CN111260414A - Advertisement inventory prediction method and device - Google Patents
Advertisement inventory prediction method and device Download PDFInfo
- Publication number
- CN111260414A CN111260414A CN202010076323.4A CN202010076323A CN111260414A CN 111260414 A CN111260414 A CN 111260414A CN 202010076323 A CN202010076323 A CN 202010076323A CN 111260414 A CN111260414 A CN 111260414A
- Authority
- CN
- China
- Prior art keywords
- advertisement
- time period
- inventory
- amount
- effective
- 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
- 238000000034 method Methods 0.000 title claims abstract description 99
- 238000007781 pre-processing Methods 0.000 claims abstract description 12
- 230000008569 process Effects 0.000 claims description 47
- 238000012549 training Methods 0.000 claims description 26
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 10
- 238000013277 forecasting method Methods 0.000 claims description 2
- 230000002159 abnormal effect Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000009499 grossing Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000000737 periodic effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 230000003442 weekly effect Effects 0.000 description 1
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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0202—Market predictions or forecasting for commercial activities
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Finance (AREA)
- Economics (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Data Mining & Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a device for predicting advertisement inventory, wherein the method comprises the following steps: when the advertisement inventory of the target advertisement space in a future time period needs to be predicted, determining the total advertisement putting amount of the target advertisement space in a historical time period; preprocessing each advertisement in the historical time period to obtain the total effective advertisement putting amount of the target advertisement position in the historical time period; adjusting the advertisement inventory model according to the configuration file and the total effective advertisement putting amount to obtain an adjusted advertisement inventory model; setting a first time parameter corresponding to a future time period, inputting the first time parameter, a configuration file and a total effective advertisement putting amount into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a first predicted value corresponding to a target advertisement space, and determining the first predicted value as an advertisement inventory corresponding to the target advertisement space. By applying the method provided by the invention, the advertisement inventory in the future time period can be accurately predicted, and the advertisement putting quantity is reasonably arranged.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a device for predicting advertisement inventory.
Background
With the rapid development of internet technology, more and more advertisers can place advertisements on various advertisement platforms. Before advertisement putting, an advertiser signs contract advertisements with an advertisement platform and specifies the number of the signed advertisements consumed by the advertisement platform in a preset time, namely, the advertisement platform needs to put the advertisements specified by the advertiser according to the number of the advertisements in a preset time period according to the content of the contract advertisements.
In the prior art, an advertisement platform predicts advertisement inventory in a future period of time according to historical advertisement delivery quantity, namely the delivery quantity of advertisements in the future period of time, so as to sign contract advertisements with advertisers according to the predicted advertisement inventory. However, due to changes in time, period and other unstable factors, the predicted advertisement inventory is often inaccurate, too high predicted advertisement inventory can cause the number of signed advertisements to be delivered to the user and too low predicted advertisement inventory can cause the advertisement shortage and affect the advertisement revenue of the advertisement platform, so that the advertisement inventory in a future period of time cannot be accurately predicted only according to the historical advertisement delivery number.
Disclosure of Invention
In view of this, the present invention provides a method for predicting advertisement inventory, by which advertisement inventory in a future time period can be accurately predicted, and advertisement delivery quantity can be reasonably arranged.
The invention also provides an advertisement inventory prediction device, which is used for ensuring the realization and the application of the method in practice.
An advertising inventory forecasting method, comprising:
when the advertisement inventory of a target advertisement position in a pre-specified future time period needs to be predicted, determining the total advertisement putting amount of the target advertisement position in a preset historical time period, wherein the historical time period comprises a plurality of time intervals, and the total advertisement putting amount is the sum of the advertisement putting amounts corresponding to the time intervals;
preprocessing each released advertisement in the historical time period to obtain the total effective advertisement release amount of the target advertisement space in the historical time period;
acquiring a preset configuration file, and adjusting a preset advertisement inventory model according to the configuration file and the total effective advertisement putting quantity to obtain an adjusted advertisement inventory model, wherein the configuration file comprises date parameters corresponding to each holiday date;
setting a first time parameter corresponding to the future time period, inputting the first time parameter, the configuration file and the total effective advertisement putting amount into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a first predicted value corresponding to the target advertisement position, and determining the first predicted value as the advertisement inventory corresponding to the target advertisement position in the future time period.
Optionally, the foregoing method, where preprocessing is performed on each delivered advertisement in the historical time period to obtain a total effective advertisement delivery amount of the target advertisement slot in the historical time period, includes:
acquiring traffic information of each released advertisement in each time interval, and judging whether an abnormally released advertisement exists or not according to the traffic information of each released advertisement;
if the abnormally-launched advertisements exist, rejecting each abnormally-launched advertisement in each time interval, and determining an effective advertisement inventory corresponding to each time interval;
and summing the effective advertisement inventory corresponding to each time interval to obtain the total effective advertisement putting amount of the target advertisement in the historical time period.
The above method, optionally, further includes:
checking the effective advertisement putting quantity corresponding to each time interval, and judging whether the effective advertisement putting quantity corresponding to the current time interval is empty or not;
if the effective advertisement putting amount corresponding to the current time interval is empty, determining the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval;
and calculating the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval to obtain an average effective advertisement putting amount, and determining the average effective advertisement putting amount as the effective advertisement putting amount corresponding to the current time interval.
Optionally, the method for adjusting a preset advertisement inventory model according to the configuration file and the total effective advertisement placement amount includes:
dividing the historical time period into a first time period and a second time period, and dividing the total effective advertisement putting quantity according to the first time period and the second time period to obtain a first putting quantity and a second putting quantity, wherein the first putting quantity is the quantity of advertisements put at the advertisement slot in the first time period, and the second putting quantity is the quantity of advertisements put at the target advertisement slot in the second time period;
setting a second time parameter corresponding to the second time period, and executing a preset training process; the training process comprises: inputting the configuration file, the second time parameter and the first input amount into the advertisement inventory model, triggering the advertisement inventory model to output a second predicted value corresponding to the second time period, and calculating the second predicted value and the second input amount to obtain an error value;
judging whether the error value is larger than a preset minimum threshold value or not;
if the error value is larger than a preset minimum threshold value, adjusting the model parameters of the advertisement inventory model according to the error value, and executing the training process until the error value is not larger than the minimum threshold value, and finishing adjusting the advertisement inventory model.
The above method, optionally, further includes:
carrying out logarithmic calculation on the effective advertisement putting quantity corresponding to each time interval according to a preset logarithmic formula to obtain each logarithmic value corresponding to the target advertisement space;
inputting the first time parameter, the configuration file and each logarithmic value into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a third predicted value corresponding to the target advertisement position, performing inverse operation on the third predicted value according to the logarithmic formula to obtain an operation value, and determining the operation value as the advertisement inventory corresponding to the target advertisement position in the future time period.
An advertising inventory forecasting apparatus, comprising:
the system comprises a determining unit, a calculating unit and a calculating unit, wherein the determining unit is used for determining the total advertisement putting quantity of a target advertisement position in a preset historical time period when the advertisement inventory of the target advertisement position in a pre-specified future time period needs to be predicted, the historical time period comprises a plurality of time intervals, and the total advertisement putting quantity is the sum of the advertisement putting quantities corresponding to the time intervals;
the processing unit is used for preprocessing each released advertisement in the historical time period to obtain the total effective advertisement release amount of the target advertisement space in the historical time period;
the adjusting unit is used for acquiring a preset configuration file, adjusting a preset advertisement inventory model according to the configuration file and the total effective advertisement putting quantity, and acquiring an adjusted advertisement inventory model, wherein the configuration file comprises date parameters corresponding to each holiday date;
and the prediction unit is used for setting a first time parameter corresponding to the future time period, inputting the first time parameter, the configuration file and the total effective advertisement putting amount into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a first prediction value corresponding to the target advertisement space, and determining the first prediction value as the advertisement inventory corresponding to the target advertisement space in the future time period.
The above apparatus, optionally, the processing unit includes:
the judging subunit is used for acquiring traffic information of each released advertisement in each time interval and judging whether an abnormally released advertisement exists or not according to the traffic information of each released advertisement;
the determining subunit is used for eliminating each abnormally-launched advertisement in each time interval if the abnormally-launched advertisement exists, and determining the effective advertisement inventory corresponding to each time interval;
and the calculating subunit is used for summing the effective advertisement inventory corresponding to each time interval to obtain the total effective advertisement putting amount of the target advertisement in the historical time period.
The above apparatus, optionally, further comprises:
the processing subunit is used for checking the effective advertisement putting quantity corresponding to each time interval and judging whether the effective advertisement putting quantity corresponding to the current time interval is empty or not; if the effective advertisement putting amount corresponding to the current time interval is empty, determining the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval; and calculating the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval to obtain an average effective advertisement putting amount, and determining the average effective advertisement putting amount as the effective advertisement putting amount corresponding to the current time interval.
The above apparatus, optionally, the adjusting unit includes:
a dividing subunit, configured to divide the historical time period into a first time period and a second time period, and divide the total effective advertisement placement amount according to the first time period and the second time period to obtain a first placement amount and a second placement amount, where the first placement amount is a quantity of advertisements placed at the advertisement slot in the first time period, and the second placement amount is a quantity of advertisements placed at the target advertisement slot in the second time period;
the training subunit is used for setting a second time parameter corresponding to the second time period and executing a preset training process; the training process comprises: inputting the configuration file, the second time parameter and the first input amount into the advertisement inventory model, triggering the advertisement inventory model to output a second predicted value corresponding to the second time period, and calculating the second predicted value and the second input amount to obtain an error value;
the judging subunit is used for judging whether the error value is larger than a preset minimum threshold value;
and the adjusting subunit is used for adjusting the model parameters of the advertisement inventory model according to the error value if the error value is greater than a preset minimum threshold value, and executing the training process until the error value is not greater than the minimum threshold value, and finishing adjusting the advertisement inventory model.
The above apparatus, optionally, further comprises:
the calculation unit is used for carrying out logarithmic calculation on the effective advertisement putting amount corresponding to each time interval according to a preset logarithmic formula to obtain each logarithmic value corresponding to the target advertisement space; inputting the first time parameter, the configuration file and each logarithmic value into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a third predicted value corresponding to the target advertisement position, performing inverse operation on the third predicted value according to the logarithmic formula to obtain an operation value, and determining the operation value as the advertisement inventory corresponding to the target advertisement position in the future time period.
A storage medium comprising stored instructions, wherein the instructions, when executed, control a device in which the storage medium is located to perform the above-mentioned advertisement inventory prediction method.
An electronic device comprising a memory, and one or more instructions, wherein the one or more instructions are stored in the memory and configured to be executed by one or more processors to perform the method of advertisement inventory prediction described above.
Compared with the prior art, the invention has the following advantages:
the invention provides an advertisement inventory prediction method, which comprises the following steps: when the advertisement inventory of a target advertisement position in a pre-specified future time period needs to be predicted, determining the total advertisement putting amount of the target advertisement position in a preset historical time period, wherein the historical time period comprises a plurality of time intervals, and the total advertisement putting amount is the sum of the advertisement putting amounts corresponding to the time intervals; preprocessing each released advertisement in the historical time period to obtain the total effective advertisement release amount of the target advertisement space in the historical time period; acquiring a preset configuration file, and adjusting a preset advertisement inventory model according to the configuration file and the total effective advertisement putting quantity to obtain an adjusted advertisement inventory model, wherein the configuration file comprises date parameters corresponding to each holiday date; setting a first time parameter corresponding to the future time period, inputting the first time parameter, the configuration file and the total effective advertisement putting amount into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a first predicted value corresponding to the target advertisement position, and determining the first predicted value as the advertisement inventory corresponding to the target advertisement position in the future time period. By applying the method provided by the invention, the advertisement inventory in the future time period can be accurately predicted, and the advertisement putting quantity is reasonably arranged.
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 flowchart of a method for predicting advertisement inventory according to an embodiment of the present invention;
FIG. 2 is a diagram illustrating an exemplary method for predicting advertisement inventory according to an embodiment of the present invention;
FIG. 3 is a diagram illustrating an example of a method for predicting advertisement inventory according to an embodiment of the present invention;
FIG. 4 is a block diagram of an advertisement inventory forecasting device according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of an electronic device 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.
In this application, relational terms such as first and second, and the like are 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, and 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 invention is operational with numerous general purpose or special purpose computing device environments or configurations. For example: personal computers, server computers, hand-held or portable devices, tablet-type devices, multi-processor apparatus, distributed computing environments that include any of the above devices or equipment, and the like.
The embodiment of the invention provides an advertisement inventory prediction method, which can be applied to various system platforms, wherein an execution subject of the method can be a computer terminal or a processor of various mobile devices, and a flow chart of the method is shown in fig. 1, and the method specifically comprises the following steps:
s101: when the advertisement inventory of a target advertisement position in a pre-specified future time period needs to be predicted, determining the total advertisement putting amount of the target advertisement position in a preset historical time period;
the historical time period comprises a plurality of time intervals, and the total advertisement putting amount is the sum of the advertisement putting amounts corresponding to the time intervals.
In the embodiment of the invention, a plurality of advertisement positions are included in the advertisement platform, such as pre-patch advertisements, spot advertisements, footmark advertisements and the like. When it is desired to predict the advertisement inventory of an advertising platform over a future time period, the advertisement inventory of each ad slot needs to be determined. The advertisement placement amount is the number of times the advertisement is shown on the target advertisement slot. And determining the advertisement space of the advertisement inventory to be predicted as a target advertisement space, and determining the total advertisement putting amount of the target advertisement space in the historical time period.
For example, if the historical time period is one year and each time interval is one day, the total advertisement placement amount of the target advertisement slot in one year time is determined according to the advertisement placement amount of each day in one year.
S102: preprocessing each released advertisement in the historical time period to obtain the total effective advertisement release amount of the target advertisement space in the historical time period;
in the embodiment of the present invention, a plurality of advertisements are placed on the target advertisement space in the historical time period, but the advertisement placed at each time of the target advertisement space is not really effective, and each placed advertisement on the target advertisement space needs to be processed to obtain the total effective advertisement placement amount of the target advertisement space in the historical time period, that is, the real advertisement number of the advertisement platform placed on the target advertisement space in the historical time period.
Specifically, the preprocessing of each delivered advertisement may be to remove the abnormally delivered advertisement or add some unregistered advertisement data.
S103: acquiring a preset configuration file, and adjusting a preset advertisement inventory model according to the configuration file and the total effective advertisement putting quantity to obtain an adjusted advertisement inventory model;
the configuration file comprises date parameters corresponding to each holiday date.
In the embodiment of the invention, the advertisement inventory model may actually be a time series model which supports a Prophet time series algorithm, and the Prophet algorithm may decompose the time series into a trend term and a period term, and simultaneously considers the influence of holidays and potential mutation quantities. In order to ensure the accuracy of the advertisement inventory model in predicting the advertisement inventory, the advertisement inventory model needs to be adjusted in advance. The advertisement inventory model is adjusted through a configuration file and the total effective advertisement putting amount, and because the configuration file comprises date parameters corresponding to a plurality of holiday dates, model parameters in the advertisement inventory model can be fitted through the date parameters.
Specifically, the relevant date parameters of each holiday in one year are written in the configuration file in advance, and the advertisement library can adjust the output prediction result according to the holiday in the adjustment process, so that the distortion of the prediction value caused by the holiday is avoided. The configuration file may include a date parameter corresponding to a holiday date, and may also include a date parameter of a common workday.
Optionally, in the process of adjusting the advertisement inventory, the error value obtained by the orthogonal verification may be used to adjust the model parameters of the advertisement inventory model.
S104: setting a first time parameter corresponding to the future time period, inputting the first time parameter, the configuration file and the total effective advertisement putting amount into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a first predicted value corresponding to the target advertisement position, and determining the first predicted value as the advertisement inventory corresponding to the target advertisement position in the future time period.
In the embodiment of the invention, when the advertisement inventory in a certain future time period needs to be predicted, a first time parameter corresponding to the future time period is set, the first time parameter, a configuration file and the total effective advertisement putting amount are input into the advertisement inventory model, after the advertisement inventory model receives the first time parameter, the advertisement inventory model is adjusted according to the date parameter in the configuration file, so that the advertisement inventory model can be fitted with each date parameter according to the first time parameter, the trend of the advertisement inventory corresponding to the future time period is adjusted, a first predicted value corresponding to a target advertisement position is output according to the change of time and period, and the first predicted value is determined as the advertisement inventory corresponding to the target advertisement position in the future time period.
For example, when the advertisement inventory of the previous patch advertisement in the next month needs to be predicted, after the advertisement inventory model is adjusted, the total effective advertisement input amount in the year ahead is input into the advertisement inventory model, the configuration file containing the date parameter of each holiday and the time parameter corresponding to the next month are input into the advertisement inventory model, the advertisement inventory model outputs a predicted value according to the total effective advertisement input amount, the configuration file and the time parameter corresponding to the next month, and the predicted value predicts the advertisement inventory of the previous patch advertisement in the next month for the advertisement inventory model.
In the method for predicting advertisement inventory, when an advertisement library of a target advertisement position in a future time period needs to be predicted, the historical total advertisement putting amount of the target advertisement position, namely the total amount of advertisements put in the historical time period of the target advertisement position, is determined, each put advertisement is preprocessed, the abnormally put advertisements are eliminated, and the total amount of the target advertisements actually put in the historical time period, namely the total effective advertisement putting amount, is obtained. The method comprises the steps of obtaining a preset configuration file according to each holiday in one year, setting a plurality of date parameters in the configuration file, adjusting an advertisement inventory model according to the date parameters and the total effective advertisement putting amount in a historical time period to obtain an adjusted advertisement inventory model, and predicting the advertisement inventory of a target advertisement space in a future time period by the adjusted advertisement inventory model.
It should be noted that the advertisement inventory model is provided with a Prophet algorithm, which can decompose the time series into a trend term and a period term, and simultaneously considers the influence of holidays and potential mutation quantity. The calculation process of the algorithm is mainly realized by the following formula:
y(t)＝g(t)+s(t)+h(t)+et
specifically, g (t) is used for fitting non-periodic changes such as piecewise linear growth or logic growth in a time series; s (t) represents a periodic variation, such as: seasonal weekly/annually; h (t) represents the irregular holiday effect; et denotes that the error term is used to reflect the abnormal variation not represented in the model.
By applying the method provided by the embodiment of the invention, the advertisement inventory of the target advertisement position in the future time period is predicted by combining the advertisement inventory model and the Prophet algorithm, so that the accuracy of the pre-stored advertisement inventory is ensured, and the advertisement delivery quantity is reasonably arranged.
It should be noted that, in the method provided by the present invention, in addition to predicting the advertisement inventory of the target ad spot in the future time period, the advertisement inventory of the whole advertisement platform may also be predicted, or the advertisement inventory may be classified into dimensions, for example, the advertisement inventory of each region is predicted.
In the method provided by the embodiment of the present invention, based on the step S102 in the above embodiment, even after the total advertisement placement amount of the target advertisement space in the historical time period is determined, in the process of placing the advertisement in the historical time period, there may be an abnormally placed advertisement, so each placed advertisement in the historical time period needs to be preprocessed to obtain the total effective advertisement placement amount of the target advertisement space in the historical time period. The process of preprocessing each delivered advertisement to obtain the total effective advertisement delivery amount is shown in fig. 2, and may specifically include:
s201: acquiring traffic information of each released advertisement in each time interval, and judging whether an abnormally released advertisement exists or not according to the traffic information of each released advertisement;
in the implementation of the invention, the abnormally-delivered advertisement represents the abnormal exposure of the advertisement in the display process, and the traffic information comprises the advertisement traffic generated by the corresponding advertisement in the delivery process. In the process of delivering advertisements on the advertisement platform, there may be an abnormal delivery process, for example, the click rate of a certain advertisement suddenly increases, and the corresponding advertisement traffic also suddenly rises, so that, according to the traffic information of each delivered advertisement, it is determined whether there is an abnormally delivered advertisement, that is, it is determined whether there is any delivered advertisement and the corresponding advertisement traffic is abnormal.
S202: if the abnormally-launched advertisements exist, rejecting each abnormally-launched advertisement in each time interval, and determining an effective advertisement inventory corresponding to each time interval;
in the embodiment of the invention, for any time interval, if the abnormally-launched advertisement exists, the abnormally-launched advertisement is removed, and the effective advertisement inventory corresponding to the time interval is obtained.
S203: and summing the effective advertisement inventory corresponding to each time interval to obtain the total effective advertisement putting amount of the target advertisement in the historical time period.
In the embodiment of the invention, after the effective advertisement inventory of each time interval is determined, the total effective advertisement inventory is obtained by summing the effective advertisement inventory.
In the advertisement inventory prediction method provided by the embodiment of the invention, in order to prevent the error of the total advertisement putting quantity in the historical time period caused by the abnormal putting advertisements, the flow information of each put advertisement in each time interval is obtained, and whether the abnormal putting advertisements exist in each put advertisement is judged through the flow information. And if the abnormally-launched advertisements exist, rejecting the abnormally-launched advertisements, updating the advertisement launching amount corresponding to each time interval, and obtaining the effective advertisement launching amount corresponding to each time interval. And summing the effective advertisement putting quantities to obtain the total effective advertisement putting quantity of the target advertisement space in the historical time period. The method provided by the invention can remove abnormal advertisements and increase the accuracy of predicting the advertisement inventory.
Specifically, after obtaining the effective advertisement putting amount corresponding to each time interval, the effective advertisement putting amount of each time interval can be further processed, and the specific process further includes:
checking the effective advertisement putting quantity corresponding to each time interval, and judging whether the effective advertisement putting quantity corresponding to the current time interval is empty or not;
if the effective advertisement putting amount corresponding to the current time interval is empty, determining the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval;
and calculating the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval to obtain an average effective advertisement putting amount, and determining the average effective advertisement putting amount as the effective advertisement putting amount corresponding to the current time interval.
In the advertisement inventory prediction method provided by the embodiment of the invention, in the process of delivering the advertisement on the target advertisement position, the times of the advertisement displayed by the target advertisement position can be counted through equipment such as a server, and if the advertisement delivery amount in a certain time interval is not counted due to network problems and the like, the advertisement delivery amount in the time interval needs to be supplemented. And after obtaining the effective advertisement putting amount corresponding to each time interval, checking the effective advertisement putting amount of each time interval, and judging whether the currently checked effective advertisement putting amount corresponding to the current time interval is empty, namely whether relevant data of the advertisement put in the current time interval is lost. If the effective advertisement putting amount corresponding to the current time interval is empty, the relevant data of advertisement putting in the current time interval can be determined to be missing. Current ad placement amounts need to be supplemented. Specifically, the average value of the effective advertisement putting quantities corresponding to the previous time interval and the next time interval can be obtained, and the effective advertisement putting quantity corresponding to the current time interval is obtained.
For example, if the historical time period is one month and each time interval is 0:00 to 24:00 of each day in one month, whether the effective advertisement putting amount of each day is zero is sequentially checked, if the effective advertisement putting amount of the day is zero, the effective advertisement putting amounts of the two days before and after are averaged, and the average value is used as the effective advertisement putting amount of the day.
Optionally, in addition to obtaining the effective advertisement placement amount corresponding to the current time interval according to the average value of the effective advertisement placement amounts corresponding to the previous time interval and the next time interval, the effective advertisement placement amount corresponding to the time interval in which the date is consistent in the pre-stored history record may be obtained according to the date corresponding to the current time interval, and the obtained effective advertisement placement amount is used as the effective advertisement placement amount corresponding to the current time interval.
By applying the method provided by the embodiment of the invention, the lost advertisement putting amount is supplemented, so that the accuracy of predicting the advertisement inventory is ensured when the advertisement inventory model is applied to predict the advertisement inventory.
In the method provided by the embodiment of the invention, before predicting the advertisement inventory of the target advertisement space, the advertisement inventory model needs to be adjusted first to ensure the accuracy of the advertisement inventory model in the prediction process. Specifically, as shown in fig. 3, the process of adjusting the advertisement inventory according to the configuration file and the total effective advertisement placement amount specifically includes:
s301: dividing the historical time period into a first time period and a second time period, and dividing the total effective advertisement putting quantity according to the first time period and the second time period to obtain a first putting quantity and a second putting quantity;
the first putting quantity is the quantity of the advertisements put in the advertisement positions in the first time period, and the second putting quantity is the quantity of the advertisements put in the target advertisement positions in the second time period.
In the embodiment of the invention, in the process of adjusting the advertisement inventory model, the advertisement inventory model needs to be trained, wherein the total effective advertisement putting amount is divided into two parts, and the specific division mode is divided according to the first time period and the second time period. The historical time period is divided into two time periods which are respectively a first time period and a second time period, and then the total effective advertisement putting amount is divided into a first putting amount and a second putting amount according to the first time period and the second time period.
For example, if the historical time period is one year, the first time period may be the first nine months, the second time period may be the last three months, the first placement amount is the total amount of ads placed in the target ad spots in the first nine months, and the second placement amount is the total amount of ads placed in the target ad spots in the last three months.
S302: setting a second time parameter corresponding to the second time period;
in the embodiment of the present invention, in order to enable the advertisement inventory model to output the predicted value corresponding to the second time period, a time parameter corresponding to the second time period needs to be set, so that the advertisement inventory model only outputs the predicted value corresponding to the second time period through the second time parameter.
S303: executing a preset training process;
wherein the training process comprises: inputting the configuration file, the second time parameter and the first input amount into the advertisement inventory model, triggering the advertisement inventory model to output a second predicted value corresponding to the second time period, and calculating the second predicted value and the second input amount to obtain an error value;
in the embodiment of the invention, the training of the advertisement inventory model is a multi-iteration process, and each training needs to be carried out according to the training process.
S304: judging whether the error value is larger than a preset minimum threshold value or not;
in the embodiment of the invention, after the training process is executed each time, the second predicted value and the second input amount which are output each time are calculated to obtain the error value, and whether the error value is larger than the preset minimum threshold value is judged.
Specifically, the calculation process of the error value is as follows: the error value is 1- (second test value/second shot size).
For example, if the actual delivery amount in the last three months is 10 hundred million, the second predicted value output by the advertisement inventory model is 9.5 million, and the error value is 0.05.
Alternatively, the error may be calculated by calculating the mean variance.
S305: if the error value is greater than the preset minimum threshold value, adjusting the model parameters of the advertisement inventory model according to the error value, and executing the step S303;
in the embodiment of the present invention, if the error value is not greater than the minimum threshold, the model parameters of the advertisement inventory model are adjusted, and the training process of the step S303 is executed again.
Specifically, the model parameter may be an inflection point smoothing factor changepoint _ prior _ scale. The change _ prior _ scale can be used to adjust the trend predicted for the advertisement inventory in the advertisement inventory model, preventing the change in trend from being over-fit or under-fit.
S306: and if the error value is not larger than the minimum threshold value, finishing adjusting the advertisement inventory model.
In the embodiment of the invention, when the error value is not greater than the minimum threshold value, the prediction capability of the advertisement inventory model is characterized to be nearly maximized, the adjustment of the advertisement inventory model is finished, the adjusted advertisement inventory model is obtained, and the advertisement inventory model can be applied to predict the advertisement inventory of the target advertisement position.
In the advertisement inventory prediction method provided by the embodiment of the invention, before predicting the advertisement inventory of the target advertisement space, the advertisement inventory model needs to be adjusted to obtain the adjusted advertisement inventory model. In the process of adjusting the advertisement inventory model, the total effective advertisement putting amount corresponding to the historical time period needs to be divided into two parts, one part is used for training the advertisement inventory model, the other part is used for solving an error value of the trained advertisement inventory model, model parameters of the advertisement inventory model are adjusted according to the error value, and when the solved error value meets a minimum threshold condition, the adjusted advertisement inventory model can be obtained.
By applying the method provided by the embodiment of the invention, the advertisement inventory model is adjusted, and the accuracy of predicting the advertisement inventory of the target advertisement space is improved.
Based on the method provided by the above embodiment, in the process of predicting the advertisement inventory of the target advertisement space, the prediction needs to be performed through an advertisement inventory model. Specifically, the process of predicting the advertisement inventory of the target advertisement space may further include:
carrying out logarithmic calculation on the effective advertisement putting quantity corresponding to each time interval according to a preset logarithmic formula to obtain each logarithmic value corresponding to the target advertisement space;
inputting the first time parameter, the configuration file and each logarithmic value into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a third predicted value corresponding to the target advertisement position, performing inverse operation on the third predicted value according to the logarithmic formula to obtain an operation value, and determining the operation value as the advertisement inventory corresponding to the target advertisement position in the future time period.
In the method provided by the embodiment of the invention, the total effective advertisement putting quantity corresponding to the historical time period comprises the effective advertisement putting quantity of each time interval. Because the effective advertisement putting amount corresponding to each time interval is large, it can be known through the variance formula that the time sequence corresponding to the advertisement inventory corresponding to each time interval is not stable, that is, the advertisement putting amount corresponding to each time interval is not stable, so that the total effective advertisement putting amount needs to be smoothed. The specific smoothing process comprises the following steps: and carrying out logarithmic calculation on each effective advertisement putting quantity according to a logarithmic formula to obtain a logarithmic value of each effective advertisement putting quantity, wherein the logarithmic formula can be lg (x), and can also be log (x) and the like. After the logarithmic calculation is performed, the variance corresponding to each time interval may be made constant. For example, the effective advertisement putting amount corresponding to january is 10 hundred million, the effective advertisement putting amount corresponding to february is 9 hundred million, the advertisement putting amount difference between two months is large, if a logarithm value is obtained according to a formula lg (x), the logarithm value corresponding to january is 9.00, and the logarithm value of february is 8.95, so that the fluctuation change between two months is small and tends to a gentle state. After obtaining each logarithm value, inputting the first time parameter, the configuration file and each logarithm value into the adjusted advertisement inventory model, outputting a third predicted value by the advertisement inventory model according to each input data, carrying out inverse operation on the third predicted value according to an original logarithmic formula to obtain an operation value, and taking the operation value as the advertisement inventory of the target advertisement position in the future time period.
Optionally, in the process of adjusting the advertisement inventory model, the advertisement inventory model may be adjusted through the respective pair numbers after the logarithm of the first and second delivery quantities respectively corresponding to the first and second time periods is measured.
By applying the method provided by the embodiment of the invention, the fluctuation of the advertisement putting amount among all time intervals can be reduced in a logarithmic conversion mode, and the prediction process of the advertisement inventory model on the target advertisement space is accelerated.
The specific implementation procedures and derivatives thereof of the above embodiments are within the scope of the present invention.
Corresponding to the method described in fig. 1, an embodiment of the present invention further provides an advertisement inventory prediction apparatus, which is used for implementing the method in fig. 1 in detail, the advertisement inventory prediction apparatus provided in the embodiment of the present invention may be applied to a computer terminal or various mobile devices, and a schematic structural diagram of the apparatus is shown in fig. 4, and specifically includes:
a determining unit 401, configured to determine, when it is required to predict an advertisement inventory of a target advertisement spot in a pre-specified future time period, a total advertisement placement amount of the target advertisement spot in a preset historical time period, where the historical time period includes a plurality of time intervals, and the total advertisement placement amount is a sum of advertisement placement amounts corresponding to the time intervals;
a processing unit 402, configured to pre-process each delivered advertisement in the historical time period, and obtain a total effective advertisement delivery amount of the target advertisement slot in the historical time period;
an adjusting unit 403, configured to obtain a preset configuration file, and adjust a preset advertisement inventory model according to the configuration file and the total effective advertisement placement amount to obtain an adjusted advertisement inventory model, where the configuration file includes date parameters corresponding to each holiday date;
a predicting unit 404, configured to set a first time parameter corresponding to the future time period, input the first time parameter, the configuration file, and the total effective advertisement placement amount into the adjusted advertisement inventory model, trigger the adjusted advertisement inventory model to output a first predicted value corresponding to the target advertisement slot, and determine the first predicted value as an advertisement inventory corresponding to the target advertisement slot in the future time period.
In the device provided by the embodiment of the invention, when the advertisement inventory of the target advertisement space in the future time slot needs to be predicted, the determining unit determines the historical total advertisement putting quantity of the target advertisement space, and the processing unit preprocesses each put advertisement to obtain the total effective advertisement putting quantity. After the advertisement inventory model is adjusted through the adjusting unit, the adjusting unit finally predicts the advertisement inventory of the target advertisement position in the future time period through triggering the adjusted advertisement inventory model.
By applying the device provided by the embodiment of the invention, the advertisement inventory in the future time period can be accurately predicted, and the advertisement putting quantity is reasonably arranged.
In the apparatus provided in the embodiment of the present invention, the processing unit 402 includes:
the judging subunit is used for acquiring traffic information of each released advertisement in each time interval and judging whether an abnormally released advertisement exists or not according to the traffic information of each released advertisement;
the determining subunit is used for eliminating each abnormally-launched advertisement in each time interval if the abnormally-launched advertisement exists, and determining the effective advertisement inventory corresponding to each time interval;
and the calculating subunit is used for summing the effective advertisement inventory corresponding to each time interval to obtain the total effective advertisement putting amount of the target advertisement in the historical time period.
The device provided by the embodiment of the invention further comprises:
the processing subunit is used for checking the effective advertisement putting quantity corresponding to each time interval and judging whether the effective advertisement putting quantity corresponding to the current time interval is empty or not; if the effective advertisement putting amount corresponding to the current time interval is empty, determining the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval; and calculating the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval to obtain an average effective advertisement putting amount, and determining the average effective advertisement putting amount as the effective advertisement putting amount corresponding to the current time interval.
In the apparatus provided in the embodiment of the present invention, the adjusting unit 403 includes:
a dividing subunit, configured to divide the historical time period into a first time period and a second time period, and divide the total effective advertisement placement amount according to the first time period and the second time period to obtain a first placement amount and a second placement amount, where the first placement amount is a quantity of advertisements placed at the advertisement slot in the first time period, and the second placement amount is a quantity of advertisements placed at the target advertisement slot in the second time period;
the training subunit is used for setting a second time parameter corresponding to the second time period and executing a preset training process; the training process comprises: inputting the configuration file, the second time parameter and the first input amount into the advertisement inventory model, triggering the advertisement inventory model to output a second predicted value corresponding to the second time period, and calculating the second predicted value and the second input amount to obtain an error value;
the judging subunit is used for judging whether the error value is larger than a preset minimum threshold value;
and the adjusting subunit is used for adjusting the model parameters of the advertisement inventory model according to the error value if the error value is greater than a preset minimum threshold value, and executing the training process until the error value is not greater than the minimum threshold value, and finishing adjusting the advertisement inventory model.
The device provided by the embodiment of the invention further comprises:
the calculation unit is used for carrying out logarithmic calculation on the effective advertisement putting amount corresponding to each time interval according to a preset logarithmic formula to obtain each logarithmic value corresponding to the target advertisement space; inputting the first time parameter, the configuration file and each logarithmic value into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a third predicted value corresponding to the target advertisement position, performing inverse operation on the third predicted value according to the logarithmic formula to obtain an operation value, and determining the operation value as the advertisement inventory corresponding to the target advertisement position in the future time period.
The specific working processes of each unit and sub-unit in the advertisement inventory prediction device disclosed in the above embodiment of the present invention can refer to the corresponding contents in the advertisement inventory prediction method disclosed in the above embodiment of the present invention, and are not described herein again.
The embodiment of the invention also provides a storage medium, which comprises a stored instruction, wherein when the instruction runs, the equipment where the storage medium is located is controlled to execute the advertisement inventory prediction method.
An electronic device is provided in an embodiment of the present invention, and the structural diagram of the electronic device is shown in fig. 5, which specifically includes a memory 501 and one or more instructions 502, where the one or more instructions 502 are stored in the memory 501, and are configured to be executed by one or more processors 503 to perform the following operations according to the one or more instructions 502:
when the advertisement inventory of a target advertisement position in a pre-specified future time period needs to be predicted, determining the total advertisement putting amount of the target advertisement position in a preset historical time period, wherein the historical time period comprises a plurality of time intervals, and the total advertisement putting amount is the sum of the advertisement putting amounts corresponding to the time intervals;
preprocessing each released advertisement in the historical time period to obtain the total effective advertisement release amount of the target advertisement space in the historical time period;
acquiring a preset configuration file, and adjusting a preset advertisement inventory model according to the configuration file and the total effective advertisement putting quantity to obtain an adjusted advertisement inventory model, wherein the configuration file comprises date parameters corresponding to each holiday date;
setting a first time parameter corresponding to the future time period, inputting the first time parameter, the configuration file and the total effective advertisement putting amount into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a first predicted value corresponding to the target advertisement position, and determining the first predicted value as the advertisement inventory corresponding to the target advertisement position in the future time period.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the system or system embodiments are substantially similar to the method embodiments and therefore are described in a relatively simple manner, and reference may be made to some of the descriptions of the method embodiments for related points. The above-described system and system embodiments are only illustrative, wherein the units described as separate parts may or may not be physically separate, and the parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both.
To clearly illustrate this interchangeability of hardware and software, various illustrative components and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
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 (10)
1. An advertising inventory forecasting method, comprising:
when the advertisement inventory of a target advertisement position in a pre-specified future time period needs to be predicted, determining the total advertisement putting amount of the target advertisement position in a preset historical time period, wherein the historical time period comprises a plurality of time intervals, and the total advertisement putting amount is the sum of the advertisement putting amounts corresponding to the time intervals;
preprocessing each released advertisement in the historical time period to obtain the total effective advertisement release amount of the target advertisement space in the historical time period;
acquiring a preset configuration file, and adjusting a preset advertisement inventory model according to the configuration file and the total effective advertisement putting quantity to obtain an adjusted advertisement inventory model, wherein the configuration file comprises date parameters corresponding to each holiday date;
setting a first time parameter corresponding to the future time period, inputting the first time parameter, the configuration file and the total effective advertisement putting amount into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a first predicted value corresponding to the target advertisement position, and determining the first predicted value as the advertisement inventory corresponding to the target advertisement position in the future time period.
2. The method of claim 1, wherein preprocessing each delivered advertisement in the historical time period to obtain a total effective advertisement placement amount of the target ad spot in the historical time period comprises:
acquiring traffic information of each released advertisement in each time interval, and judging whether an abnormally released advertisement exists or not according to the traffic information of each released advertisement;
if the abnormally-launched advertisements exist, rejecting each abnormally-launched advertisement in each time interval, and determining an effective advertisement inventory corresponding to each time interval;
and summing the effective advertisement inventory corresponding to each time interval to obtain the total effective advertisement putting amount of the target advertisement in the historical time period.
3. The method of claim 2, further comprising:
checking the effective advertisement putting quantity corresponding to each time interval, and judging whether the effective advertisement putting quantity corresponding to the current time interval is empty or not;
if the effective advertisement putting amount corresponding to the current time interval is empty, determining the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval;
and calculating the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval to obtain an average effective advertisement putting amount, and determining the average effective advertisement putting amount as the effective advertisement putting amount corresponding to the current time interval.
4. The method of claim 1, wherein adjusting a preset advertisement inventory model based on the configuration file and the total effective advertisement placement amount comprises:
dividing the historical time period into a first time period and a second time period, and dividing the total effective advertisement putting quantity according to the first time period and the second time period to obtain a first putting quantity and a second putting quantity, wherein the first putting quantity is the quantity of advertisements put at the advertisement slot in the first time period, and the second putting quantity is the quantity of advertisements put at the target advertisement slot in the second time period;
setting a second time parameter corresponding to the second time period, and executing a preset training process; the training process comprises: inputting the configuration file, the second time parameter and the first input amount into the advertisement inventory model, triggering the advertisement inventory model to output a second predicted value corresponding to the second time period, and calculating the second predicted value and the second input amount to obtain an error value;
judging whether the error value is larger than a preset minimum threshold value or not;
if the error value is larger than a preset minimum threshold value, adjusting the model parameters of the advertisement inventory model according to the error value, and executing the training process until the error value is not larger than the minimum threshold value, and finishing adjusting the advertisement inventory model.
5. The method of any one of claims 1 to 4, further comprising:
carrying out logarithmic calculation on the effective advertisement putting quantity corresponding to each time interval according to a preset logarithmic formula to obtain each logarithmic value corresponding to the target advertisement space;
inputting the first time parameter, the configuration file and each logarithmic value into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a third predicted value corresponding to the target advertisement position, performing inverse operation on the third predicted value according to the logarithmic formula to obtain an operation value, and determining the operation value as the advertisement inventory corresponding to the target advertisement position in the future time period.
6. An advertisement inventory forecasting device, comprising:
the system comprises a determining unit, a calculating unit and a calculating unit, wherein the determining unit is used for determining the total advertisement putting quantity of a target advertisement position in a preset historical time period when the advertisement inventory of the target advertisement position in a pre-specified future time period needs to be predicted, the historical time period comprises a plurality of time intervals, and the total advertisement putting quantity is the sum of the advertisement putting quantities corresponding to the time intervals;
the processing unit is used for preprocessing each released advertisement in the historical time period to obtain the total effective advertisement release amount of the target advertisement space in the historical time period;
the adjusting unit is used for acquiring a preset configuration file, adjusting a preset advertisement inventory model according to the configuration file and the total effective advertisement putting quantity, and acquiring an adjusted advertisement inventory model, wherein the configuration file comprises date parameters corresponding to each holiday date;
and the prediction unit is used for setting a first time parameter corresponding to the future time period, inputting the first time parameter, the configuration file and the total effective advertisement putting amount into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a first prediction value corresponding to the target advertisement space, and determining the first prediction value as the advertisement inventory corresponding to the target advertisement space in the future time period.
7. The apparatus of claim 6, wherein the processing unit comprises:
the judging subunit is used for acquiring traffic information of each released advertisement in each time interval and judging whether an abnormally released advertisement exists or not according to the traffic information of each released advertisement;
the determining subunit is used for eliminating each abnormally-launched advertisement in each time interval if the abnormally-launched advertisement exists, and determining the effective advertisement inventory corresponding to each time interval;
and the calculating subunit is used for summing the effective advertisement inventory corresponding to each time interval to obtain the total effective advertisement putting amount of the target advertisement in the historical time period.
8. The apparatus of claim 7, further comprising:
the processing subunit is used for checking the effective advertisement putting quantity corresponding to each time interval and judging whether the effective advertisement putting quantity corresponding to the current time interval is empty or not; if the effective advertisement putting amount corresponding to the current time interval is empty, determining the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval; and calculating the effective advertisement putting amount corresponding to the previous time interval and the effective advertisement putting amount corresponding to the next time interval to obtain an average effective advertisement putting amount, and determining the average effective advertisement putting amount as the effective advertisement putting amount corresponding to the current time interval.
9. The apparatus of claim 6, wherein the adjusting unit comprises:
a dividing subunit, configured to divide the historical time period into a first time period and a second time period, and divide the total effective advertisement placement amount according to the first time period and the second time period to obtain a first placement amount and a second placement amount, where the first placement amount is a quantity of advertisements placed at the advertisement slot in the first time period, and the second placement amount is a quantity of advertisements placed at the target advertisement slot in the second time period;
the training subunit is used for setting a second time parameter corresponding to the second time period and executing a preset training process; the training process comprises: inputting the configuration file, the second time parameter and the first input amount into the advertisement inventory model, triggering the advertisement inventory model to output a second predicted value corresponding to the second time period, and calculating the second predicted value and the second input amount to obtain an error value;
the judging subunit is used for judging whether the error value is larger than a preset minimum threshold value;
and the adjusting subunit is used for adjusting the model parameters of the advertisement inventory model according to the error value if the error value is greater than a preset minimum threshold value, and executing the training process until the error value is not greater than the minimum threshold value, and finishing adjusting the advertisement inventory model.
10. The apparatus of claim 6, further comprising:
the calculation unit is used for carrying out logarithmic calculation on the effective advertisement putting amount corresponding to each time interval according to a preset logarithmic formula to obtain each logarithmic value corresponding to the target advertisement space; inputting the first time parameter, the configuration file and each logarithmic value into the adjusted advertisement inventory model, triggering the adjusted advertisement inventory model to output a third predicted value corresponding to the target advertisement position, performing inverse operation on the third predicted value according to the logarithmic formula to obtain an operation value, and determining the operation value as the advertisement inventory corresponding to the target advertisement position in the future time period.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010076323.4A CN111260414B (en) | 2020-01-23 | 2020-01-23 | Advertisement inventory prediction method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010076323.4A CN111260414B (en) | 2020-01-23 | 2020-01-23 | Advertisement inventory prediction method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111260414A true CN111260414A (en) | 2020-06-09 |
| CN111260414B CN111260414B (en) | 2023-04-07 |
Family
ID=70951022
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010076323.4A Active CN111260414B (en) | 2020-01-23 | 2020-01-23 | Advertisement inventory prediction method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111260414B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112418907A (en) * | 2020-10-29 | 2021-02-26 | 当趣网络科技(杭州)有限公司 | Advertisement putting method and device, electronic equipment and medium |
| CN113011910A (en) * | 2021-01-18 | 2021-06-22 | 腾讯科技（深圳）有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN113837814A (en) * | 2021-11-01 | 2021-12-24 | 北京字节跳动网络技术有限公司 | Method and device for predicting quantity of released resources, readable medium and electronic equipment |
| CN114422826A (en) * | 2020-10-28 | 2022-04-29 | 腾讯科技（深圳）有限公司 | Multimedia content playing control method, device, equipment and medium |
| CN115456529A (en) * | 2022-09-13 | 2022-12-09 | 上海数禾信息科技有限公司 | Material inventory early warning method and system, computer equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100082442A1 (en) * | 2008-10-01 | 2010-04-01 | Yahoo! Inc. | Demand forecasting system and method for online advertisements |
| US20150095166A1 (en) * | 2013-02-19 | 2015-04-02 | ORIOLE MEDIA CORPORATION dbc Juice Mobile | System, method and computer program for providing qualitative ad bidding |
| CN104778608A (en) * | 2015-04-13 | 2015-07-15 | 合一信息技术（北京）有限公司 | N+ advertisement putting and optimizing 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 | Method for estimating stock of front-mounted advertisements |
| CN108428161A (en) * | 2018-06-20 | 2018-08-21 | 北京奇艺世纪科技有限公司 | Advertisement can injected volume prediction technique, device and electronic equipment |
| CN109741104A (en) * | 2018-12-28 | 2019-05-10 | 微梦创科网络科技（中国）有限公司 | A kind of preordering method and device of ad traffic resource |
| CN109978221A (en) * | 2019-01-02 | 2019-07-05 | 阿里巴巴集团控股有限公司 | A kind of advertisement inventory method for pre-distributing and device |
| CN110009383A (en) * | 2019-01-02 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of advertisement inventory method for pre-distributing and device |
-
2020
- 2020-01-23 CN CN202010076323.4A patent/CN111260414B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100082442A1 (en) * | 2008-10-01 | 2010-04-01 | Yahoo! Inc. | Demand forecasting system and method for online advertisements |
| US20150095166A1 (en) * | 2013-02-19 | 2015-04-02 | ORIOLE MEDIA CORPORATION dbc Juice Mobile | System, method and computer program for providing qualitative ad bidding |
| CN104778608A (en) * | 2015-04-13 | 2015-07-15 | 合一信息技术（北京）有限公司 | N+ advertisement putting and optimizing 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 | Method for estimating stock of front-mounted advertisements |
| CN108428161A (en) * | 2018-06-20 | 2018-08-21 | 北京奇艺世纪科技有限公司 | Advertisement can injected volume prediction technique, device and electronic equipment |
| CN109741104A (en) * | 2018-12-28 | 2019-05-10 | 微梦创科网络科技（中国）有限公司 | A kind of preordering method and device of ad traffic resource |
| CN109978221A (en) * | 2019-01-02 | 2019-07-05 | 阿里巴巴集团控股有限公司 | A kind of advertisement inventory method for pre-distributing and device |
| CN110009383A (en) * | 2019-01-02 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of advertisement inventory method for pre-distributing and device |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114422826A (en) * | 2020-10-28 | 2022-04-29 | 腾讯科技（深圳）有限公司 | Multimedia content playing control method, device, equipment and medium |
| CN114422826B (en) * | 2020-10-28 | 2023-06-27 | 腾讯科技（深圳）有限公司 | Multimedia content playing control method, device, equipment and medium |
| CN112418907A (en) * | 2020-10-29 | 2021-02-26 | 当趣网络科技(杭州)有限公司 | Advertisement putting method and device, electronic equipment and medium |
| CN113011910A (en) * | 2021-01-18 | 2021-06-22 | 腾讯科技（深圳）有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN113011910B (en) * | 2021-01-18 | 2023-08-08 | 腾讯科技（深圳）有限公司 | Data processing method and device, electronic equipment and storage medium |
| CN113837814A (en) * | 2021-11-01 | 2021-12-24 | 北京字节跳动网络技术有限公司 | Method and device for predicting quantity of released resources, readable medium and electronic equipment |
| CN113837814B (en) * | 2021-11-01 | 2024-08-27 | 北京字节跳动网络技术有限公司 | Method and device for predicting quantity of released resources, readable medium and electronic equipment |
| CN115456529A (en) * | 2022-09-13 | 2022-12-09 | 上海数禾信息科技有限公司 | Material inventory early warning method and system, computer equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111260414B (en) | 2023-04-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111260414B (en) | Advertisement inventory prediction method and device | |
| CN107958382A (en) | Abnormal behaviour recognition methods, device, electronic equipment and storage medium | |
| US10379502B2 (en) | Control system with machine learning time-series modeling | |
| CN112215546B (en) | Object page generation method, device, equipment and storage medium | |
| CN110400184B (en) | Method and apparatus for generating information | |
| CN109191202A (en) | Resource allocation method, device, electronic equipment and computer readable storage medium | |
| US20150081398A1 (en) | Determining a performance target setting | |
| CN111310988B (en) | Advertisement flow prediction method and device | |
| CN110889725B (en) | Online advertisement CTR estimation method, device, equipment and storage medium | |
| CN111080276A (en) | Payment method, device, equipment and storage medium for withholding order | |
| CN115270013B (en) | Method and device for evaluating emission reduction measures during activity and electronic equipment | |
| CN113034171A (en) | Business data processing method and device, computer and readable storage medium | |
| CN111932314A (en) | Method, device and equipment for pushing recommended content and readable storage medium | |
| CN110191119B (en) | Method and device for determining APP (application) generating abnormal traffic | |
| CN113139826A (en) | Method and device for determining distribution authority of advertisement space and computer equipment | |
| CN116823471A (en) | Transaction policy return method and device, electronic equipment and storage medium | |
| JP7101403B2 (en) | Electric power transaction support device and electric power transaction support method | |
| US20080167940A1 (en) | Method and structure for increasing revenue for on-demand environments | |
| CN115170167A (en) | Advertisement diagnosis method, device, computer equipment and storage medium | |
| CN111815374B (en) | Advertisement putting method and device, electronic equipment and medium | |
| CN109816414A (en) | Advertisement effectiveness data prediction technique and device | |
| CN113129127A (en) | Early warning method and device | |
| CN113888200A (en) | Data processing method, display method and electronic equipment | |
| CN113177838A (en) | Loan interest rate pricing method and device | |
| CN113139825A (en) | Method and device for determining distribution authority of advertisement space and computer equipment |
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 |