CN113379461B - Advertisement exposure prediction method based on deep learning - Google Patents
Advertisement exposure prediction method based on deep learning Download PDFInfo
- Publication number
- CN113379461B CN113379461B CN202110711389.0A CN202110711389A CN113379461B CN 113379461 B CN113379461 B CN 113379461B CN 202110711389 A CN202110711389 A CN 202110711389A CN 113379461 B CN113379461 B CN 113379461B
- Authority
- CN
- China
- Prior art keywords
- advertisement
- predicted
- bidding
- fine
- module
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000013135 deep learning Methods 0.000 title claims abstract description 26
- 230000002860 competitive effect Effects 0.000 claims description 18
- 239000000284 extract Substances 0.000 claims description 3
- 238000013136 deep learning model Methods 0.000 abstract description 4
- 238000005070 sampling Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011282 treatment Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000008685 targeting Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0242—Determining effectiveness of advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
- G06Q30/0275—Auctions
-
- 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
- G06Q30/0277—Online advertisement
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Economics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Molecular Biology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for estimating advertisement exposure of deep learning, which comprises the following steps: step 1, acquiring future bidding opportunities of advertisements to be predicted; step 2, judging whether the future bidding opportunity of the advertisement to be predicted is a valid bidding opportunity, if so, adding the advertisement to be predicted into a fine queuing queue of the valid bidding opportunity; step 3, predicting the exposure rate of the effective bidding opportunities of the advertisement to be predicted from the refined queue of the effective bidding opportunities through deep learning; and step 4, adding the estimated exposure rates of the advertisement to be predicted in all effective bidding opportunities to obtain the final exposure number of the advertisement to be predicted. The method has the advantage of automatic modeling of diversity and fine-ranking competitiveness due to the deep learning model, and is suitable for the exposure estimated scene of the online display advertisement with complex constraint.
Description
Technical Field
The invention relates to the field of online advertisement information processing, in particular to a method for advertisement exposure prediction based on deep learning.
Background
Advertisement exposure estimation techniques are important to online advertising systems that can provide advertisers with advertising effects at a certain bid. It is also the underlying technology for many advertising tools, such as bid simulators, potential advertisement screening, and the like. The accurate estimated value can bring better user experience, and is beneficial to the long-term development of the platform.
The existing methods mainly comprise two methods: the first method is based on the playback of historical data, namely, the historical bid data of the advertisement is found, the bid ordering score of the advertisement is recalculated according to the bid value, and the first ordering times are counted to be used as the final advertisement predicted value; the second method is based on Bayesian sampling and dynamic linear model, firstly establishing Bayesian network according to historical bidding data, then generating future bidding data of advertisements by sampling Bayesian network, wherein the sampling data is obtained by dynamic linear model, and the method is mainly aimed at searching advertisements, and the dynamic linear model can establish sequence model for each advertisement keyword. However, for online advertising, the existing methods have at least the following limitations:
(1) Effective inventory predictions for appropriate treatments are not considered: the first approach uses the historical data entirely without consideration of the possible future fluctuations in advertisement bid inventory. The second method uses a dynamic linear model to estimate inventory sequences, and has no way to deal with the problem of high-dimensional time sequence estimation caused by complex targeting in the presentation of advertisements.
(2) Advertisement diversity constraints are not considered: both existing methods are essentially to judge whether the advertisement can obtain the first name of the order in the bidding opportunities under different bid settings, however, the actual advertisement system also has a filtering mechanism to ensure that the advertisement types put to the same user are rich enough to avoid aesthetic fatigue of the user.
Disclosure of Invention
Aiming at the problems of the existing method, the invention aims to provide the advertisement exposure prediction method based on deep learning, which can model the diversity of advertisements while generating the effective bidding opportunities of the displayed advertisements, thereby realizing higher prediction precision.
The invention aims at realizing the following technical scheme:
the embodiment of the invention provides a method for advertisement exposure estimation based on deep learning, which comprises the following steps:
step 1, acquiring future bidding opportunities of advertisements to be predicted;
step 2, judging whether the future bidding opportunity of the advertisement to be predicted is a valid bidding opportunity, if so, adding the advertisement to be predicted into a fine queuing queue of the valid bidding opportunity;
step 3, predicting the exposure rate of the effective bidding opportunities of the advertisement to be predicted from the refined queue of the effective bidding opportunities through deep learning;
and step 4, adding the estimated exposure rates of the advertisement to be predicted in all effective bidding opportunities to obtain the final exposure number of the advertisement to be predicted.
As can be seen from the technical scheme provided by the invention, the method for estimating advertisement exposure based on deep learning provided by the embodiment of the invention has the beneficial effects that:
through effective bidding inventory estimation and combination of deep learning exposure rate estimation, the display advertisement estimation considering diversity is realized, compared with an estimation mode of feature extraction, complicated feature engineering is avoided, the advertisement diversity can be automatically modeled, and the estimation precision of higher advertisement exposure rate is realized.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for advertisement exposure estimation based on deep learning according to an embodiment of the present invention;
FIG. 2 is a flowchart of a method for deep learning based advertisement exposure estimation according to an embodiment of the present invention;
fig. 3 is a schematic diagram of an advertisement exposure estimation model of a method for advertisement exposure estimation based on deep learning according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will clearly and fully describe the technical solutions of the embodiments of the present invention in conjunction with the specific contents of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to fall within the scope of the invention. What is not described in detail in the embodiments of the present invention belongs to the prior art known to those skilled in the art.
As shown in fig. 1 and 2, the embodiment of the invention provides a method for estimating advertisement exposure based on deep learning, which realizes the estimation of advertisement showing considering diversity through effective bidding inventory estimation and exposure rate estimation based on deep learning, and comprises the following steps:
step 1, acquiring future bidding opportunities of advertisements to be predicted;
step 2, judging whether the future bidding opportunity of the advertisement to be predicted is a valid bidding opportunity, if so, adding the advertisement to be predicted into a fine queuing queue of the valid bidding opportunity;
step 3, predicting the exposure rate of the effective bidding opportunities of the advertisement to be predicted from the refined queue of the effective bidding opportunities through deep learning;
and step 4, adding the estimated exposure rates of the advertisement to be predicted in all effective bidding opportunities to obtain the final exposure number of the advertisement to be predicted.
In step 1 of the above method, the future bid opportunity of the advertisement is obtained according to the advertisement inventory estimation module of the existing contract advertisement system.
In step 2 of the above method, determining whether the future bid opportunity of the advertisement to be predicted is a valid bid opportunity includes:
and calculating the eCPM score of each future bidding opportunity of the advertisement to be predicted, judging whether the eCPM score is smaller than the lowest eCPM score in the fine queuing queue of the effective bidding opportunities, if so, the future bidding opportunity is the effective bidding opportunity, and if so, the future bidding opportunity is the ineffective bidding opportunity.
In step 2 of the above method, adding the advertisement to be predicted to the fine queuing of effective bidding opportunities comprises the following steps:
and finding the ordering position of the advertisement in the fine queuing list of the effective bidding opportunities according to the eCPM score of the advertisement to be predicted, and inserting the advertisement to be predicted into the ordering position of the fine queuing list of the effective bidding opportunities.
In step 3 of the above method, the exposure of the effective bidding opportunity of the advertisement to be predicted is estimated from the refined queue of effective bidding opportunities through deep learning in the following manner, including:
acquiring a fine queuing queue of effective bidding opportunities;
extracting user access information of each advertisement in a fine ranking queue of valid bidding opportunities;
establishing a correlation model among advertisements in the fine ranking queue of the effective bidding opportunities based on a self-attention mechanism by combining user access information of each advertisement through a deep-learning advertisement exposure rate estimation model;
and predicting the exposure rate of each advertisement in the fine-ranking queue of the effective bidding opportunities according to the correlation model among the advertisements.
In step 4 of the method, the exposure rates of the advertisements to be predicted estimated in step 3 are all added to obtain the final exposure number of the advertisements to be predicted.
In step 3 of the above method, the advertisement exposure rate estimation model (see fig. 3) includes: a base module, a competitive module and a timing module; wherein,
the output end of the timing sequence module is connected with the vector of the output end of the competitive module, and the timing sequence module can extract the user access information of each advertisement in the fine-ranking queue;
the output end of the competitive module is connected with the output end of the basic module in an addition way, and the competitive module can establish a correlation model among advertisements in the fine-ranking queue by combining the user access information of each advertisement output by the time sequence module through a self-attention mechanism;
the base module, the output end of which is connected with the output end of the time sequence module and the output end adding connection end of the competitive module, can estimate the exposure rate of each advertisement in the fine-ranking queue of the effective bidding opportunity according to the interrelation model among advertisements.
Specifically, the data of the timing module and the output of the competitive module are added to obtain a competitive factor, and then the competitive factor is multiplied by the output of the full-connection layer of the base module to obtain the final exposure rate.
In the advertisement exposure rate estimation model described above,
the time sequence module adopts an LSTM network;
the competitive module adopts a self-attention mechanism network; preferably, the Self-Attention mechanism network adopts a Self-Attention network;
the base module adopts a deep FM network.
According to the method, the display advertisement estimation considering diversity is realized through the effective bidding inventory estimation and the exposure rate estimation based on deep learning, and the method has the advantage of automatically modeling diversity and fine-ranking competitiveness due to the utilization of a deep learning model, so that the method is suitable for an exposure estimation scene of online display advertisements with complex constraints.
Embodiments of the present invention are described in detail below.
Referring to fig. 1 and 2, an embodiment of the present invention provides a method for advertisement exposure estimation based on deep learning, which mainly includes the following steps:
step 1, acquiring future bidding opportunities of advertisements to be predicted according to an advertisement inventory estimation module of a contract advertisement system; the method comprises the steps of carrying out a first treatment on the surface of the
Step 2, judging whether the advertisement can enter a fine queuing queue of bidding opportunities or not, wherein the advertisement can enter the fine queuing queue of bidding opportunities to become effective bidding opportunities;
step 3, generating the exposure rate of the advertisement to the effective bidding opportunities by using an advertisement exposure rate estimation model (namely a deep learning model);
and 4, adding the exposure rates of all effective bidding opportunities of the advertisement to obtain the final exposure number of the advertisement to be predicted.
Specifically, the steps in the method are as follows:
step 2, for each opportunity, the system searches its refined rank, and first calculates the score S of the advertisement C to be predicted at each opportunity according to the eCPM score calculation formula of the advertisement to be predicted q Since each advertisement in the fine-ranking queue has an eCPM score, determining S q Whether the advertisement is smaller than the lowest eCPM score in the bid opportunities, if so, the bid opportunities are invalid bid opportunities, otherwise, the advertisement is valid bid opportunities, and for the valid bid opportunities, the advertisement is judged according to S q Find the ranking position of the advertisement in the refined rank and insert it into that position.
Step 3, after the fine queuing of the effective bidding opportunities in step 2 is obtained, inputting the fine queuing into an advertisement exposure rate estimation model, wherein the fine queuing consists of three parts, namely a basic module, a competitive module and a time sequence module, as shown in figure 3; the basic module adopts a deep FM network, can output an exposure rate for each advertisement in the fine-ranking queue, and the competitive module adopts a Self-Attention mechanism network (Self-Attention network) and can model the correlation between the advertisements in the fine-ranking queue. Because their interrelationships are also affected by the historical access records of users, the time sequence module adopts an LSTM network, extracts user access information through the time sequence module, and inputs the user access information into the competitive module to realize modeling of advertisement diversity. For example, with respect to the constraint of freshness of the user, if the user has seen the sports shoe advertisement, the competitiveness of the sports shoe advertisement will decrease, and not the advertisement competitiveness of the sports shoe will increase, for the next chance of bidding he triggers.
And 4, for each effective bidding opportunity in the step 2, the advertisement exposure rate estimation model in the step 3 can generate an exposure rate, and all the exposure rates are added to obtain the final exposure number of the advertisement to be predicted.
The prediction method of the invention has the advantage of automatic modeling of diversity and fine-ranking competitiveness due to the utilization of the deep learning model, and is suitable for the exposure prediction scene of the online display advertisement with complex constraint.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the scope of the present invention should be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.
Claims (5)
1. A method for advertisement exposure prediction based on deep learning, comprising:
step 1, acquiring future bidding opportunities of advertisements to be predicted;
step 2, judging whether the future bidding opportunity of the advertisement to be predicted is a valid bidding opportunity, if so, adding the advertisement to be predicted into a fine queuing queue of the valid bidding opportunity; determining whether a future bid opportunity for an advertisement to be predicted is a valid bid opportunity comprises:
calculating the eCPM score of each future bidding opportunity of the advertisement to be predicted, judging whether the eCPM score is smaller than the lowest eCPM score in the fine queuing queue of the effective bidding opportunities, if so, the future bidding opportunities are effective bidding opportunities, and if so, the future bidding opportunities are ineffective bidding opportunities;
and step 3, predicting the exposure rate of the advertisement to be predicted in the effective bidding opportunity from the refined queuing of the effective bidding opportunity through deep learning, wherein the method comprises the following steps: acquiring a fine queuing queue of effective bidding opportunities;
extracting user access information of each advertisement in a fine ranking queue of valid bidding opportunities;
establishing a correlation model among advertisements in the fine ranking queue of the effective bidding opportunities based on a self-attention mechanism by combining user access information of each advertisement through a deep-learning advertisement exposure rate estimation model;
predicting the exposure rate of each advertisement in the fine-ranking queue of the effective bidding opportunities according to the interrelation model among the advertisements;
the advertisement exposure rate estimation model comprises the following steps: a base module, a competitive module and a timing module; wherein,
the output end of the timing sequence module is connected with the vector of the output end of the competitive module, and the timing sequence module can extract the user access information of each advertisement in the fine-ranking queue;
the output end of the competitive module is connected with the output end of the basic module in an addition way, and the competitive module can establish a correlation model among advertisements in the fine-ranking queue by combining the user access information of each advertisement output by the time sequence module through a self-attention mechanism;
the output end of the basic module is connected with the output end of the timing sequence module and the output end addition connecting end of the competitive module, and the basic module can estimate the exposure rate of each advertisement in the fine-ranking queue of the effective bidding opportunities according to the interrelation model among advertisements;
and step 4, adding all the estimated exposure rates of the advertisement to be predicted in the step 3 in all effective bidding opportunities to obtain the final exposure rate of the advertisement to be predicted.
2. The method according to claim 1, wherein in step 1, the future bid opportunity of the advertisement is obtained according to an advertisement inventory estimation module of the existing contract advertisement system.
3. The method for deep learning based advertisement exposure estimation according to claim 1, wherein in the step 2, adding the advertisement to be predicted to the fine queuing of effective bidding opportunities comprises:
and finding the ordering position of the advertisement in the fine queuing list of the effective bidding opportunities according to the eCPM score of the advertisement to be predicted, and inserting the advertisement to be predicted into the ordering position of the fine queuing list of the effective bidding opportunities.
4. The method for deep learning based advertisement exposure prediction of claim 1, wherein,
the time sequence module adopts an LSTM network;
the competitive module adopts a self-attention mechanism network;
the base module adopts a deep FM network.
5. The method for deep learning based advertisement exposure estimation according to claim 4, wherein the Self-Attention mechanism network adopts Self-Attention network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110711389.0A CN113379461B (en) | 2021-06-25 | 2021-06-25 | Advertisement exposure prediction method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110711389.0A CN113379461B (en) | 2021-06-25 | 2021-06-25 | Advertisement exposure prediction method based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113379461A CN113379461A (en) | 2021-09-10 |
| CN113379461B true CN113379461B (en) | 2023-12-01 |
Family
ID=77579153
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110711389.0A Active CN113379461B (en) | 2021-06-25 | 2021-06-25 | Advertisement exposure prediction method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113379461B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101324106B1 (en) * | 2012-07-26 | 2013-10-31 | 김수현 | Advertise exposure ranking management apparatus and method using client position |
| CN109615442A (en) * | 2019-01-23 | 2019-04-12 | 上海旺翔文化传媒股份有限公司 | RTB real time bid method based on excitation video ads |
| CN110033314A (en) * | 2019-03-18 | 2019-07-19 | 北京品友互动信息技术股份公司 | Advertisement data processing method and device |
| CN112734504A (en) * | 2021-04-02 | 2021-04-30 | 北京智慧易科技有限公司 | Intelligent traffic routing method, device and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10748173B2 (en) * | 2017-05-11 | 2020-08-18 | Jivox Corporation | Scoring users based on intent for online advertising |
-
2021
- 2021-06-25 CN CN202110711389.0A patent/CN113379461B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101324106B1 (en) * | 2012-07-26 | 2013-10-31 | 김수현 | Advertise exposure ranking management apparatus and method using client position |
| CN109615442A (en) * | 2019-01-23 | 2019-04-12 | 上海旺翔文化传媒股份有限公司 | RTB real time bid method based on excitation video ads |
| CN110033314A (en) * | 2019-03-18 | 2019-07-19 | 北京品友互动信息技术股份公司 | Advertisement data processing method and device |
| CN112734504A (en) * | 2021-04-02 | 2021-04-30 | 北京智慧易科技有限公司 | Intelligent traffic routing method, device and system |
Non-Patent Citations (4)
| Title |
|---|
| AlAli Moneera 等.Click through Rate Effectiveness Prediction on Mobile Ads Using Extreme Gradient Boosting.Computers Materials& Continua.2021,全文. * |
| 基于多级注意力机制网络的app流行度预测;张艺璇;郭斌;刘佳琪;欧阳逸;於志文;;计算机研究与发展(第05期);全文 * |
| 基于深度学习的搜索广告排序应用;钟小勇;电子技术与软件工程;第171卷(第1期);全文 * |
| 基于特征学习的广告点击率预估技术研究;张志强;周永;谢晓芹;潘海为;;计算机学报(第04期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113379461A (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108521439B (en) | Message pushing method and device | |
| CN106372249B (en) | A kind of clicking rate predictor method, device and electronic equipment | |
| US9817896B2 (en) | Search query scoring and popularity monitoring and systems and methods therefor | |
| US9245252B2 (en) | Method and system for determining on-line influence in social media | |
| CN102262647B (en) | Signal conditioning package, information processing method and program | |
| KR100911014B1 (en) | System and method of allowing sponsor match advertising | |
| WO2019148982A1 (en) | Sorting | |
| JP6267344B2 (en) | Content selection using quality control | |
| CN110796477A (en) | Advertisement display method and device, electronic equipment and readable storage medium | |
| CN111401963B (en) | Method and device for training user behavior prediction model | |
| US20100228558A1 (en) | Aggregate Content-Based Advertising | |
| CN103198086A (en) | Information processing device, information processing method, and program | |
| US20160379243A1 (en) | Method and system for forecasting a campaign performance using predictive modeling | |
| CN103942236A (en) | System and method for serving electronic content | |
| CN107222526A (en) | Push method, device, equipment and the computer-readable storage medium of promotion message | |
| US20120084142A1 (en) | Bid landscape forecasting in online advertising | |
| WO2015175835A1 (en) | Click through ratio estimation model | |
| CN106296287A (en) | For the method and apparatus adjusting the bid of promotion message | |
| CN110880127A (en) | Consumption level prediction method and device, electronic equipment and storage medium | |
| KR20170037760A (en) | System and method for bidding automatically using correlation of bidding related data in keyword advertisement, and bidding management server | |
| KR20170036874A (en) | Method and apparatus for recommendation of social event based on users preference | |
| US20130097053A1 (en) | Method and system for recommending a combined service by taking into account situation information on a target user and the degree of complementarity of a service | |
| JP2019074988A (en) | Room charge setting device, room charge setting method, and program | |
| CN113379461B (en) | Advertisement exposure prediction method based on deep learning | |
| US20160125454A1 (en) | Systems and methods for managing advertising campaigns |
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 |