CN106815749A - Advertisement putting map construction method and advertisement putting collection of illustrative plates deliver detection method extremely - Google Patents
Advertisement putting map construction method and advertisement putting collection of illustrative plates deliver detection method extremely Download PDFInfo
- Publication number
- CN106815749A CN106815749A CN201710041416.1A CN201710041416A CN106815749A CN 106815749 A CN106815749 A CN 106815749A CN 201710041416 A CN201710041416 A CN 201710041416A CN 106815749 A CN106815749 A CN 106815749A
- Authority
- CN
- China
- Prior art keywords
- collection
- illustrative plates
- media
- advertisement putting
- user
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0242—Determining effectiveness of advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
Abstract
The advertisement putting map construction method that the present invention is provided, media are mapped as by ordinate by the user accesses data of media, using the time as abscissa, by the dispensing index of different media is not plotted in corresponding position and obtains delivering collection of illustrative plates in the same time, therefore advertisement release process can be visualized by delivering collection of illustrative plates, advertiser and advertisement analysis personnel are able to observe that within the whole dispensing cycle, advertisement in injected volume not in the same time on different media, deliver the indexs such as cost；The advertisement putting collection of illustrative plates that the present invention is provided delivers detection method extremely, the dispensing collection of illustrative plates obtained using advertisement putting map construction method of the invention can be detected, surge detection according to collection of illustrative plates is delivered delivered Traffic Anomaly, jettison system failure or the sudden cheating of middle generation to advertisement putting collection of illustrative plates, convenient that launch process is tracked and controlled.
Description
Technical field
The invention belongs to Internet technical field, and in particular to a kind of advertisement putting map construction method and advertisement putting figure
Spectrum is abnormal to deliver detection method.
Background technology
Internet advertising, because proportion of its interactive strong, accuracy advantage high in advertisement structure rises year by year, and
There is the trend for continuing to expand.2011 to 2014 years, the market scale of Internet advertising had been over newspaper advertisement scale, ranks
Second, market scale continuously keeps rapid growth.
RTB (Real time Bidding, real time bid) is a kind of each net of utilization third party technology in substantial amounts
The technology of bidding for being estimated and bidding for the displaying behavior of each user advertising on standing.Real time bid has been evaded invalid
Audient reaches, and advertiser or its agency can carry out flow purchase for significant user.The core of RTB is DSP
(Demand Side Platform, party in request's platform) and Ad Exchange (ADX, advertisement trade market).
DSP incorporates the on-line advertisement resource of separate sources, and advertiser or its agency can be allowed unified by one
Platform is bought, manages advertising resource.By DSP, it is possible to achieve unified orientation is delivered, data management, optimize and effect report
Accuse.
ADX is a place similar to stock exchange, by ad serving side (site media) and party in request (advertiser
Or its agency) flock together transaction online advertisement stock, itself does not buy and sell online advertisement stock, and merely provides
The place of transaction, collects the charges by dealing money.ADX is one of important on-line advertisement resource source of DSP.In a word,
ADX is exactly one can allow advertiser in the correct time, and correct media exposure is transported to the place of suitable user in transaction
RTB technologies can allow advertiser or its act on behalf of DSP auctions advertisement each time and show chance in front of the user.
But, at present, during dispensing, people can not intuitively observe advertisement on certain moment media for advertisement
Injected volume and deliver indicator of costs etc., be not easy to advertiser and advertisement analysis personnel hold market trend, adjust traffic assignments
Strategy.And can not in time find Traffic Anomaly, jettison system failure or burst that advertisement occurs during dispensing
Property cheating etc. phenomenon, it has not been convenient to launch process is tracked and is controlled.
The content of the invention
The present invention is carried out to solve the above problems, it is therefore intended that offer one kind is able to observe that entirely delivers
In cycle, advertisement in injected volume not in the same time on different media, deliver the advertisement putting map construction method of the indexs such as cost,
And it is capable of detecting when that Traffic Anomaly, jettison system failure or sudden cheating occur in launch process, it is convenient to delivering
The advertisement putting collection of illustrative plates dispensing detection method extremely that journey is tracked and controls.
The invention provides a kind of advertisement putting map construction method, it is characterised in that comprise the following steps：Step 1, will
Any two media obtain the user's registration between two media based on pre-defined rule, and matchmaker is obtained according to user's registration
Body similarity matrix；
Step 2, projection vector is obtained according to the media similarity matrix based on pre-defined rule；
Step 3, the projection coordinate of each media is obtained according to the projection vector based on pre-defined rule；
Step 4, with the time as abscissa, media are ordinate, and the dispensing index of advertisement is depicted as to deliver collection of illustrative plates.
Further, in the advertisement putting map construction method that the present invention is provided, can also have the feature that：Its
In, in step 1, the user that each media obtains occurring on the media according to real time bidding data, and table with a matrix type
Show：
Domain=(user1 user2 … usern)
Domain is mediamatrix, and n is positive integer.
Further, in the advertisement putting map construction method that the present invention is provided, can also have the feature that：Its
In, the user's registration between two media is tried to achieve by Jacobi similarity formula：
The media similarity matrix is：
mpqIt is the user's registration between two media, M is p row q column matrix, and p, q are positive integer.
Further, in the advertisement putting map construction method that the present invention is provided, can also have the feature that：Its
In, in step 2, the projection vector is tried to achieve based on PCA, and solution formula is as follows：
arg maxμTMμ
s.t.μTμ=1
μ is projection vector, μTIt is the transposed vector of μ, M is media similarity matrix.
Further, in the advertisement putting map construction method that the present invention is provided, can also have the feature that：Its
In, in step 3, the projection coordinate is tried to achieve by below equation：
ki=μTdomaini
domaini=(user1 user2 … usern)
kiIt is the projection coordinate of certain media, μ is projection vector, domainiIt is certain mediamatrix.
Further, in the advertisement putting map construction method that the present invention is provided, can also have the feature that：Its
In, in step 4, draw dispensing collection of illustrative plates and comprise the steps of：
Step 4-1, calculate under synchronization, the projection coordinate of all media, and each media represents and clicks on index,
Step 4-2, by projection coordinate's identical media under synchronization represent and click on index merge,
Step 4-3, the time is abscissa, and media are ordinate, by representing of being obtained in step 4-2 and clicks on index and draws
Obtain delivering collection of illustrative plates at corresponding position.
Detection method is delivered extremely present invention also offers a kind of advertisement putting collection of illustrative plates, it is characterised in that comprising following step
Suddenly：Step 1, decomposes SVD is according to the dispensing collection of illustrative plates obtained by advertisement putting map construction method recited above；
Step 2, chooses low-rank recovery coefficient, and then carrying out low-rank reconstruction to dispensing collection of illustrative plates obtains rebuilding collection of illustrative plates；
Step 3, reconstruction error is obtained according to collection of illustrative plates is delivered with collection of illustrative plates is rebuild based on pre-defined rule；
Step 4, judges the size of the reconstruction error and threshold value, if the reconstruction error is less than the threshold value, chooses
The low-rank recovery coefficient of step 2, otherwise, return to step 2, and increase the value of the low-rank recovery coefficient；
Whether step 5, the low-rank recovery factor analysis advertisement putting collection of illustrative plates chosen according to step 4 delivers exception occurs.
Further, in the advertisement putting collection of illustrative plates that the present invention is provided delivers detection method extremely, can also have such
Feature：Wherein, in step 1, the formula that the dispensing collection of illustrative plates does SVD decomposition is as follows：
X=U Σ VT
X is to deliver collection of illustrative plates；U is p × p rank unitary matrice；Σ is positive semidefinite p × q rank diagonal matrix；VT, i.e., the conjugation of V turn
Put, be q × q rank unitary matrice.
Further, in the advertisement putting collection of illustrative plates that the present invention is provided delivers detection method extremely, can also have such
Feature：Wherein, in step 2, low-rank reconstruction is carried out to delivering collection of illustrative plates by formula below：
To rebuild collection of illustrative plates, c is low-rank recovery coefficient, UcThe preceding c component of representing matrix U, ΣcThe preceding c of representing matrix Σ
Individual component,Representing matrix VTPreceding c component.
Further, in the advertisement putting collection of illustrative plates that the present invention is provided delivers detection method extremely, can also have such
Feature：Wherein, in step 3, the computing formula of reconstruction error is as follows：
ε is reconstruction error, and X is dispensing collection of illustrative plates,To rebuild collection of illustrative plates.
Advantages of the present invention is as follows：
According to advertisement putting map construction method involved in the present invention, media are reflected by the user accesses data of media
It is ordinate to penetrate, using the time as abscissa, by the dispensing index of different media is not plotted in corresponding position and obtains in the same time
Collection of illustrative plates is delivered, therefore advertisement release process can be visualized by delivering collection of illustrative plates, advertiser and advertisement analysis personnel can
It was observed that within the whole dispensing cycle, advertisement in injected volume not in the same time on different media, deliver the indexs such as cost.
Advertisement putting collection of illustrative plates involved in the present invention delivers detection method extremely, can be to using advertisement putting of the invention
The dispensing collection of illustrative plates that map construction method is obtained is detected, delivered to advertisement putting collection of illustrative plates according to the surge detection for delivering collection of illustrative plates
Traffic Anomaly, jettison system failure or the sudden cheating of middle generation are crossed, it is convenient that launch process is tracked and controlled.
Brief description of the drawings
Fig. 1 is the flow chart of advertisement putting map construction method in the present invention；
Fig. 2 is the flow chart of advertisement putting collection of illustrative plates method for detecting abnormality in sample in the present invention.
Specific embodiment
In order that technological means, creation characteristic, reached purpose and effect that the present invention is realized are easy to understand, it is real below
Apply example combination accompanying drawing and detection method work is delivered extremely specifically to advertisement putting map construction method of the present invention and advertisement putting collection of illustrative plates
Illustrate.
Advertisement putting map construction method is used to be presented advertisement release process in the way of collection of illustrative plates, can be visual
Change.As shown in figure 1, advertisement putting map construction method is comprised the steps of：
Step S1-1, user's registration that any two media are obtained between two media based on pre-defined rule, according to
User's registration obtains media similarity matrix.
In advertising business, media are typically represented with domain name (domain) and advertisement position (adslot).Each media is from RTB
Can obtain the user occurred on the media in (Real time Bidding, real time bid) broadcast data, and by each matchmaker
Body is represented with a matrix type：
Domain=(user1 user2 … usern)
Wherein, domain is mediamatrix, and n is positive integer.
In the present embodiment, the user for occurring on each media is as shown in table 1.
The user occurred on table 1. each media
| User 1 | User 2 | … | User n | |
| Domain 1 | 1 | 0 | … | 0 |
| Domain 2 | 1 | 1 | … | 1 |
| … | … | … | … | … |
| Domain P | 0 | 0 | … | 1 |
User's registration between two media is tried to achieve by Jacobi similarity formula：
Media similarity matrix is：
Wherein, mpqIt is the user's registration between two media, M is p row q column matrix, and p, q are positive integer.
In the present embodiment, the user's registration two-by-two between media is as shown in table 2.
The user's registration two-by-two between media of table 2.
Then media similarity matrix is：
Step S1-2, projection vector μ is obtained according to media similarity matrix based on pre-defined rule.Projection vector μ is based on master
Componential analysis are tried to achieve, and solution formula is as follows：
arg maxμTMμ
s.t.μTμ=1
μ is projection vector, μTIt is the transposed vector of μ.
Step S1-3, the projection coordinate of each media is obtained according to projection vector based on pre-defined rule.Projection coordinate kiBy
Below equation is tried to achieve：
ki=μTdomaini
domaini=(user1 user2 … usern)
domainiIt is certain mediamatrix.
The corresponding projection coordinate of each media being calculated using said process is met:User's registration between media
The projection coordinate for obtaining high approaches, and the low projection coordinate for obtaining of the user's registration between media separates.
Step S1-4, with the time as abscissa, media are ordinate, and the dispensing index of advertisement is depicted as to deliver collection of illustrative plates X.
It is two dimensional image to deliver collection of illustrative plates X, and t represents time, k presentation mediums, Xk,tRepresent dispensing index of the t on media k.Deliver
Index is included and represents (impression) and click (click) index.Collection of illustrative plates X drawing process is delivered to comprise the following steps：
Step S1-4-1, calculates under synchronization, the projection coordinate of all media, and each media represents
(impression) and click on (click) index.In the present embodiment, synchronization difference media projection coordinate, represent and
The data for clicking on index are as shown in table 3.
The synchronization of table 3. difference media projection coordinate, represent and click on index
| Media | Projection coordinate | impression | click |
| domain1 | k1 | 3000 | 10 |
| domain2 | k2 | 2000 | 20 |
| domain3 | k1 | 1000 | 30 |
| … | … | … | … |
Step S1-4-2, by projection coordinate's identical media under synchronization represent and click on index merge.Such as table 4
It is shown.
It is corresponding under synchronization difference coordinate in the coordinate system of table 4. to show and click on index
| Projection coordinate | impression | click |
| k1 | 4000 | 40 |
| k2 | 2000 | 20 |
| … | … | … |
Step S1-4-3, the time is abscissa, and media are ordinate, by representing of being obtained in step S1-4-2 and click refers to
Mark is plotted at corresponding position and obtains delivering collection of illustrative plates.
Collection of illustrative plates can be illustrated in the whole dispensing cycle, the index such as injected volume, the dispensing cost of advertisement on different media.
Advertisement putting collection of illustrative plates delivers method for detecting abnormality can be according to the surge detection of dispensing collection of illustrative plates to advertisement putting collection of illustrative plates
Delivering Traffic Anomaly, jettison system failure or the sudden cheating of middle generation, it is convenient launch process is tracked and
Control.
As shown in Fig. 2 advertisement putting collection of illustrative plates delivers method for detecting abnormality comprising the steps of：
Step S2-1, decomposes SVD is according to the dispensing collection of illustrative plates obtained by advertisement putting map construction method.
The formula that SVD is decomposed is as follows：
X=U Σ VT
X is to deliver collection of illustrative plates；U is p × p rank unitary matrice；Σ is positive semidefinite p × q rank diagonal matrix；VT, i.e., the conjugation of V turn
Put, be q × q rank unitary matrice.
Step S2-2, chooses reconstructed coefficients c, and then carrying out low-rank reconstruction to dispensing collection of illustrative plates X obtains rebuilding collection of illustrative platesDeliver
It is as follows that collection of illustrative plates X carries out low-rank reconstruction formula：
To rebuild collection of illustrative plates, c is reconstructed coefficients, UcThe preceding c component of representing matrix U, ΣcPreceding c point of representing matrix Σ
Amount,Representing matrix VTPreceding c component.
Step S2-3, reconstruction error is obtained according to collection of illustrative plates is delivered with collection of illustrative plates is rebuild based on pre-defined rule.The meter of reconstruction error ε
Calculate formula as follows：
Step S2-4, judges the size of the reconstruction error and threshold value, if the reconstruction error is less than the threshold value, choosing
The reconstructed coefficients c of step S2-2 is taken, otherwise, return to step S2-2, and increase the value of reconstructed coefficients c.
Step S2-5, whether the low-rank recovery factor c analysis advertisement putting collection of illustrative plates chosen according to step S2-4 is delivered occurs
It is abnormal.
Reconstructed coefficients c illustrates media situation of change of the advertisement when delivering, if media are changed greatly, reconstructed coefficients
The value of c is larger, and otherwise, the value of reconstructed coefficients c is smaller.If in certain advertisement release process, reconstructed coefficients c is very stable always, but
It is that mutation is there occurs at certain time point, then illustrates that the media of advertisement putting there occurs change, and then can determine that dispensing there occurs
It is abnormal.
For example, intercepting t respectively in collection of illustrative plates is delivered0~t1、t0~t2、t0~t3、t0~t4Collection of illustrative plates, according to advertisement putting
Collection of illustrative plates delivers method for detecting abnormality and obtains t1、t2、t3、t4Four reconstructed coefficients c at moment1、c2、c3、c4。c1、c2、c3Approximate phase
Deng, and c4Significantly greater than c1、c2、c3, then t is illustrated3~t4Moment advertisement there occurs abnormal dispensing.
Above-mentioned implementation method is preferred case of the invention, is not intended to limit protection scope of the present invention.
Claims (10)
1. a kind of advertisement putting map construction method, it is characterised in that comprise the following steps：
Step 1, user's registration that any two media are obtained between two media based on pre-defined rule, according to the user
Registration obtains media similarity matrix；
Step 2, projection vector is obtained according to the media similarity matrix based on pre-defined rule；
Step 3, the projection coordinate of each media is obtained according to the projection vector based on pre-defined rule；
Step 4, with the time as abscissa, media are ordinate, and the dispensing index of advertisement is depicted as to deliver collection of illustrative plates.
2. advertisement putting map construction method according to claim 1, it is characterised in that：
Wherein, in step 1, the user that each media obtains occurring on the media according to real time bidding data, and with matrix
Form is represented：
Domain=(user1 user2 … usern)
Domain is mediamatrix, and n is positive integer.
3. advertisement putting map construction method according to claim 2, it is characterised in that：
Wherein, the user's registration between two media is tried to achieve by Jacobi similarity formula：
The media similarity matrix is：
mpqIt is the user's registration between two media, M is p row q column matrix, and p, q are positive integer.
4. advertisement putting map construction method according to claim 1, it is characterised in that：
Wherein, in step 2, the projection vector is tried to achieve based on PCA, and solution formula is as follows：
arg maxμTMμ
s.t.μTμ=1
μ is projection vector, μTIt is the transposed vector of μ, M is media similarity matrix.
5. advertisement putting map construction method according to claim 1, it is characterised in that：
Wherein, in step 3, the projection coordinate is tried to achieve by below equation：
ki=μTdomaini
domaini=(user1 user2 … usern)
kiIt is the projection coordinate of certain media, μ is projection vector, domainiIt is certain mediamatrix.
6. advertisement putting map construction method according to claim 1, it is characterised in that：
Wherein, in step 4, draw dispensing collection of illustrative plates and comprise the steps of：
Step 4-1, calculate under synchronization, the projection coordinate of all media, and each media represents and clicks on index,
Step 4-2, by projection coordinate's identical media under synchronization represent and click on index merge,
Step 4-3, the time is abscissa, and media are ordinate, represents what is obtained in step 4-2 and clicks on index and be plotted in phase
Obtain delivering collection of illustrative plates at the position answered.
7. a kind of advertisement putting collection of illustrative plates delivers method for detecting abnormality, it is characterised in that comprise the steps of：
Step 1, obtained by the advertisement putting map construction method according to any claim in claim 1 to 6
Deliver collection of illustrative plates and do SVD decomposition；
Step 2, chooses reconstructed coefficients, and then carrying out low-rank reconstruction to dispensing collection of illustrative plates obtains rebuilding collection of illustrative plates；
Step 3, reconstruction error is obtained according to collection of illustrative plates is delivered with collection of illustrative plates is rebuild based on pre-defined rule；
Step 4, judges the size of the reconstruction error and threshold value, if the reconstruction error is less than the threshold value, selecting step 2
Reconstructed coefficients, otherwise, return to step 2, and increase the value of the reconstructed coefficients；
Whether step 5, the low-rank recovery factor analysis advertisement putting collection of illustrative plates chosen according to step 4 delivers exception occurs.
8. advertisement putting collection of illustrative plates according to claim 7 delivers method for detecting abnormality, it is characterised in that：
Wherein, in step 1, the formula that the dispensing collection of illustrative plates does SVD decomposition is as follows：
X=U Σ VT
X is to deliver collection of illustrative plates；U is p × p rank unitary matrice；Σ is positive semidefinite p × q rank diagonal matrix；VT, i.e. the conjugate transposition of V is q
× q rank unitary matrice.
9. advertisement putting collection of illustrative plates according to claim 7 delivers method for detecting abnormality, it is characterised in that：
Wherein, in step 2, low-rank reconstruction is carried out to delivering collection of illustrative plates by formula below：
To rebuild collection of illustrative plates, c is reconstructed coefficients, UcThe preceding c component of representing matrix U, ΣcThe preceding c component of representing matrix Σ,Representing matrix VTPreceding c component.
10. advertisement putting collection of illustrative plates according to claim 7 delivers method for detecting abnormality, it is characterised in that：
Wherein, in step 3, the computing formula of reconstruction error is as follows：
ε is reconstruction error, and X is dispensing collection of illustrative plates,To rebuild collection of illustrative plates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710041416.1A CN106815749B (en) | 2017-01-20 | 2017-01-20 | Advertisement putting map construction method and advertisement putting map abnormal putting detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710041416.1A CN106815749B (en) | 2017-01-20 | 2017-01-20 | Advertisement putting map construction method and advertisement putting map abnormal putting detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106815749A true CN106815749A (en) | 2017-06-09 |
| CN106815749B CN106815749B (en) | 2020-09-18 |
Family
ID=59111943
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710041416.1A Active CN106815749B (en) | 2017-01-20 | 2017-01-20 | Advertisement putting map construction method and advertisement putting map abnormal putting detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106815749B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080103883A1 (en) * | 2006-10-25 | 2008-05-01 | Google Inc. | Providing Feedback to an Offer for Advertising Space |
| CN103136696A (en) * | 2013-03-26 | 2013-06-05 | 明日互动（北京）广告传媒有限公司 | Management method of media placement and system thereof |
| CN103679510A (en) * | 2013-12-17 | 2014-03-26 | 深圳市东信时代信息技术有限公司 | Mobile Internet advertisement monitoring method and system |
| CN105787767A (en) * | 2016-03-03 | 2016-07-20 | 上海珍岛信息技术有限公司 | Method and system for obtaining advertisement click-through rate pre-estimation model |
-
2017
- 2017-01-20 CN CN201710041416.1A patent/CN106815749B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080103883A1 (en) * | 2006-10-25 | 2008-05-01 | Google Inc. | Providing Feedback to an Offer for Advertising Space |
| CN103136696A (en) * | 2013-03-26 | 2013-06-05 | 明日互动（北京）广告传媒有限公司 | Management method of media placement and system thereof |
| CN103679510A (en) * | 2013-12-17 | 2014-03-26 | 深圳市东信时代信息技术有限公司 | Mobile Internet advertisement monitoring method and system |
| CN105787767A (en) * | 2016-03-03 | 2016-07-20 | 上海珍岛信息技术有限公司 | Method and system for obtaining advertisement click-through rate pre-estimation model |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN109615442B (en) * | 2019-01-23 | 2021-07-20 | 上海旺翔文化传媒股份有限公司 | RTB real-time bidding method based on incentive video advertisement |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106815749B (en) | 2020-09-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bergquist et al. | Combining climate, economic, and social policy builds public support for climate action in the US | |
| Jimenez | Human and physical infrastructure: Public investment and pricing policies in developing countries | |
| Bracking et al. | Migrant remittances and household wellbeing in urban Zimbabwe | |
| Mohamad et al. | Service quality, customer satisfaction and customer loyalty of the hotel industry in United Arab Emirates (UAE): A measurement model | |
| Chang et al. | Volatility spillovers for spot, futures, and ETF prices in agriculture and energy | |
| Kumar et al. | District-level estimates of fertility and implied sex ratio at birth in India | |
| Sorenson et al. | The extent, nature, and utility of evaluating consumer satisfaction in community mental health centers | |
| WO2019080588A1 (en) | Transaction object matching method and system | |
| Pradana et al. | Gap analysis of Indonesian state-owned bank internet banking website | |
| CN106815749A (en) | Advertisement putting map construction method and advertisement putting collection of illustrative plates deliver detection method extremely | |
| He et al. | A developed framework for the multi-district ecological compensation standards integrating ecosystem service zoning in an urban area in China | |
| Makarenko et al. | Investments support for Sustainable Development Goal 7: Research gaps in the context of post-COVID-19 recovery | |
| Li et al. | Consumers’ willingness to pay for the solar photovoltaic system in the post-subsidy era: a comparative analysis under an urban-rural divide | |
| WO2019080585A1 (en) | Method and system for scoring value of social circle | |
| Fedushko et al. | E-Commerce and E-Health Strategies and Implementation Activities in the United Kingdom: Review Study | |
| Zheng et al. | Are nutrition labels useful for the purchase of a familiar food? Evidence from Chinese consumers’ purchase of rice | |
| CN105956914A (en) | Option type auction data processing method and system | |
| Lee et al. | The effect of privacy concerns and consumer trust on consumer response to online behavioral advertising for travel products | |
| Hussain et al. | The use of high-speed internet as a platform for compulsive online buying: A case of study of post-graduate students in Malaysia | |
| CN106875168A (en) | A kind of big data value classifying method based on block chain | |
| Candra et al. | Analysis of Consumer Risk Perception on Online Auction Features | |
| Wang et al. | Credit allocation for research institutes | |
| Xu et al. | Consumers’ attitudes of e-commerce in China | |
| Shagari et al. | The impact of board autonomy and information and communication technology on tax administration efficiency: Evidence from Nigeria | |
| Al-Guthmy et al. | Financial implications of mobile phone-based personal carbon trading: a case study of Safaricom |
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 |