CN108985810B - Method and device for advertising on demand side platform - Google Patents
Method and device for advertising on demand side platform Download PDFInfo
- Publication number
- CN108985810B CN108985810B CN201710407472.2A CN201710407472A CN108985810B CN 108985810 B CN108985810 B CN 108985810B CN 201710407472 A CN201710407472 A CN 201710407472A CN 108985810 B CN108985810 B CN 108985810B
- Authority
- CN
- China
- Prior art keywords
- advertisement
- time
- exposure
- locking
- cost
- 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
- G06Q30/0273—Determination of fees for advertising
-
- 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
Abstract
The invention discloses a method and a device for advertising on a demand side platform, and relates to the technical field of computers. One embodiment of the method comprises: extracting feature data from the received advertisement putting request; calculating the exposure times of the advertisement according to the rules stored in advance and the extracted characteristic data; calculating the cost of the advertisement for the release according to the exposure times of the advertisement; when the budget of the advertisement is not less than the cost, the cost is locked from the budget and the advertisement is delivered. The implementation method can solve the problems of repeated charging and over-budget putting of the demand side platform in the process of carrying out advertisement putting, and enables the consumption of the advertisement budget to be more reasonable and controllable.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a method and a device for advertising on a demand side platform.
Background
In recent years, mobile internet advertisers have become popular, and the advertisement is expressed as a primary advertisement (also called as a stream advertisement). The native advertisement emphasizes the concept of "content, i.e. advertisement", which can be naturally fused with the computer application app (application) in terms of appearance, form and content, and thus is well received by advertisers.
Different from the traditional internet advertisement, due to the instability of the mobile internet, each APP can adopt a special caching strategy to cache content and advertisements at a client according to the use habits of users so as to optimize the user experience. However, this causes two problems:
1. because there is a time difference between the request and the exposure of the advertisement, the Demand-Side Platform DSP (Demand-Side Platform) deducts the fee from the advertiser and delays the fee compared with the exposure time of the advertisement;
2. the user slides up and down in the screen, which causes the problem of repeated exposure of the advertisement content returned by the same advertisement request.
At present, for information flow advertisement, most demand side platforms and advertisers charge according to the exposure times of the advertisement, and the difficulty of charging the advertisers by the demand side platforms is increased by delaying fee deduction and repeated exposure: how to maximize the advertisement putting effect of an advertiser and ensure accurate budget control on the premise of ensuring the profit of a demand side platform is a problem to be solved urgently. In addition, the complexity of the problem is further increased due to the diversity of the charging modes of the demand side platform and each media side.
Generally, when a user accesses an APP, an advertising media party of the APP sends an advertising delivery request to a demand party platform; after receiving an advertisement putting request sent by a media party, the demand party platform can put advertisements to the media party in a pack-section buying mode according to time periods, and can also return determined advertisement content and quotation to the media party together to request the media party to put the advertisements in a bidding mode; for the information flow advertisement bought in the package section, the media party puts in the advertisement according to the agreement with the demand party platform, and for the information flow advertisement bidding in real time, the media party decides which advertisement should be displayed according to the quotation of different demand party platforms. For the information flow advertisement of real-time bidding, the media party determines the demand party platform winning the bid and the advertisement to be displayed according to the rules of the secondary price sealed bidding auction, and carries out the fee deduction standard on the demand party platform. By a sealed bid auction, it is meant that the bidder submits a sealed bid and the highest bid obtains the item, but the price is not equal to his bid price, but is only next to the second highest bid price of his bid. After the advertisement is exposed (namely the advertisement is displayed), the media party informs the demand party platform and deducts the fee of the demand party platform, and then the demand party platform carries out corresponding fee deduction operation on the advertiser.
At present, the method for deducting the fee of the advertiser by the demand side platform is correspondingly as follows:
1. for the information flow advertisement which is bought in a packet mode according to the time period, directly deducting fees of an advertiser according to the buying cost and the price of bidding by a demand side platform when the request is made;
2. for the information flow advertisement of real-time bidding, the price macro is added in the advertisement exposure, and the media side platform carries out price replacement. After receiving the exposure request, the platform of the demand side decrypts the transaction price and directly deducts the fee of the advertiser.
In the process of implementing the invention, the inventor finds that at least the following problems exist in the prior art:
1. the charging strategy of the media party is not considered, for example, the charging of some media parties can be repeated by exposure for many times, and the charging of some media parties can be repeated by exposure for many times only once, which can cause the problem that the platform of the demand party can charge the advertiser for many times or repeatedly in order to ensure the benefit of the demand party;
2. the uncertainty of the advertisement exposure times can be caused without considering the browsing habits of different users, so that the fluctuation range of the charging is larger, and the difficulty of the main control budget of the advertisement is increased; moreover, due to the delayed exposure characteristic of the information flow advertisement, the problem of over-budget playing is caused by uncertainty of deduction after the advertisement plan is suspended and the budget is used up.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method and an apparatus for advertisement delivery by a demander platform, which can solve the problems of repeated charging and over-budget delivery during advertisement delivery by the demander platform, so that consumption of advertisement budget is more reasonable and controllable.
To achieve the above object, according to an aspect of the embodiments of the present invention, a method for advertisement delivery by a demander platform is provided.
A method for advertisement placement of a demander platform comprises the following steps: extracting feature data from the received advertisement putting request; calculating the exposure times of the advertisement according to a rule stored in advance and the extracted feature data; calculating the cost of the advertisement putting according to the exposure times; locking the fee from the budget and delivering the advertisement when the budget of the advertisement is not less than the fee.
Optionally, the feature data comprises user behavior feature data, media side feature data and time feature data.
Optionally, the pre-stored rule is obtained by performing offline model training on the feature data of a recent time period, where the recent time period has a predefined duration.
Optionally, after calculating the number of exposures of the advertisement, the method further includes: and calculating a charging coefficient of each exposure according to the exposure times so as to correct the exposure times.
Optionally, after locking the fee from the budget and placing the advertisement, the method further includes: setting a lock expiration time for a cost of the lock.
Optionally, after setting a lock expiration time of the cost of the lock, the method further includes: within the locking expiration time, if the advertisement is exposed, carrying out exposure deduction from the locking cost, and updating the locking cost; and releasing the expense of the locking after the expiration time of the locking.
Optionally, the number of exposures includes the number of exposures of the advertisement in different time interval intervals; and the lock expiration time is set according to the length of the time interval.
According to another aspect of the embodiment of the invention, a device for advertisement delivery of a demand side platform is provided.
An apparatus for advertisement placement on a demand side platform, comprising: the characteristic extraction module is used for extracting characteristic data from the received advertisement putting request; the exposure frequency estimation module is used for calculating the exposure frequency of the advertisement according to the prestored rule and the extracted characteristic data; the delivery cost calculation module is used for calculating the cost of the advertisement delivery according to the exposure times; and the delivery judging module is used for locking the cost from the budget and delivering the advertisement when the budget of the advertisement is not less than the cost.
Optionally, the feature data comprises user behavior feature data, media side feature data and time feature data.
Optionally, the pre-stored rule is obtained by performing offline model training on the feature data of a recent time period, where the recent time period has a predefined duration.
Optionally, the system further comprises an exposure number correction module, configured to: after the exposure times of the advertisement are calculated, calculating the charging coefficient of each exposure according to the exposure times so as to correct the exposure times.
Optionally, the system further comprises a lock time setting module, configured to: setting a lockout expiration time for the locked out fee after the fee is locked out from the budget and the advertisement is placed.
Optionally, the system further comprises a locking cost control module, configured to: after the locking expiration time of the locking expense is set, if the advertisement is exposed within the locking expiration time, carrying out exposure deduction from the locking expense, and updating the locking expense; and releasing the expense of the locking after the expiration time of the locking.
Optionally, the number of exposures includes the number of exposures of the advertisement in different time interval intervals; and the lock expiration time is set according to the length of the time interval.
According to another aspect of the embodiment of the invention, a terminal device for advertising on a demand side platform is provided.
A terminal device for advertisement delivery by a demander platform, comprising: one or more processors; the storage device is used for storing one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method for advertisement delivery by the demander platform provided by the embodiment of the invention.
According to yet another aspect of embodiments of the present invention, a computer-readable medium is provided.
A computer readable medium, on which a computer program is stored, when the program is executed by a processor, the program implements a method for advertisement delivery by a demander platform provided by an embodiment of the present invention.
One embodiment of the above invention has the following advantages or benefits: by predicting the exposure times and time of each advertisement delivery based on the user behavior data, the media party data and the time data, the cost of each advertisement delivery is estimated, and the advertisement budget is pre-locked, the budget can be accurately controlled, so that the problems of repeated charging and over-budget delivery of a demand party platform in the advertisement delivery process are solved, and the consumption of the advertisement budget is more reasonable and controllable.
Further effects of the above-mentioned non-conventional alternatives will be described below in connection with the embodiments.
Drawings
The drawings are included to provide a better understanding of the invention and are not to be construed as unduly limiting the invention. Wherein:
FIG. 1 is a schematic diagram of the main steps of a method for advertisement placement by a demander platform according to an embodiment of the invention;
FIG. 2 is a system architecture diagram of one embodiment of the present invention;
FIG. 3 is a budget control flow diagram for one embodiment of the present invention;
FIG. 4 is a schematic diagram of the main modules of an apparatus for advertisement placement by a demander platform according to an embodiment of the invention;
FIG. 5 is an exemplary system architecture diagram in which embodiments of the present invention may be employed;
fig. 6 is a schematic block diagram of a computer system suitable for use in implementing a terminal device or server of an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present invention are described below with reference to the accompanying drawings, in which various details of embodiments of the invention are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
In order to solve the problems in the prior art, embodiments of the present invention provide a method for advertisement delivery by a demander platform, which can more reasonably deduct fees for advertisers, implement accurate budget control, and solve the problems of repeated billing and over-budget delivery during advertisement delivery by the demander platform.
Fig. 1 is a schematic diagram of main steps of a method for advertisement delivery by a demander platform according to an embodiment of the present invention. As shown in fig. 1, the method for advertisement delivery by a demander platform according to the embodiment of the present invention mainly includes the following steps S101 to S104.
Step S101: feature data is extracted from the received advertisement placement request.
The feature data to be extracted mainly comprises user behavior feature data, media side feature data and time feature data.
The feature data of the user, the media party, the time and the like can be extracted from the real-time advertisement putting request sent by the media party, and the feature data mainly comprises the following steps: user identification ID, App ID, user's location data, operator data, mobile phone operating system, mobile phone model, App type, advertisement location, channel ID, billing type, time of request (hourly accurate), day of week, etc. data. App types may include books, business, merchandise guides, education, entertainment, finance, gourmet, games, fitness, life, medical, music, navigation, news, press magazines, photography and video, efficiency, references, social, sports, travel, tools, weather, etc.; channel IDs may include financial, sports, entertainment, etc.; the charging type refers to a charging settlement mode of a media party and a demand party platform, and comprises exposure charging before duplication elimination and exposure charging after duplication elimination. Because the frequency of repeated exposure of the advertisement can obviously change along with the time, the data analysis can be better carried out by introducing the time characteristic.
Step S102: and calculating the exposure times of the advertisement according to the rules stored in advance and the extracted characteristic data.
The pre-stored rule is obtained by performing offline model training on the feature data of the latest time period, and the latest time period has predefined duration. By analyzing and machine learning the feature data of the user, the media party, the time and the like in the latest period of time, a data model can be obtained for estimating the exposure times of the advertisement after each advertisement delivery according to the feature data. For example: and performing offline model training on the feature data in the last month to obtain a data model corresponding to the feature data so as to realize the pre-stored rule.
The off-line model training process can be mainly realized through the following units: the device comprises a log receiving unit, a log splicing unit, a feature extraction unit, a model training unit and a model updating unit.
And the log receiving unit is responsible for receiving an advertisement request log and an advertisement browsing log of a terminal user on the App, a real-time fee deduction log and an offline fee deduction log of a media party for advertisement display and the like, performing simple time detection on the logs, filtering abnormal logs and the like. The abnormal logs mainly refer to cheating advertisement browsing and fee deduction logs, network request data caused by a web crawler and the like. The advertisement charging strategy of the media party can be obtained by analyzing the data of the real-time fee deduction log, the off-line fee deduction log and the like of the media party for advertisement display.
When the user uses the APP, the media party initiates an advertisement request to the demand party platform to inform the demand party platform to deliver the advertisement. And when the user browses the advertisement, the App also sends an advertisement exposure message to the demand side platform so that the demand side platform can deduct the fee of the advertiser according to the exposure message. These data record the user's browsing timestamp on each ad (indexed by ad request ID), browsing times, and the source of the requested traffic (from which media party platform), APP name, ad slot characteristics: such as which tag ("sports", "finance", etc.) of the news APP.
A log splicing unit: the method is mainly used for splicing all logs received by the log receiving module according to the identification ID of the advertisement request.
A feature extraction unit: and performing feature extraction on the spliced log so as to provide input sample data for a subsequent machine learning module. According to the technical scheme of the embodiment of the invention, modeling is mainly carried out based on the behavior data of the user, and the used characteristics comprise the following groups:
1. user behavior feature data
The characteristics included in behavior data of an advertisement request, an advertisement exposure, an advertisement click and the like of a user mainly include: user identification ID, App ID, geographic position, operator data, mobile phone operating system, mobile phone model and the like;
2. media side characteristic data
Mainly media related data features, for example, include: app type, ad placement, channel ID (finance, sports, entertainment, etc.), billing type, billing variance gap. The charging type refers to a charging settlement mode of a media party and a demand party platform, and mainly comprises exposure charging before duplication elimination and exposure charging after duplication elimination. The charging difference gap is the difference (gap) between the cost counted by the media party and the cost counted by the platform of the demand party, and because of the factors such as network transmission, log transmission, anti-cheating and the like, the statistical error of data occurs between the two parties, and the part of the error can be used as a part of the characteristics for subsequent calculation;
3. temporal feature data
From the data, the frequency of repeated exposures varied significantly over time within each day. And model training can be better performed by introducing the time characteristics. The time characteristics mainly include: hours, days of the week, time intervals between repeated exposures, etc. The time interval of the repeated exposure refers to the difference between the time stamp of the multiple exposures of the same request and the request time. To better identify the feature of the time interval of the repeated exposure, the time interval may be discretized, for example, into the following time interval intervals: (0-10 minutes), (10 minutes-30 minutes), (30 minutes-1 hour), (1 hour-2 hours), (2 hours-12 hours), (12 hours-1 day), (1 day or more) for 7 time periods, and the "hour" and "day of the week" inside the feature refer to the point in time when the request occurs, not the point in time when the advertisement is exposed. In practical use, time interval intervals with different numbers and lengths can be set for different systems and applications.
The data after the feature extraction process is arranged into a feature matrix, and the data during each advertisement exposure corresponds to one row in the matrix. Each row of data comprises 3 parts of data of user behavior characteristics, media side characteristics and time characteristics.
A model training unit: and predicting the exposure times of each time interval by performing feature extraction on the log and utilizing a regression algorithm in machine learning.
And obtaining a training sample set through feature extraction, and selecting a proper machine learning algorithm to train a model. For each input X, the corresponding output Y is calculated through the Model obtained through training. According to the embodiment of the invention, the input X is a characteristic data matrix obtained by the characteristic extraction module, and the output Y is the advertisement exposure times in different time interval intervals. The prediction of the number of advertisement exposures in each discrete time period by using the machine learning Model, i.e. the training process of the Model, can be regarded as a regression problem. That is, for each input X, a corresponding Y is determined. The problem can be described as:
where Yj represents the number of exposures in the jth time period, Xi is the input feature, θiIs the parameter corresponding to Xi. To simplify the representation, let X in the input features01, then:
given a training set, to obtain the parameter θ. We define a loss function J (θ) that describes hθ(x)iAnd corresponding yiThe degree of proximity therebetween:
theta is adjusted by using a gradient descent method so that J (theta) takes a minimum value.
A model updating unit: the unit is used for importing the off-line trained model into an on-line real-time computing environment. The off-line model training system generates a piece of model data every day, and after the model data is generated, the model data is imported to the on-line mode through the model updating unit. For the stability of the system, before the newly generated model is imported to the online, the existing data of the current day is used as a test set to verify the effect of the model, and then the new model is updated to the online system.
The characteristic data can be trained off line through the log receiving unit, the log splicing unit, the characteristic extracting unit, the model training unit and the model updating unit to obtain a characteristic data model and update the characteristic data model to an on-line system. The feature data model is a pre-stored rule mentioned in the embodiment of the present invention.
After receiving an advertisement putting request sent by a media party, the demand party platform can estimate the exposure times of the advertisement after the advertisement is put according to the characteristic data model and the online extracted characteristic data.
As can be seen from the above description, the number of exposures in embodiments of the present invention includes the number of exposures of an advertisement in different time interval intervals. Assuming that the number of exposures in each time interval is represented by Fi (floating point number), the number of exposures during each time interval in which the advertisement is delivered can be obtained, as shown in table 1 below.
TABLE 1
| Interval of time | Number of exposures |
| T1(0-10 minutes) | F1 |
| T2(10 min-30 min) | F2 |
| T3(30 minutes-1 hour) | F3 |
| T4(1 hr-2 hr) | F4 |
| T5(2 hours-12 hours) | F5 |
| T6(12 hours-1 day) | F6 |
| T7 (more than 1 day) | F7 |
In table 1, for the number of exposures in each time interval, a charging weight may be correspondingly allocated to correspond to different charging policies. When all the billing weights are the default value 1, the total exposure times of the advertisement in the advertisement delivery is the sum of the exposure times in each time interval. Those skilled in the art will appreciate that the length, number, etc. of the time interval partitions can be set according to the actual usage, and the contents shown in table 1 are only examples.
In addition, after the exposure times of the advertisements are calculated, the charging coefficient of each exposure can be calculated according to the exposure times so as to correct the exposure times.
Because of different charging strategies of media parties, repeated charging of multiple exposures of advertisements within a short time is possible, which is unreasonable for advertisers; moreover, because the charging difference gap may exist between the media party and the platform of the demand party due to factors such as network transmission, log transmission, anti-cheating, etc., the platform of the demand party can correct the estimated exposure times under the condition, and the corresponding public expression of the charging coefficient R is as follows:
wherein, gap is counted in the feature data of the media side extracted by the feature extraction unit; n is the number of time interval intervals, in this embodiment, n is 7; wiIs the charging weight for each time interval, and is 1 by default. WiThe system is a configuration item of the system, and can be specifically set according to different media parties, different users and different time periods so as to be better used for correcting the advertisement exposure times.
Step S103: and calculating the cost of the advertisement for the advertisement putting according to the exposure times.
According to the exposure times of the advertisement obtained in step S102 and the exposure unit price bid _ price (i.e., the cost of exposing the advertisement once) agreed by the requesting platform and the advertiser, the cost P of the advertisement delivered at this time can be estimated, where P is bid _ price F R, where F is the exposure times and R is the charging coefficient.
Step S104: when the budget of the advertisement is not less than the cost, the cost is locked from the budget and the advertisement is delivered.
If the cost P of the advertisement for this placement is determined in step S103, the cost P can be locked from the budget of the advertisement and the advertisement can be placed when the budget of the advertisement is not less than P. After the cost is locked from the budget, the partial cost in the budget of the corresponding advertisement becomes unavailable, that is, the partial cost P should be deducted from the budget of the advertisement.
In addition, after locking the fee from the budget and placing the advertisement, the method further comprises the following steps: a lock expiration time for the cost of the lock is set. As can be seen from the description in step S102, the number of exposures in the embodiment of the present invention includes the number of exposures of the advertisement in different time interval intervals. Then, the lock expiration time is set according to the length of the time interval. Generally, the lock expiration time is taken as an upper limit value of the corresponding time interval. For example: the lock expiration time for the charge P1 required for the exposure F1 may be set to 10 minutes corresponding to T1(0-10 minutes) shown in table 1; the lock expiration time for the exposure time F2 required fee P2 can be set to 30 minutes corresponding to T2(10 minutes to 30 minutes) shown in table 1, and so on.
After the locking expiration time of the locking expense is set, if the advertisement is exposed within the locking expiration time, carrying out exposure deduction from the locking expense and updating the locking expense; after the lock expiration time expires, the cost of the lock is released.
Such as: assuming that the predicted exposure F1 is 5 times for T1(0-10 minutes) shown in Table 1 and the required cost is 5 dollars, the ad is placed with 5 dollars from its budget and a lock expiration time of 10 minutes. If the ad is exposed 4 times in 10 minutes, the exposure charge required can be deducted from the 5 dollars in the lock, while the cost of renewing the lock is 1 dollar, but the lock expiration time is unchanged. After 10 minutes, the lock expiration time expires, then the cost 1 dollar opportunity for the lock is released and returned to the budget for the advertisement. In a special case, if the advertisement is exposed 6 times within 10 minutes of the lock expiration time, and 6 dollars are to be deducted, the exposure fee of 6 dollars will be first deducted from the locked 5 dollars, and the excess exposure fee of 1 dollar will be deducted from the unlocked budget of the advertisement. In particular, if the unlocked budget of the advertisement is not sufficient to deduct the excess exposure charges, the advertisement may be deducted from the costs locked in other time intervals, or may not be deducted, and all the costs to be locked are released and then deducted.
According to the steps S101 to S104, the exposure times of the advertisement can be calculated according to the characteristic data and the offline training model, so that the cost of advertisement delivery is obtained, and the cost is locked from the budget, so that the problem of over-budget delivery can be solved; because the feature data comprises the media party data, the problem of repeated charging caused by different charging strategies of different media parties is solved, and the consumption of the advertisement budget is more reasonable and controllable.
Fig. 2 is a system architecture diagram of one embodiment of the present invention. As shown in fig. 2, the technical solution of this embodiment is mainly divided into two parts, namely, offline and real-time online. The off-line part mainly works for data analysis and model calculation, and the real-time on-line part mainly performs cost prediction on a real-time advertisement delivery request to determine advertisement delivery.
The off-line part is mainly realized by the following units: the device comprises a log receiving unit, a log splicing unit, a feature extraction unit, a model training unit and a model updating unit. For a specific implementation process, reference may be made to the description in step S102, and details are not described here.
And when the user accesses the App, the media server is triggered to send an advertisement putting request to the demander platform, the advertisement putting control system of the real-time online part estimates the cost of the advertisement putting according to the received advertisement putting request and the offline training data model, and determines the advertisement to be put according to the estimated cost and the advertisement budget. And after the advertisement is put, deducting the fee of the advertiser according to the exposure times of the advertisement so as to control the budget of the advertiser in real time.
The advertiser's budget is easily exceeded because of the delayed exposure feature of the information flow advertisement. After the exposure times and the charging coefficient are estimated, the cost of the advertisement put at this time can be calculated, the budget of the advertiser is pre-locked, and the locking expiration time is set according to the exposure time interval. When all budgets for the advertisement are locked, delivery of the advertisement is stopped. Pre-locking refers to the initial deduction of the advertiser's budget without actual ad exposure, hence the deduction of the budget. If exposure occurs within the lock expiration time, the pre-charging fee is converted into a real charging fee; if there is no exposure within the lock expiration time, the pre-paid fee is released and returned to the advertiser.
How embodiments of the present invention perform budget control is described below in conjunction with FIG. 3.
FIG. 3 is a budget control flow diagram of one embodiment of the invention. As shown in fig. 3, the estimated exposure times Fi and charging coefficients Ri of the advertisement in different time interval intervals Ti are obtained, the cost Pi of the advertisement in the current delivery in the time interval intervals Ti is calculated to be bid _ price Fi Ri, and then the sum of all Pi is obtained to obtain the cost P of the advertisement in the current delivery. And when the advertisement budget is not less than P, putting the advertisement, locking the cost P from the budget, and setting a locking expiration time, wherein each part Pi in the cost P sets the locking expiration time as Ti according to the interval length of the corresponding time interval Ti. After the advertisement is put, if the real exposure occurs in the locking period, the real fee deduction is carried out on the advertisement and the locking fee is updated, when the locking expiration time is expired, the locking fee is released, the budget of the advertisement is updated, and the advertisement putting is finished. Each time an advertisement placement request from a media party is received, a decision is made as to whether or not to place the advertisement by comparing the estimated cost of advertisement placement with the advertisement budget. If the budget is not sufficient, the advertisement is not put. Therefore, the advertisement budget can be controlled, so that the advertisement putting decision can be better carried out, and the exceeding of the advertisement budget is avoided.
Fig. 4 is a schematic diagram of main modules of an apparatus for advertisement delivery by a demander platform according to an embodiment of the present invention. As shown in fig. 4, the apparatus 400 for advertisement delivery on a demand side platform according to the embodiment of the present invention mainly includes a feature extraction module 401, an exposure number estimation module 402, a delivery cost calculation module 403, and a delivery determination module 404.
The feature extraction module 401 is configured to extract feature data from the received advertisement delivery request;
the exposure times estimating module 402 is used for calculating the exposure times of the advertisement according to the pre-stored rules and the extracted feature data;
the placement cost calculation module 403 is configured to calculate the cost of the advertisement for this placement according to the exposure times;
the placement determination module 404 is used to lock out the cost from the budget and place the advertisement when the budget of the advertisement is not less than the cost.
In an embodiment of the present invention, the feature data may mainly include user behavior feature data, media side feature data, and time feature data.
According to the technical scheme of the embodiment of the invention, the pre-stored rule is obtained by performing off-line model training on the feature data of the latest time period, and the latest time period has the predefined duration. For example: and performing offline model training on the feature data in the last month to obtain a data model corresponding to the feature data so as to realize the pre-stored rule.
In addition, the device 400 for advertisement delivery on the demander platform according to the embodiment of the present invention may further include an exposure number correction module (not shown in the figure), configured to: after the exposure times of the advertisement are calculated, the charging coefficient of each exposure is calculated according to the exposure times so as to correct the exposure times.
The device 400 for advertisement delivery of the demander platform according to the embodiment of the present invention may further include a locking time setting module (not shown in the figure), configured to: after the cost is locked out from the budget and the advertisement is placed, a lock expiration time for the locked out cost is set.
Moreover, the apparatus 400 for advertisement delivery by the demander platform according to the embodiment of the present invention may further include a locking cost control module (not shown in the figure), configured to: after the locking expiration time of the locking expense is set, if the advertisement is exposed within the locking expiration time, carrying out exposure deduction from the locking expense and updating the locking expense; after the lock expiration time expires, the cost of the lock is released.
According to the technical scheme of the embodiment of the invention, the exposure times can comprise the exposure times of the advertisement in different time interval intervals; and, the lock expiration time is set according to the length of the time interval.
Fig. 5 illustrates an exemplary system architecture 500 of a method for advertisement placement by a demander platform or an apparatus for advertisement placement by a demander platform to which embodiments of the present invention may be applied.
As shown in fig. 5, the system architecture 500 may include terminal devices 501, 502, 503, a network 504, and a server 505. The network 504 serves to provide a medium for communication links between the terminal devices 501, 502, 503 and the server 505. Network 504 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 501, 502, 503 to interact with a server 505 over a network 504 to receive or send messages or the like. The terminal devices 501, 502, 503 may have installed thereon various communication client applications, such as shopping-like applications, web browser applications, search-like applications, instant messaging tools, mailbox clients, social platform software, etc. (by way of example only).
The terminal devices 501, 502, 503 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop portable computers, desktop computers, and the like.
The server 505 may be a server providing various services, such as a background management server (for example only) providing support for shopping websites browsed by users using the terminal devices 501, 502, 503. The backend management server may analyze and perform other processing on the received data such as the product information query request, and feed back a processing result (for example, target push information, product information — just an example) to the terminal device.
It should be noted that the method for advertisement delivery by the demander platform provided by the embodiment of the present invention is generally executed by the server 505, and accordingly, the apparatus for advertisement delivery by the demander platform is generally disposed in the server 505.
It should be understood that the number of terminal devices, networks, and servers in fig. 5 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation.
According to the introduction of the prior art, the problems of repeated charging, over-budget playing and the like of an advertisement charging system are caused due to the characteristics of multiple exposure and delayed exposure of the App information stream advertisement. In order to solve the problems, embodiments of the present invention provide a method for advertisement delivery on a demand side platform, which is used for estimating the cost of each advertisement delivery by analyzing user behavior data, media side data and time data, and can overcome the problem that budget cannot be controlled due to different browsing habits of users and different APP cache policies, so that an advertisement delivery decision can be made according to the budget of an advertiser, so that the budget consumption of the advertiser is more reasonable and controllable, and the problems of billing abnormality and over-budget playing are solved.
Referring now to FIG. 6, a block diagram of a computer system 600 suitable for use with a terminal device implementing an embodiment of the invention is shown. The terminal device shown in fig. 6 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present invention.
As shown in fig. 6, the computer system 600 includes a Central Processing Unit (CPU)601 that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM)602 or a program loaded from a storage section 608 into a Random Access Memory (RAM) 603. In the RAM 603, various programs and data necessary for the operation of the system 600 are also stored. The CPU 601, ROM 602, and RAM 603 are connected to each other via a bus 604. An input/output (I/O) interface 605 is also connected to bus 604.
The following components are connected to the I/O interface 605: an input portion 606 including a keyboard, a mouse, and the like; an output portion 607 including a display such as a Cathode Ray Tube (CRT), a Liquid Crystal Display (LCD), and the like, and a speaker; a storage section 608 including a hard disk and the like; and a communication section 609 including a network interface card such as a LAN card, a modem, or the like. The communication section 609 performs communication processing via a network such as the internet. The driver 610 is also connected to the I/O interface 605 as needed. A removable medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 610 as necessary, so that a computer program read out therefrom is mounted in the storage section 608 as necessary.
In particular, according to the embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication section 609, and/or installed from the removable medium 611. The computer program performs the above-described functions defined in the system of the present invention when executed by the Central Processing Unit (CPU) 601.
It should be noted that the computer readable medium shown in the present invention can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present invention, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In the present invention, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams or flowchart illustration, and combinations of blocks in the block diagrams or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present invention may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor comprises a feature extraction module, an exposure frequency estimation module, a delivery cost calculation module and a delivery judgment module. The names of these modules do not in some cases constitute a limitation of the module itself, for example, the feature extraction module may also be described as a "module for extracting feature data from a received advertisement placement request".
As another aspect, the present invention also provides a computer-readable medium that may be contained in the apparatus described in the above embodiments; or may be separate and not incorporated into the device. The computer readable medium carries one or more programs which, when executed by a device, cause the device to comprise: extracting feature data from the received advertisement putting request; calculating the exposure times of the advertisement according to a rule stored in advance and the extracted feature data; calculating the cost of the advertisement putting according to the exposure times; locking the fee from the budget and delivering the advertisement when the budget of the advertisement is not less than the fee.
According to the technical scheme of the embodiment of the invention, the exposure times and time of each advertisement delivery are predicted based on the user behavior data, the media party data and the time data, so that the cost of each advertisement delivery is estimated, the advertisement budget is pre-locked, and the budget can be accurately controlled, so that the problems of repeated charging and over-budget delivery of a demand party platform in the process of advertisement delivery are solved, and the consumption of the advertisement budget is more reasonable and controllable.
The technical scheme of the embodiment of the invention fully considers the caching strategy of each APP and the use habits of different users, fully considers the flow acquisition and purchase mode of a demand side platform, and sets different charging coefficients according to the historical delay fee deduction time distribution; for repeated exposure outside the delay fee deduction time, fee deduction is carried out on the advertiser according to the weight and the delay factor, so that the experience of the advertiser is better; a feedback mechanism is automatically formed, manual intervention is not needed, the price of each advertisement display is more reasonably determined, the advertisement putting effect is improved as much as possible on the premise of ensuring a certain profit margin of a demand side platform, and the user experience of an advertiser is improved; moreover, the situation of over-budget advertisement delivery can be avoided.
The above-described embodiments should not be construed as limiting the scope of the invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions can occur, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (12)
1. A method for advertisement delivery of a demand side platform is characterized by comprising the following steps:
extracting characteristic data from the received advertisement putting request, wherein the characteristic data comprises time characteristic data, user behavior characteristic data and media side characteristic data, and the time characteristic data comprises time intervals of repeated exposure;
calculating the exposure times of the advertisement according to a pre-stored rule and the extracted feature data, wherein the exposure times comprise the exposure times of the advertisement in different time interval intervals;
calculating the cost of the advertisement putting according to the exposure times;
locking the cost from the budget and delivering the advertisement when the budget of the advertisement is not less than the cost;
after calculating the number of times of exposure of the advertisement, the method further comprises the following steps: calculating a charging coefficient of each exposure according to the exposure times to correct the exposure times, wherein a formula corresponding to a charging system R is as follows:
wherein, gap is the charging difference counted in the media side feature data extracted by the feature extraction unit; n is the number of time interval intervals; fiIs the number of exposures per time interval; wiThe charging weight of each time interval is set according to different media parties, different users and different time periods.
2. The method of claim 1, wherein the pre-saved rules are derived by off-line model training of the feature data for a recent period of time, the recent period of time having a predefined duration.
3. The method of claim 1, wherein after locking the fee from the budget and placing the advertisement, further comprising:
setting a lock expiration time for a cost of the lock.
4. The method of claim 3, wherein setting a lock expiration time for the cost of the lock further comprises:
within the locking expiration time, if the advertisement is exposed, carrying out exposure deduction from the locking cost, and updating the locking cost;
and releasing the expense of the locking after the expiration time of the locking.
5. The method according to claim 1 or 3, wherein the lock expiration time is set according to the length of the time interval.
6. The utility model provides a demand side platform carries out device of advertisement putting which characterized in that includes:
the system comprises a characteristic extraction module, a characteristic analysis module and a characteristic analysis module, wherein the characteristic extraction module is used for extracting characteristic data from a received advertisement putting request, the characteristic data comprises time characteristic data, user behavior characteristic data and media side characteristic data, and the time characteristic data comprises a time interval of repeated exposure;
the exposure frequency estimation module is used for calculating the exposure frequency of the advertisement according to a pre-stored rule and the extracted characteristic data, wherein the exposure frequency comprises the exposure frequency of the advertisement in different time interval intervals;
the delivery cost calculation module is used for calculating the cost of the advertisement delivery according to the exposure times;
a delivery decision module for locking the cost from the budget and delivering the advertisement when the budget of the advertisement is not less than the cost;
an exposure number correction module for: after the exposure times of the advertisement are calculated, calculating a charging coefficient of each exposure according to the exposure times so as to correct the exposure times, wherein a formula corresponding to a charging system R is as follows:
wherein, gap is the charging difference counted in the media side feature data extracted by the feature extraction unit; n is the number of time interval intervals; fiIs the number of exposures per time interval; wiThe charging weight of each time interval is set according to different media parties, different users and different time periods.
7. The apparatus of claim 6, wherein the pre-saved rules are obtained by performing offline model training on the feature data for a recent period of time, the recent period of time having a predefined duration.
8. The apparatus of claim 6, further comprising a lock time setting module to:
setting a lockout expiration time for the locked out fee after the fee is locked out from the budget and the advertisement is placed.
9. The apparatus of claim 8, further comprising a locking cost control module to:
after the locking expiration time of the locking expense is set, if the advertisement is exposed within the locking expiration time, carrying out exposure deduction from the locking expense, and updating the locking expense;
and releasing the expense of the locking after the expiration time of the locking.
10. The apparatus according to claim 6 or 8, wherein the lock expiration time is set according to a length of the time interval.
11. The utility model provides a terminal equipment that demander platform carried out advertisement and puts, its characterized in that includes:
one or more processors;
storage means for storing one or more programs;
when executed by the one or more processors, cause the one or more processors to implement the method of any one of claims 1-5.
12. A computer-readable medium, on which a computer program is stored, which, when being executed by a processor, carries out the method according to any one of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710407472.2A CN108985810B (en) | 2017-06-02 | 2017-06-02 | Method and device for advertising on demand side platform |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710407472.2A CN108985810B (en) | 2017-06-02 | 2017-06-02 | Method and device for advertising on demand side platform |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108985810A CN108985810A (en) | 2018-12-11 |
| CN108985810B true CN108985810B (en) | 2022-03-04 |
Family
ID=64501595
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710407472.2A Active CN108985810B (en) | 2017-06-02 | 2017-06-02 | Method and device for advertising on demand side platform |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108985810B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110148021A (en) * | 2019-05-13 | 2019-08-20 | 北京乐信圣文科技有限责任公司 | Bid methodologies and system are launched in advertisement |
| CN111080357B (en) * | 2019-12-12 | 2023-08-25 | 恩亿科(北京)数据科技有限公司 | Method and device for determining product release duty ratio, electronic equipment and storage medium |
| CN110942355B (en) * | 2019-12-13 | 2023-05-23 | 北京搜狐新媒体信息技术有限公司 | Advertisement duplication removing method and device |
| CN113506139B (en) * | 2021-07-19 | 2024-03-01 | 成都屏盟科技有限公司 | Advertisement putting pricing method and device and computer equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102110265A (en) * | 2009-12-23 | 2011-06-29 | 深圳市腾讯计算机系统有限公司 | Network advertisement effect estimating method and network advertisement effect estimating system |
| CN104268644A (en) * | 2014-09-23 | 2015-01-07 | 新浪网技术(中国)有限公司 | Method and device for predicting click frequency of advertisement at advertising position |
| CN106127528A (en) * | 2016-06-30 | 2016-11-16 | 北京小米移动软件有限公司 | Advertisement placement method and device |
| CN106204136A (en) * | 2016-07-08 | 2016-12-07 | 合网络技术（北京）有限公司 | A kind of advertisement delivery system and advertisement placement method |
| CN106658069A (en) * | 2016-12-16 | 2017-05-10 | 华扬联众数字技术股份有限公司 | Method and device for advertisement putting |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10223703B2 (en) * | 2010-07-19 | 2019-03-05 | Mediamath, Inc. | Systems and methods for determining competitive market values of an ad impression |
| WO2014043882A1 (en) * | 2012-09-20 | 2014-03-27 | 华为技术有限公司 | Advertisement launching method, device and system |
| TW201516918A (en) * | 2013-10-16 | 2015-05-01 | Utechzone Co Ltd | System for managing advertising effectiveness and method therefore |
| CN104636945B (en) * | 2013-11-11 | 2018-03-27 | 秒针信息技术有限公司 | A kind of method and device for realizing DSP advertisement puttings |
| KR20170006128A (en) * | 2015-07-07 | 2017-01-17 | 에스케이플래닛 주식회사 | Advertisement transmission service providing device for processing log in DSP interworking, user equipment, mediation service providing device, DSP service providing device, system comprising the same, control method thereof and computer readable medium having computer program recorded therefor |
| CN106204109B (en) * | 2016-06-28 | 2019-02-26 | 腾讯科技（深圳）有限公司 | Media file pulls method and apparatus |
| CN106372959A (en) * | 2016-08-22 | 2017-02-01 | 广州图灵科技有限公司 | Internet-based user access behavior digital marketing system and method |
| CN106097025A (en) * | 2016-08-24 | 2016-11-09 | 苏州新科兰德科技有限公司 | A kind of advertisement placement method based on DSP and device |
-
2017
- 2017-06-02 CN CN201710407472.2A patent/CN108985810B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102110265A (en) * | 2009-12-23 | 2011-06-29 | 深圳市腾讯计算机系统有限公司 | Network advertisement effect estimating method and network advertisement effect estimating system |
| CN104268644A (en) * | 2014-09-23 | 2015-01-07 | 新浪网技术(中国)有限公司 | Method and device for predicting click frequency of advertisement at advertising position |
| CN106127528A (en) * | 2016-06-30 | 2016-11-16 | 北京小米移动软件有限公司 | Advertisement placement method and device |
| CN106204136A (en) * | 2016-07-08 | 2016-12-07 | 合网络技术（北京）有限公司 | A kind of advertisement delivery system and advertisement placement method |
| CN106658069A (en) * | 2016-12-16 | 2017-05-10 | 华扬联众数字技术股份有限公司 | Method and device for advertisement putting |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108985810A (en) | 2018-12-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10649818B2 (en) | Multi-touch attribution model for valuing impressions and other online activities | |
| CN108985810B (en) | Method and device for advertising on demand side platform | |
| CN107767164B (en) | Information processing method, distribution method and device for push information order | |
| US20170103429A1 (en) | Intelligent ad auction and sla compliance techniques | |
| US20160275570A1 (en) | Methods and systems for dynamic auction floors | |
| Geyik et al. | Multi-touch attribution based budget allocation in online advertising | |
| US10217118B2 (en) | Systems and methods for implementing bid adjustments in an online advertisement exchange | |
| US20150100438A1 (en) | Selecting among advertisements competing for a slot associated with electronic content delivered over a network based upon predicted latency | |
| US20200134663A1 (en) | Automatic resource adjustment based on resource availability | |
| CN109949064B (en) | Open interface call charging method and device | |
| US20210150585A1 (en) | Artificial intelligence techniques for bid optimization used for generating dynamic online content | |
| JP2015060463A (en) | Service provision device, service provision method, and service provision program | |
| KR101430648B1 (en) | Method and apparatus for exposing targeting advertisement | |
| US10565613B2 (en) | Simulating advertising campaigns | |
| CN113256330A (en) | Information delivery effect attribution method and device | |
| US20150081455A1 (en) | Service providing apparatus and service providing method | |
| US11093861B2 (en) | Controlling item frequency using a machine-learned model | |
| KR20130126282A (en) | Apparatus and method for providing ranking information of advertisement | |
| CN111724181A (en) | Information pushing method, device, medium and electronic equipment | |
| CN115936764A (en) | Product promotion method and device | |
| US20130246161A1 (en) | Systems and methods for optimization-aware delivery pacing adjustment in advertisement serving | |
| CN114445128A (en) | Card ticket management method and device, electronic equipment and computer readable medium | |
| EP3013055A1 (en) | Video frame set processing cost management method, apparatus and related computer program product | |
| KR20200048401A (en) | A method for providing transaction services of advertisement traffic networks | |
| US20160019583A1 (en) | Systems and methods for smooth and effective budget delivery in online advertising |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |