking statistics of the newly delivered advertisement amount per day and the daily advertisement inventory amount per day from xxxx × year × 11 days to 31 days in a certain channel are shown, wherein each row in the table represents the advertisement inventory amount per day after the advertisement is delivered on the current day. It can be seen that the ad placement data for 11 x-months has tracked 20 days, during which the data change does not follow the monotone decreasing statistical rule, but there is an upward fluctuation on the 6 th and 13 th days, which is very disadvantageous for us to count all the stock change trends on the 11 th day. In addition, in order to know how much daily advertisement investment (activation amount) finally obtains output (how many advertisement inventory is viewed by the activated users in total), we hope to completely count all inventory changes generated by investment of a channel in a certain day, but because a part of good-quality channels have long survival time, the inventory changes which are not negligible when tracking over the whole day are still remained, and the time consumption of only counting the data is too large. But the original data is directly predicted by a mathematical method, and the inventory change fluctuation generated by single-day activation is more obvious, so that the embodiment of the invention carries out data preprocessing by a regular average smoothing mode.
The specific preprocessing is to carry out batch summation and averaging on data of a corresponding date and a certain period after the date to form corresponding period statistical data. The embodiment of the invention specifically adopts the statistical data of the weekly average value of the advertisement stock in a set time interval after the advertisement is put in from a target date as the calculation basis of the prediction function of the advertisement stock. It is to be understood that the advertisement inventory on the nth day in the advertisement inventory weekly average statistical data is an average of advertisement inventory on the nth day after advertisement placement every day in the week from the target date, and N is an integer greater than zero.
Specifically, as shown in fig. 2, in order to count the advertisement inventory quantity weekly average statistical data starting from the day of × month 11, at least the advertisement inventory quantity data for each day within 7 days starting from the day of × month 11 is required. It is understood that the number of days included in the set time interval is at least greater than the number of days required to calculate the mean value statistics from the target date. The following description will be given taking the data in fig. 2 as an example to calculate the advertisement inventory weekly average statistical data starting from day × month 11.
The set time period is 20 days from day × 11 month shown in fig. 2. When calculating the advertisement inventory weekly average statistical data starting from the 11 th day of × month, the advertisement inventory statistical data on the nth day from the 11 th day of × month to the 17 th day of × month in the rectangular frame area in fig. 2 are summed and averaged by columns, and the obtained data is the advertisement inventory on the nth day of the advertisement inventory weekly average statistical data starting from the 11 th day of × month. For example, assuming that the advertisement inventory amount of the first day after 11 days × month is calculated, the data of the first day from 11 days × month to 17 days × month in the column of the first day after 11 days × month in fig. 2 are summed and averaged to obtain the advertisement inventory amount of the first day of the advertisement inventory amount weekly average statistical data starting from 11 days × month; assuming that the advertisement inventory amount of the second day after 11 days of × month is calculated, summing and averaging the data from 11 days of × month to 17 days of × month in a column of the second day after 11 days of × month in fig. 2 to obtain the advertisement inventory amount of the second day of the advertisement inventory amount weekly average statistical data starting from 11 days of × month; by analogy, the advertisement inventory amount of each day in the advertisement inventory amount week average statistical data starting from 11 days × month can be calculated.
In addition, in the calculation of the advertisement inventory quantity week-average statistical data on the 25 th day from the 11 th day of × month in fig. 2, since the above-mentioned set time period takes only the data of 20 days from the 11 th day of × month, that is, only 31 days of × month, and the statistical data on the 25 th day from the 11 th day of × month is less than the data of seven days, the week-average statistical data cannot be calculated, and the advertisement inventory quantity week-average statistical data on the 25 th day from the 11 th day of × month can be calculated only after the advertisement inventory quantity data on the first day of the next month, that is, the advertisement inventory quantity on the 25 th day from the 17 th day of × month is counted again.
According to the calculation method, the statistical data of the weekly average value, the monthly average value, the quarterly average value and the like of the advertisement inventory stock from each day can be calculated. After the mean statistical data are calculated, the calculation result and the original statistical data are stored in the sql database, and when the return on investment of the advertisement in a certain day needs to be predicted, corresponding data are read from the sql database. For example, assuming that the return on investment for advertisement placement for 11 days xxxx × year × month is calculated by using the technical solution of the embodiment of the present invention, advertisement placement amount data for 11 days xxxx × year and advertisement inventory weekly average statistical data for a set time interval after advertisement placement for 11 days xxxx × month, for example, advertisement inventory weekly average statistical data for 20 days after advertisement placement for 11 days xxxx × year, are read from a previously stored sql database.
Specifically, when data is read from the sql database, the data can be read through a json string, wherein a 'time' field of the json string represents a target date and a 'name' field represents a time interval in which the data needs to be read.
S302, calculating to obtain an advertisement inventory prediction function after the advertisement is put on the target date according to the weekly average statistical data of the advertisement inventory;
specifically, after the week-average statistical data of the advertisement inventory quantity of the target date is obtained, the advertisement inventory quantity y of the ith date is obtainediAs a prediction target, for the advertisement inventory yiLinear fitting is carried out on the logarithm of the number of the advertisement inventory and the statistical number of days i, and the slope and the intercept of a fitting function are calculated by adopting a least square method, so that the advertisement inventory y is determinediThe logarithm of (d) and the number of statistical days i, the advertisement inventory y is determinediAnd calculating to obtain the function of predicting the stock quantity of the advertisements after the advertisements are placed on the target date.
The form of the prediction function of the logarithm of the calculated advertisement inventory quantity is as follows:
ln(yi)＝α*i+β
further, the prediction function of the logarithm of the advertisement stock quantity is subjected to form transformation, so that the prediction function of the advertisement stock quantity can be finally determined:
yi＝eα*i+β
s303, obtaining historical advertisement inventory data of the target channel in a set time period before the target date and advertisement inventory weekly average statistical similarity data with the minimum difference with the advertisement inventory weekly average statistical data;
specifically, the advertisement inventory amount history data of the channel in the set time period before the target date refers to advertisement inventory amount statistical data of the channel every day in the set time period before the target date. For example, the advertisement inventory amount history data in three months before 11 days of the channel xxx × year × month refers to advertisement inventory amount statistical data for each day of the channel counted from 90 days before 11 days of the channel xxx × year × month to 11 days of the channel xxx × month.
The acquisition of the historical data can be directly read from the database according to the time interval.
The similar data of the advertisement inventory week mean statistics with the minimum difference with the advertisement inventory week mean statistics refers to all the prediction data yiAdvertisement inventory on day one1And after normalization, selecting the data with the minimum difference between the previous 7 days and the predicted data from the data with the time length of more than 50 days in the statistical data of the weekly average value of the advertisement stock.
In an embodiment of the present invention, the difference between the statistical data and the prediction data can be expressed as:
wherein, yiIndicating the advertisement inventory amount on the ith day predicted by the prediction function obtained in step S202; y isiAdvertisement inventory indicating day i of the statistics; h is the set of the first seven days (H ═ {1,2,3,4,5,6,7}), | H | represents the number of data in which the correspondence is greater than 0.
And calculating the difference between the statistical data and the predicted data according to the formula, namely selecting the data with the minimum difference between the previous 7 days and the predicted data as the advertisement stock weekly average statistical similarity data of the advertisement stock weekly average statistical data.
S304, determining a stable period starting point of the advertisement inventory of the target date according to the advertisement inventory prediction function and the historical data of the advertisement inventory;
specifically, the advertisement inventory prediction function calculated in step S302 is close to the logarithmic curve form, but the advertisement inventory development in the later stage of advertisement delivery is closer to the power curve, and it can be considered that the advertisement inventory change in the earlier stage after advertisement delivery is not steady. Therefore, the embodiment of the invention considers that the advertisement inventory can be accurately predicted after the change of the advertisement inventory enters the stationary phase, and the accurate advertisement inventory prediction function can be obtained only by utilizing the advertisement inventory stationary phase data.
Therefore, in the embodiment of the present invention, after the advertisement inventory prediction function is preliminarily obtained in step S302, the starting point of the advertisement inventory stationary period after the advertisement is delivered in the target date in the channel is further determined according to the preliminarily obtained advertisement inventory prediction function and the advertisement inventory historical data.
Specifically, the embodiment of the present invention calculates the difference between the predicted data and the actual data of the different stationary period starting points of the historical data of the advertisement inventory amount by enumeration, and finds the stationary period starting point with the minimum difference from the predicted data and the actual data, that is, the stationary period starting point of the advertisement inventory amount of the target date.
The difference between the forecast data and the actual data of the historical data of the advertisement inventory at different stationary period starting points can be expressed as:
wherein f (i) represents the advertisement inventory amount on the ith day calculated by using the advertisement inventory amount prediction function obtained in step S302; y isiReal advertisement inventory indicating the ith day; t represents the late stationary phase, which in the present embodiment is {50,60,70,80,90,100}, and | T | represents y in the late stationary phaseiA number of data greater than 0.
According to the formula, the minimum D is calculatedTAnd the corresponding T is the starting point of the stable period of the advertisement inventory.
It should be noted that, in the embodiment of the present invention, the data within 90 days before the target date is used as the advertisement inventory history data to calculate the start point of the advertisement inventory stationary period, and when the embodiment of the present invention is implemented, the time interval of the advertisement inventory history data may be flexibly set according to the data amount. Theoretically, the time interval of the advertisement inventory historical data cannot be less than 60 days, if the time interval of the advertisement inventory historical data of the channel is set to be less than 60 days, the channel is considered to have no credible historical data, at this time, the stationary period starting point can be assigned, for example, the stationary period starting point can be set to 5, namely, the advertisement inventory enters the stationary period from the 5 th day after the advertisement is delivered.
S305, calculating to obtain a first advertisement inventory prediction function after the advertisement is put on the target date according to the data after the stationary period starting point in the advertisement inventory weekly average statistical data;
specifically, after the start point of the stationary period of the advertisement stock amount is determined, the calculation of the advertisement stock amount prediction function is performed again for the data after the start point of the stationary period in the advertisement stock amount weekly average statistical data acquired in step S301. The specific calculation process is the same as the calculation process in step S302, and is not described here again. In order to distinguish from the advertisement inventory quantity prediction function calculated in step S302, the advertisement inventory quantity prediction function calculated in this step is named a first advertisement inventory quantity prediction function.
The form of the first advertisement inventory quantity prediction function is as shown in the formula in step S302:
yi＝eα*i+β
s306, calculating to obtain a second advertisement inventory prediction function according to the historical advertisement inventory data, and calculating to obtain a third advertisement inventory prediction function according to similar data counted by the weekly average of the advertisement inventory;
specifically, after the advertisement inventory historical data and the advertisement inventory weekly average statistical similarity data are obtained, since both the historical data and the advertisement inventory weekly average statistical similarity data have long time longer than 60 days, the corresponding fitting result, that is, the corresponding advertisement inventory prediction function, can be directly obtained by firstly obtaining the logarithm and then performing the least square fitting introduced in step S302. In order to facilitate distinguishing, the advertisement inventory prediction function calculated according to the historical advertisement inventory data is named as a second advertisement inventory prediction function, and the advertisement inventory prediction function calculated according to the weekly average statistical similarity of advertisement inventory data is named as a third advertisement inventory prediction function.
S307, calculating to obtain a correction slope according to the slopes of the second advertisement inventory prediction function and the third advertisement inventory prediction function;
specifically, in the embodiment of the present invention, the slopes of the second advertisement inventory prediction function and the third advertisement inventory prediction function are weighted to obtain a corrected slope.
The specific weighting method is shown in the following formula:
wherein alpha istarCalculating the obtained correction slope; alpha is alphahistAnd alphasimThe slopes of the second advertisement inventory prediction function and the third advertisement inventory prediction function, respectively;
and
the data difference of the stationary period starting point of the second advertisement inventory quantity prediction function and the stationary period starting point of the third advertisement inventory quantity prediction function are respectively.
S308, calculating to obtain a correction intercept according to the week average statistical data of the advertisement inventory;
specifically, the correction slope α is calculated in step S307tarThen, the corrected slope is substituted into the advertisement stock quantity prediction function calculated in step S305 to obtain an advertisement stock quantity prediction function:
then, the advertisement inventory data of any day in the obtained advertisement inventory weekly average statistical data and the day (i.e. the value of i) of the day after the day is the target day are substituted into the advertisement inventory prediction function, and the correction intercept beta is calculatedtar。
S309, correcting the first advertisement stock prediction function by using the correction slope and the correction intercept to obtain a corrected advertisement stock prediction function;
specifically, the correction slope α calculated in the above steps S307 and S308 is usedtarAnd correcting the intercept betatarThe advertisement inventory amount prediction function y obtained in step S305 is replaced with each otheri＝eα*i+βThe slope alpha and the intercept beta in the step (b) are used for obtaining a corrected advertisement inventory prediction function:
s310, determining the daily advertisement inventory after the advertisements are put on the target date according to the advertisement inventory prediction function; the daily advertisement inventory is data obtained by rounding the daily advertisement inventory;
specifically, the function of predicting the advertisement inventory amount after the correction determined in step S309
The method can be substituted into different dates i to calculate the daily advertisement inventory after the advertisements are placed on the target dates. Since the value of the exponential function is never attenuated to 0, which is inconsistent with reality, the embodiment of the present invention rounds the calculated daily advertisement inventory, and when the calculated advertisement inventory is 0, it is considered that the advertisement delivery on the target date does not generate any more revenue.
It should be noted that, since the embodiment of the present invention is an advertisement inventory quantity prediction function determined by advertisement inventory quantity statistical data of a target date in a short period (one week), and then predicts advertisement inventory quantity data of a later period that has not been counted by using the prediction function, advertisement inventory quantity data after the target date that has been counted can be directly used, and advertisement inventory quantity data of a later period that has not occurred can be predicted by the above formula.
S311, calculating the sum of all the advertisement inventory amounts from the time when the advertisement is released on the target date to the time when the advertisement inventory amount is reduced to zero, and taking the sum as the advertisement release income of the target date;
specifically, the daily advertisement inventory after the advertisement is delivered from the target date is determined by step S310 until the advertisement inventory is reduced to 0. And summing the advertisement inventory quantity of each day after the target date (including the statistical advertisement inventory quantity data and the predicted advertisement inventory quantity data), and taking the obtained total quantity of the advertisement inventory quantity as the advertisement putting income of the target date.
S312, calculating the return on investment of the advertisement putting on the target date according to the advertisement putting income on the target date and the advertisement putting amount data on the target date.
Specifically, the return on investment of advertisement delivery on the target date can be obtained by dividing the advertisement delivery income on the target date by the advertisement delivery amount data on the target date.
The embodiment of the invention also discloses a device for predicting the return on investment of the advertisement, which is shown in figure 4 and comprises the following components:
a first data obtaining unit 100, configured to obtain advertisement delivery amount data of a target date of a target channel and advertisement inventory weekly average statistical data in a set time interval after advertisement delivery from the target date; the advertisement inventory on the Nth day in the advertisement inventory weekly average statistical data is the average value of the advertisement inventory on the Nth day after advertisements are put every day in a week starting from the target date, and N is an integer greater than zero;
a function calculating unit 110, configured to calculate, according to the weekly average statistical data of advertisement inventory, an advertisement inventory prediction function after advertisement delivery on the target date;
a profit determination unit 120, configured to determine advertisement delivery profits of the target date according to the advertisement inventory prediction function;
and the return on investment calculation unit 130 is configured to calculate the return on investment of advertisement delivery on the target date according to the advertisement delivery income on the target date and the advertisement delivery amount data on the target date.
Specifically, please refer to the contents of the above method embodiments for the specific working contents of each unit in this embodiment, which are not described herein again.
In another embodiment of the invention, as shown in fig. 5, the apparatus further comprises:
a second data obtaining unit 140, configured to obtain advertisement inventory historical data of the target channel in a set time period before the target date and advertisement inventory weekly average statistical similarity data with a smallest difference from the advertisement inventory weekly average statistical data;
and a function correcting unit 150, configured to correct the advertisement inventory prediction function according to the historical advertisement inventory data and the weekly average statistical similarity data of the advertisement inventory.
Wherein, the function modification unit 150 includes:
the stationary period determining unit is used for determining the starting point of the stationary period of the advertisement inventory quantity of the target date according to the advertisement inventory quantity prediction function and the historical data of the advertisement inventory quantity;
the first calculation unit is used for calculating and obtaining a first advertisement inventory prediction function after the advertisement is delivered on the target date according to the data after the stationary period starting point in the advertisement inventory weekly average statistical data;
the second calculation unit is used for calculating to obtain a correction slope according to the historical advertisement stock data and the weekly average statistical similarity data of the advertisement stock;
the third calculating unit is used for calculating to obtain a correction intercept according to the week average statistical data of the advertisement stock;
and the correction unit is used for correcting the first advertisement stock prediction function by using the correction slope and the correction intercept to obtain a corrected advertisement stock prediction function.
The second calculating unit is specifically configured to, when calculating to obtain a correction slope according to the advertisement inventory historical data and the advertisement inventory weekly average statistical similarity data:
calculating to obtain a second advertisement inventory prediction function according to the historical advertisement inventory data, and calculating to obtain a third advertisement inventory prediction function according to the weekly average statistical similarity data of the advertisement inventory; and calculating to obtain a correction slope according to the slopes of the second advertisement inventory prediction function and the third advertisement inventory prediction function.
Specifically, please refer to the contents of the above method embodiment for the specific working contents of each unit in the above embodiment, which are not described herein again.
In another embodiment of the present invention, when the profit determining unit 120 determines the advertisement putting profit of the target date according to the advertisement inventory prediction function, it is specifically configured to:
determining the daily advertisement inventory after the advertisements are put on the target date according to the advertisement inventory prediction function; the daily advertisement inventory is data obtained by rounding the daily advertisement inventory; and calculating the total inventory of all advertisements from the time when the advertisements are released on the target date to the time when the inventory of the advertisements is reduced to zero, and taking the total inventory of all advertisements as the advertisement release yield of the target date.
Specifically, please refer to the content of the method embodiment for the specific working content of the benefit determining unit 120 in this embodiment, which is not described herein again.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (8)
1. A method for predicting return on investment for advertising, comprising:
acquiring advertisement putting quantity data of a target date of a target channel and advertisement inventory weekly average statistical data in a set time interval after advertisements are put from the target date; the advertisement inventory on the Nth day in the advertisement inventory weekly average statistical data is the average value of the advertisement inventory on the Nth day after advertisements are put every day in a week starting from the target date, and N is an integer greater than zero;
calculating to obtain an advertisement inventory prediction function after the advertisement is put on the target date according to the weekly average statistical data of the advertisement inventory; the advertisement inventory prediction function represents the functional relation between the advertisement inventory and the number of statistical days, wherein the statistical days are counted days after the target date;
determining the advertisement putting income of the target date according to the advertisement inventory prediction function;
calculating the return on investment of the advertisement delivery on the target date by dividing the advertisement delivery income on the target date by the advertisement delivery amount data on the target date;
after calculating and obtaining an advertisement inventory prediction function after the advertisement is delivered on the target date according to the weekly average statistical data of the advertisement inventory, the method further comprises the following steps:
acquiring historical advertisement inventory data of the target channel in a set time period before the target date and advertisement inventory weekly average statistical similarity data with the minimum difference with the advertisement inventory weekly average statistical data;
correcting the advertisement stock prediction function according to the historical advertisement stock data and the similar advertisement stock weekly average statistical data;
wherein the difference is
Wherein, yiThe advertisement inventory quantity of the ith day obtained by adopting the advertisement inventory quantity prediction function is represented; y isiAdvertisement inventory indicating day i of the statistics; h is the set of the first seven days (H ═ {1,2,3,4,5,6,7}), | H | represents the number of data in which the correspondence is greater than 0.
2. The method of claim 1, wherein said modifying said advertisement inventory prediction function based on said advertisement inventory historical data and said advertisement inventory weekly average statistical similarity comprises:
determining the starting point of the stable period of the advertisement inventory on the target date according to the advertisement inventory prediction function and the historical data of the advertisement inventory;
calculating to obtain a first advertisement inventory prediction function after the advertisement is put on the target date according to the data after the stationary period starting point in the advertisement inventory weekly average statistical data;
counting similar data according to the historical data of the advertisement inventory and the weekly average value of the advertisement inventory, and calculating to obtain a correction slope;
calculating to obtain a correction intercept according to the week average statistical data of the advertisement inventory;
and correcting the first advertisement inventory prediction function by using the correction slope and the correction intercept to obtain a corrected advertisement inventory prediction function.
3. The method of claim 2, wherein calculating a correction slope based on the historical advertisement inventory data and the weekly average advertisement inventory statistics similarity comprises:
calculating to obtain a second advertisement inventory prediction function according to the historical advertisement inventory data, and calculating to obtain a third advertisement inventory prediction function according to the weekly average statistical similarity data of the advertisement inventory;
and calculating to obtain a correction slope according to the slopes of the second advertisement inventory prediction function and the third advertisement inventory prediction function.
4. The method of claim 1, wherein determining revenue from placement of advertisements on the target date based on the advertisement inventory prediction function comprises:
determining the daily advertisement inventory after the advertisements are put on the target date according to the advertisement inventory prediction function; the daily advertisement inventory is data obtained by rounding the daily advertisement inventory;
and calculating the total inventory of all advertisements from the time when the advertisements are released on the target date to the time when the inventory of the advertisements is reduced to zero, and taking the total inventory of all advertisements as the advertisement release yield of the target date.
5. An apparatus for predicting return on investment for advertising, comprising:
the system comprises a first data acquisition unit, a second data acquisition unit and a third data acquisition unit, wherein the first data acquisition unit is used for acquiring advertisement putting quantity data of a target date of a target channel and advertisement inventory weekly average statistical data in a set time interval after advertisements are put on the target date; the advertisement inventory on the Nth day in the advertisement inventory weekly average statistical data is the average value of the advertisement inventory on the Nth day after advertisements are put every day in a week starting from the target date, and N is an integer greater than zero;
the function calculation unit is used for calculating and obtaining an advertisement inventory prediction function after the advertisement is put on the target date according to the week-average statistical data of the advertisement inventory; the advertisement inventory prediction function represents the functional relation between the advertisement inventory and the number of statistical days, wherein the statistical days are counted days after the target date;
the profit determining unit is used for determining the advertisement putting profit of the target date according to the advertisement inventory prediction function;
the return on investment rate calculating unit is used for calculating the return on investment rate of the advertisement delivery on the target date by dividing the advertisement delivery income on the target date by the advertisement delivery amount data on the target date;
the device further comprises:
the second data acquisition unit is used for acquiring historical advertisement stock data of the target channel in a set time period before the target date and similar advertisement stock weekly average statistical data with the minimum difference with the advertisement stock weekly average statistical data;
the function correction unit is used for correcting the advertisement stock prediction function according to the historical advertisement stock data and the similar advertisement stock weekly average statistical data;
wherein the difference is
Wherein, yiThe advertisement inventory quantity of the ith day obtained by adopting the advertisement inventory quantity prediction function is represented; y isiRepresenting statistical dataAdvertisement inventory on day i; h is the set of the first seven days (H ═ {1,2,3,4,5,6,7}), | H | represents the number of data in which the correspondence is greater than 0.
6. The apparatus of claim 5, wherein the function modification unit comprises:
the stationary period determining unit is used for determining the starting point of the stationary period of the advertisement inventory quantity of the target date according to the advertisement inventory quantity prediction function and the historical data of the advertisement inventory quantity;
the first calculation unit is used for calculating and obtaining a first advertisement inventory prediction function after the advertisement is delivered on the target date according to the data after the stationary period starting point in the advertisement inventory weekly average statistical data;
the second calculation unit is used for calculating to obtain a correction slope according to the historical advertisement stock data and the weekly average statistical similarity data of the advertisement stock;
the third calculating unit is used for calculating to obtain a correction intercept according to the week average statistical data of the advertisement stock;
and the correction unit is used for correcting the first advertisement stock prediction function by using the correction slope and the correction intercept to obtain a corrected advertisement stock prediction function.
7. The apparatus according to claim 6, wherein the second calculating unit is specifically configured to, when calculating the correction slope according to the advertisement inventory historical data and the advertisement inventory weekly average statistical similarity data:
calculating to obtain a second advertisement inventory prediction function according to the historical advertisement inventory data, and calculating to obtain a third advertisement inventory prediction function according to the weekly average statistical similarity data of the advertisement inventory; and calculating to obtain a correction slope according to the slopes of the second advertisement inventory prediction function and the third advertisement inventory prediction function.
8. The apparatus according to claim 5, wherein the profit determining unit is configured to, when determining the profit of advertisement placement on the target date according to the advertisement inventory prediction function, specifically:
determining the daily advertisement inventory after the advertisements are put on the target date according to the advertisement inventory prediction function; the daily advertisement inventory is data obtained by rounding the daily advertisement inventory; and calculating the total inventory of all advertisements from the time when the advertisements are released on the target date to the time when the inventory of the advertisements is reduced to zero, and taking the total inventory of all advertisements as the advertisement release yield of the target date.
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 CN108564404A (en) | 2018-09-21 |
| CN108564404B true 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) |
Families Citing this family (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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108564404A (en) | 2018-09-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108564404B (en) | Method and device for predicting return on investment of advertisement | |
| CN106960359B (en) | Full-automatic bidding optimization method and system based on streaming calculation | |
| US8473339B1 (en) | Automatically switching between pricing models for services | |
| US20130018722A1 (en) | System and method for generating a keyword bid | |
| CN109657132B (en) | Recommended information cost control method, recommended information cost control device, computer equipment and storage medium | |
| US20140207564A1 (en) | System and method for serving electronic content | |
| US20110231239A1 (en) | Method and system for attributing an online conversion to multiple influencers | |
| US20110208596A1 (en) | Method and system for providing advertising in which the bid price per unit time is adjusted in accordance with advertising traffic | |
| CN106296247B (en) | Network information resource online ordering method and device | |
| CN111162925B (en) | Network flow prediction method and device, electronic equipment and storage medium | |
| CN109636477B (en) | Early warning method and device for advertisement delivery shortage | |
| WO2008016591A2 (en) | System and method for scheduling online keyword auctions subject to budget constraints | |
| CN111899041A (en) | Information delivery processing method, information delivery device, information delivery equipment and storage medium | |
| CN109087136B (en) | Method and device for adjusting advertisement resource value | |
| CN110880127A (en) | Consumption level prediction method and device, electronic equipment and storage medium | |
| JP5904987B2 (en) | Advertising device, prediction method, and prediction program | |
| US20220108336A1 (en) | Methods and apparatus to determine reach with time dependent weights | |
| CN117011031A (en) | Intelligent financial risk monitoring and early warning platform using big data information processing technology | |
| CN112070564A (en) | Advertisement pulling method, device and system and electronic equipment | |
| CN110245969B (en) | Promotion information processing method and device, electronic equipment and computer readable medium | |
| JP2012504270A (en) | Method and system for managing quality of advertising documents | |
| CN109740060B (en) | Frequency control method and device for information pushing | |
| US20160086226A1 (en) | Extraction apparatus, extraction method, and non-transitory computer readable storage medium | |
| CN117853167A (en) | Advertisement putting effect measuring method and system | |
| CN116976998A (en) | Method and device for determining bid strategy of effect advertisement putting |
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 |