CN102346899A - Method and device for predicting advertisement click rate based on user behaviors - Google Patents
Method and device for predicting advertisement click rate based on user behaviors Download PDFInfo
- Publication number
- CN102346899A CN102346899A CN201110302456XA CN201110302456A CN102346899A CN 102346899 A CN102346899 A CN 102346899A CN 201110302456X A CN201110302456X A CN 201110302456XA CN 201110302456 A CN201110302456 A CN 201110302456A CN 102346899 A CN102346899 A CN 102346899A
- Authority
- CN
- China
- Prior art keywords
- user
- classification
- advertisement
- clicking rate
- predicted value
- 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.)
- Pending
Links
Images
Abstract
The invention discloses a method and device for predicting advertisement click rate based on user behaviors. The method comprises the following steps of: behavior directional prediction: training a behavior directional model by analyzing the user behaviors, and predicting the click rates of the user on different types of advertisements to obtain a predicted value of the click rate; in-type sorting: sorting the advertisements of each type according to the current behavior type of the user to obtain an in-type advertisement sorting list; and comprehensive sorting: sorting the click rates of all advertisements according to the predicted value of click rate and the in-type advertisement sorting list to obtain a comprehensive sorting list. Through the method disclosed by the invention, a prediction model can be established for each type of advertisements so as to sort the advertisements more accurately according to the type characteristics of the advertisements.
Description
Technical field
The present invention relates to ad click rate prediction, user behavior analysis, regression model, the network browsing or the search behavior that are specifically related to attempt through analysis user are carried out modeling analysis to its behavior, thereby predict its clicking rate to advertisement delivery.
Background technology
The rise of internet makes people can when browsing same page, see different advertisement, realizes that the personalization of advertisement is showed.In this way, can show the interested advertisement of its possibility more accurately to each user, thereby improve the clicking rate of advertisement, improve advertisement delivery effect.
Existing advertisement accurately is thrown in strategy and is mainly comprised following three types: search trigger (sponsored search), content match (content match), behavioral targeting (behavioral targeting).
Wherein, the search trigger strategy comprises that the keyword of submitting to search engine according to the user carries out the advertisement retrieval, because keyword has directly reflected the interest that the user is current, so can push the advertisement relevant with the current search content to the user.
The content match strategy then comprises the content of the webpage of user's browsing is carried out modeling analysis, shows the advertisement close with web page contents to the user.
The behavioral targeting strategy comprises that the historical behavior record according to the user carries out modeling analysis to its interest, shows the advertisement that meets its interest characteristics to the user.When different user was browsed the same page or submitted identical query requests to, the advertisement of being seen also was different.
Two types of technology of search trigger and content match and consideration user's useless personalized interest; When different user is searched for same keyword or is browsed same webpage; The advertisement of being showed is identical often; And in practical application; When two users browse the same page or carry out same inquiry, its focus maybe and inequality.Preference in the then less mass advertising of consideration user in same classification of behavioral targeting method, promptly same user may represent identical advertisement when browsing the different page.Therefore, advertisement putting strategy of the prior art is not considered user personalized interest preferably, and the user that can not comparatively calculate to a nicety is to the clicking rate of advertisement, thereby can not throw in the user's interest advertisement accurately.
Summary of the invention
In order to solve the problems of the technologies described above; The invention provides a kind of two stage ad click rate Forecasting Methodology and systems that the behavioral targeting method is combined with content match/search trigger method based on user behavior; Purpose is that the more convenient location of help advertiser maybe be to its advertisement users interest colony; To user's advertisement delivery the time, take into full account simultaneously user's current browsing or the point of interest that the retrieval behavior is reflected; To strengthen advertisement delivery effect; The present invention combines the behavioral targeting method with content match/search trigger method, to make full use of the advantage of distinct methods.
The invention provides a kind of ad click rate Forecasting Methodology based on user behavior; It is characterized in that; May further comprise the steps: the behavioral targeting prediction steps; Train the behavioral targeting model through the analysis user behavior; According to the clicking rate of the said user of said behavioral targeting model prediction, obtain the clicking rate predicted value to different classes of advertisement; Classification internal sort step to the order ads in each classification, obtains order ads tabulation in the classification according to said user's current behavior type; Integrated ordered step sorts based on the clicking rate of order ads tabulation in said clicking rate predicted value and the said classification to all advertisements, obtains integrated ordered tabulation.
Further, this method comprises that also said behavioral targeting prediction steps specifically may further comprise the steps: steps A, obtain the relevant user's characteristic information of behavior of clicking advertisement with said user; Step B; Obtain user behavior information and fusion feature information respectively according to said user's characteristic information; Obtain user behavior information and fusion feature information respectively; Wherein said user behavior information is and the relevant information of user's interest advertisement classification that said fusion feature information is for to merge the information that obtains with each category feature in the said user's characteristic information; Step C is based on said user behavior information and said fusion feature information training behavioral targeting model; Step D utilizes the clicking rate of the said user of said behavioral targeting model prediction to different classes of advertisement, obtains said clicking rate predicted value.
Further, this method comprises that also said classification internal sort step specifically may further comprise the steps: step e, according to said clicking rate predicted value obtain said user the advertising listing of interested each classification; Step F, the behavior type current based on said user sorts to the advertisement in the advertising listing of said each classification respectively, obtains order ads tabulation in the said classification.
Further, this method comprises that also said integrated ordered step further comprises: according to said integrated ordered tabulation, show the highest advertisement of clicking rate of predetermined quantity to said user.
Further, this method also comprises, comprises among the said step C and shows the forecast model training and click the forecast model training, to obtain clicking the weight matrix and the weight matrix of showing forecast model of forecast model.
Further, this method comprises that also said click forecast model training may further comprise the steps: step C11, other weight matrix of initialization commercial paper; Step C12 calculates the number of clicks predicted value of said user to each classification advertisement based on said weight matrix; Step C13 according to said number of clicks predicted value, upgrades said weight matrix; Step C14, repeating step C12 and C13 carry out iterative processing, until process convergence or reach predetermined iterations, obtain the weight matrix of said click forecast model.
Further, this method comprises that also said displaying forecast model training may further comprise the steps: step C21, other weight matrix of initialization commercial paper; Step C22 calculates the displaying number of times predicted value of said user to each classification advertisement based on said weight matrix; Step C23 according to said displaying number of times predicted value, upgrades said weight matrix; Step C24, repeating step C22 and C23 carry out iterative processing, until process convergence or reach predetermined iterations, obtain the weight matrix of said displaying forecast model.
Further; This method also comprises; In said step D; Weight matrix according to said click forecast model calculates click prediction number of times; According to the weight matrix calculations show prediction number of times of said displaying forecast model, obtain said clicking rate predicted value based on said click prediction number of times and said displaying prediction number of times.
Further, this method comprises that also in said step F, said user's current behavior type is searched for or browsing page for utilizing search engine; If the current behavior of said user is to utilize search engine search, then adopt the search trigger model that advertisement in the classification is sorted by clicking rate according to the query word that said user submitted to; If the current behavior of said user is a browsing page,, select for use the content match model that advertisement in the classification is sorted by clicking rate then according to the web page contents that said user browsed.
Further, this method also comprises, in said integrated ordered step, the clicking rate value in said clicking rate predicted value and the tabulation of said order ads is carried out product, linear weighted function or nonlinear combination, obtains said integrated ordered tabulation.
The present invention also provides a kind of ad click rate prediction unit based on user behavior; Said device comprises: the behavioral targeting predicting unit; Be used for training the behavioral targeting model through the analysis user behavior; According to the clicking rate of the said user of said behavioral targeting model prediction, obtain the clicking rate predicted value to different classes of advertisement; Classification internal sort unit is used for according to said user's current behavior type the order ads in each classification, obtains order ads tabulation in the classification; Integrated ordered unit is used for sorting based on the clicking rate of order ads tabulation in said clicking rate predicted value and the said classification to all advertisements, obtains integrated ordered tabulation.
Compared with prior art, the present invention has the following advantages at least:
The present invention through on the basis of behavior orientation method, introduce the second step user to classification in the click prediction of concrete ad entry; Make method of the present invention set up forecast model, thereby advertisement is sorted more accurately according to the classification characteristics of advertisement to the advertisement of each classification.
Other features and advantages of the present invention will be set forth in instructions subsequently, and, partly from instructions, become apparent, perhaps understand through embodiment of the present invention.The object of the invention can be realized through the structure that in instructions, claims and accompanying drawing, is particularly pointed out and obtained with other advantages.
Description of drawings
Accompanying drawing is used to provide further understanding of the present invention, and constitutes the part of instructions, is used to explain the present invention with embodiments of the invention, is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the process flow diagram based on the ad click rate Forecasting Methodology of user behavior according to the embodiment of the invention one;
Fig. 2 is the process flow diagram according to the training of the embodiment of the invention one and prediction behavior recursive model;
Fig. 3 is the process flow diagram according to order ads in the class of utilizing the search trigger model of the embodiment of the invention one;
Fig. 4 is the ad click rate prediction unit structured flowchart based on user behavior according to the embodiment of the invention two.
Embodiment
Below will combine accompanying drawing and embodiment to describe embodiment of the present invention in detail, how the application technology means solve technical matters to the present invention whereby, and the implementation procedure of reaching technique effect can make much of and implement according to this.Need to prove that only otherwise constitute conflict, each embodiment among the present invention and each characteristic among each embodiment can mutually combine, formed technical scheme is all within protection scope of the present invention.
In addition; Can in computer system, carry out in the step shown in the process flow diagram of accompanying drawing such as a set of computer-executable instructions, and, though logical order has been shown in process flow diagram; But in some cases, can carry out step shown or that describe with the order that is different from here.
Embodiment one
Fig. 1 shows the process flow diagram based on the ad click rate Forecasting Methodology of subscriber network access behavior (abbreviating user behavior among this paper as) according to the embodiment of the invention one, describes the detailed step of present embodiment method below in detail with reference to Fig. 1.
The method of present embodiment totally comprised for two stages, and as shown in fig. 1, the phase one of part is the behavioral targeting stage shown in the fine line, and the subordinate phase shown in the heavy line is to utilize the classification internal sort stage of content match/search trigger method.
Phase one specifically may further comprise the steps:
Step S110 obtains the relevant user's characteristic information of behavior of clicking advertisement with the user.
Concrete; According to user's historical search message, user to webpage browse message and the previous advertisement message of having clicked or having showed of user, obtain user's user's characteristic information such as query word characteristic, website visiting characteristic and webpage correlated characteristic, advertisement correlated characteristic respectively.
Correlated characteristic when wherein query word is characterized as the user and utilized search engine to retrieve in the past, as carry out relevant word frequency TF or the word frequency-contrary document frequency TF*IDF characteristic of query word behind the participle, statistical natures such as the average length of the query word that uses, number etc.
The website visiting characteristic comprises that the user is to statistical natures such as the access times of different web sites, the residence time.
The webpage correlated characteristic comprises user's represented characteristic set of content of the webpage of browsing in the past, TF, the TF*IDF characteristic relevant like content, the classification of institute's browsing page, the residence time etc.
The advertisement correlated characteristic comprises displaying number of times, the number of clicks of user to dissimilar advertisements, and the user clicks/showed the content characteristic of landing page landing page itself of the speech of bidding, content description correlated characteristic and the advertisement of advertisement.
Step S120 according to the user's characteristic information of obtaining, obtains user behavior information and fusion feature information respectively.
Concrete; According to the user's that step S110 obtained query word characteristic, website visiting characteristic and webpage correlated characteristic, advertisement correlated characteristic etc.; Behavior to the user is analyzed; To obtain user behavior information; For example; The interested advertisement classification of analysis user, with information that will be relevant with user's interest advertisement classification as user behavior information.Different character in the user's characteristic information of obtaining is merged to obtain fusion feature information; Wherein, Feature Fusion is meant user-dependent each category feature is merged; User behavior being described from each different angles, as the user used keyword, the list of websites of visiting, the webpage correlation properties of visiting, user's interest advertisement classification that behavioural analysis obtains etc.Wherein keyword, website and webpage correlated characteristic can directly obtain from message or daily record, and user's interest advertisement classification can't directly obtain from related data, need from message, obtain through analyzing by the user behavior analysis method.
Step S130 is based on user behavior information and fusion feature information training behavioral targeting model.On the basis of the user behavior information that user behavior analysis obtained, utilize the fusion feature information that Feature Fusion obtained, make up the fusion feature vector, be desired value with the user to the click data of each classification advertisement, train the behavioral targeting model.
The regression model that the selecting of behavioral targeting model used always, for example Poisson regression model, support vector machine SVM model, this special model of logic and linear regression model (LRM) etc.In the present embodiment, be that example describes only with the Poisson regression model.With reference to Fig. 2, specify the process of utilizing Poisson regression model training behavioral targeting model.Shown in Fig. 2 the first half, the model training process can be divided into the displaying forecast model training of advertisement and click forecast model training two parts.
In clicking the forecast model training process; The user's that obtains in the present embodiment fusion feature vector is designated as x; The user is designated as y to the number of clicks vector of each classification advertisement, and the weight matrix of clicking the Poisson model of prediction is designated as ω, and the model training process comprises following substep:
S131, initialization step is according to following other weight matrix of formula initialization commercial paper.
Wherein, ω
KjThe weight of expression advertisement classification k on the j dimensional feature space, y
IkThe number of times of the classification k advertisement that expression user i was clicked, x
IjBe the eigenwert of user i on the j dimensional feature space.x
Ij 'The eigenwert of the j ' dimension of expression user i.The ω that calculates
K, jConstituted weight matrix ω.
Step S132 concentrates the number of clicks predicted value of user to each classification advertisement based on the weight matrix calculation training.
According to weight matrix ω, calculation training is concentrated the number of clicks predicted value of user to each classification advertisement, and formula is following:
λ=ω
Tx (2)
Wherein λ is a Parameter for Poisson Distribution, and promptly the user is to the predicted value of the number of clicks of advertisement.
Step S133 according to calculating the number of clicks predicted value, upgrades weight matrix, and concrete formula is following:
ω ' wherein
KjThe weight of classification k on the j dimensional feature after expression is upgraded, λ
IkThe user i that expression obtains according to the last iteration result is to the predicted value of classification k number of clicks.
Step S134, repeating step S132 and S133 carry out iterative processing, until process convergence or reach predetermined iterations, obtain clicking the weight matrix of forecast model.
The difference of twice resulting weight matrix ω of iteration was less than specific threshold value σ before and after wherein the process convergence was meant.Thus, accomplished click forecast model training process.
The training process that forecast model is showed in advertisement is with to click the forecast model process approximate, just desired value is used be in the training set user to the displaying number of times of different classes of advertisement (corresponding to the y that clicks in the forecast model process
Ik), be not described further at this.
Thus, accomplished training to the behavioral targeting model.
Step S140 utilizes the clicking rate of behavioral targeting model prediction user to different classes of advertisement, obtains the clicking rate predicted value.
Utilize the behavioral targeting model that follow-up user browsing behavior is predicted, concrete, predictive user is to the clicking rate value of different classes of advertisement.Wherein, follow-up user browsing behavior can be that previous new behavior of browsing that the user initiated or new Client-initiated browsed behavior.
Concrete, after training obtains the click forecast model of advertisement and shows forecast model, can be used for new user's clicking rate prediction, process is shown in Fig. 2 the latter half, and the proper vector of supposing new user is x ',
1) the ad click forecast model that utilizes training to obtain calculates the number of clicks of this user to each classification advertisement, promptly
λ
c=ω
cx′ (4)
Wherein, ω
cFor clicking the weight matrix of forecast model, λ
cFor clicking the prediction number of times.
2) advertisement that utilizes training to obtain shows that forecast model calculates the displaying number of times of this user to each classification advertisement, promptly
λ
v=ω
vx′ (5)
Wherein, ω
vFor showing the weight matrix of forecast model, λ
vFor showing the prediction number of times.
3) the final clicking rate predicted value of calculating is following:
CTRx=(λ
c+α)/(λ
v+β)(6)
Wherein CTRx is the clicking rate predicted value, and α, β are the smoothing parameter that is provided with in advance, is 0 situation to avoid CTRx.
Above step S110 to S140 is the detailed process in the behavioral targeting stage of present embodiment phase one, and through the phase one, the user who obtains being predicted through the behavioral targeting model is to the clicking rate predicted value of different classes of advertisement.
After the phase one, get into subordinate phase, that is, and content match/search trigger stage.Subordinate phase specifically may further comprise the steps:
Step S150, according to the resulting clicking rate predicted value of behavioral targeting model obtain the user the advertising listing of interested each classification.
Step S160, the behavior type current based on the user sorts to the advertisement in the advertising listing in each classification, obtains the order ads tabulation.
Concrete, according to user's current behavior type selecting proper model the advertisement in each classification is sorted.Just utilizing search engine etc. to search for if the user is current, then adopting search trigger sponsored search model that advertisement in the classification is sorted by clicking rate according to the query word that the user submitted to.If the current positive browsing page of user then according to the web page contents that the user browsed, selects for use content match Content Match model that advertisement in the classification is sorted by clicking rate.Obtain the order ads tabulation through ordering.
With the search trigger model is example, and concrete model can be selected the Bayesian regression model, and idiographic flow comprises following substep as shown in Figure 3:
Step S161 clicks situation according to the history of advertisement in this classification, makes up training set, carries out model training.
Historical click situation refers to that the user is to the click statistics of advertisement in this classification in the actual advertisement launch process; The training set schedule of samples is shown Xi=(xi; Yi) wherein xi is the proper vector of sample i; Yi is the desired value of sample i; Each sample is a user search-advertisement click events; Yi=1 representes that this incident takes place, and the yi=0 presentation of events does not take place.The affair character vector comprises three parts, i.e. query word correlated characteristic, contextual information and characteristic of advertisement.Wherein the query word correlated characteristic comprises user's expansion entry that employed query word, query word are correlated with in this incident etc.; Contextual information then comprises user position information, incident, user's search history record etc.; Characteristic of advertisement comprises the landing page landing page content information etc. of keyword, ad title, descriptor and the advertisement of advertisement.
Step S162, according to the training set that last step makes up, training Bayesian regression model, wherein the training process of Bayesian regression model can adopt the correlation method of the prior art that those skilled in the art know, and repeats no more at this.
Step S163 to new user's search behavior, copies step S161 to make up the proper vector of user-advertised event.
Step S164, the proper vector constructed to step S163, the probability that this incident of Bayesian regression model prediction of utilizing training to obtain takes place is as the click probability of user to certain bar advertisement.
Step S165 to all advertisements in certain advertisement classification, calculates user-ad click probability respectively, and with this advertisement is sorted, thereby realizes order ads in the classification, obtains order ads tabulation in the classification.
According to above step,, obtain order ads tabulation in the classification based on user behavior type selecting different model.
Step S170 sorts based on order ads tabulation in the classification and the clicking rate predicted value clicking rate to all advertisements, obtains integrated ordered tabulation, shows the advertisement of the predetermined quantity that the clicking rate of predetermined quantity is the highest to the user.
In this step; The user that the user that need obtain according to phase one step S140 obtains other clicking rate predicted value of each commercial paper and subordinate phase step S160 is to clicking rate ranking results two parts content of advertisement in each classification; Calculate the clicking rate of user to every relevant advertisements; And successively the relevant advertisements of all categories is carried out the integral body ordering from high to low by clicking rate; Obtain integrated ordered tabulation; User side is given in the advertisement pushing of earlier predetermined quantity of will sorting at last, shows to the user.
Concrete, suppose advertisement a
jAffiliated classification is c
k, the user is to c
kThe clicking rate predicted value of classification is CTR
Ck, and when in this classification, carrying out in the classification order ads, the user is CTR to the clicking rate predicted value of this advertisement
Aj, when then the user carries out current search or browses behavior, click advertisement a
jClicking rate predicted value CTR can be expressed as CTR
CkWith CTR
AjCombination, combinatorial formula can be chosen arbitrary forms such as product, linear weighted function and nonlinear combination.
The product combination formula may be calculated:
CTR=CTR
ck*CTR
aj (7)
The linear weighted function combinatorial formula may be calculated:
CTR=γ*CTR
ck+(1-γ)CTR
aj?(8)
Wherein γ is predefined weight pondage factor.
Thus; Present embodiment has obtained embodying the clicking rate predicted value CTR of the user of user interest to advertisement, thereby can sort to advertisement according to predicted value through two phase process; The advertisement that clicking rate is higher is showed to the user, makes the input of advertisement more accurate.Present embodiment can be set up forecast model to the advertisement of each classification, thereby according to the classification characteristics of advertisement advertisement is sorted more accurately.
Embodiment two
Fig. 4 is the ad click rate prediction unit structured flowchart based on user behavior according to the embodiment of the invention two, and the composition of this device is described according to Fig. 4 below.
This device comprises behavioral targeting predicting unit, classification internal sort unit, integrated ordered unit.
The behavioral targeting predicting unit is used for training the behavioral targeting model through the analysis user behavior, according to the clicking rate of behavioral targeting model prediction user to different classes of advertisement, obtains the clicking rate predicted value.
Classification internal sort unit is used for according to user's current behavior type the order ads in each classification, obtains the order ads tabulation in the classification.
Integrated ordered unit is used for sorting based on the clicking rate of clicking rate predicted value and said order ads tabulation to all advertisements, obtains integrated ordered tabulation.
Those skilled in the art should be understood that; Above-mentioned each module of the present invention or each step can realize with the general calculation device; They can concentrate on the single calculation element; Perhaps be distributed on the network that a plurality of calculation element forms; Alternatively; They can be realized with the executable program code of calculation element; Thereby; They can be stored in the memory storage and carry out by calculation element; Perhaps they are made into each integrated circuit modules respectively, perhaps a plurality of modules in them or step are made into the single integrated circuit module and realize.Like this, the present invention is not restricted to any specific hardware and software combination.
Though the embodiment that the present invention disclosed as above, the embodiment that described content just adopts for the ease of understanding the present invention is not in order to limit the present invention.Technician under any the present invention in the technical field; Under the prerequisite of the spirit and scope that do not break away from the present invention and disclosed; Can do any modification and variation what implement in form and on the details; But scope of patent protection of the present invention still must be as the criterion with the scope that appending claims was defined.
Claims (11)
1. the ad click rate Forecasting Methodology based on user behavior is characterized in that, may further comprise the steps:
The behavioral targeting prediction steps is trained the behavioral targeting model through the analysis user behavior, according to the clicking rate of the said user of said behavioral targeting model prediction to different classes of advertisement, obtains the clicking rate predicted value;
Classification internal sort step to the order ads in each classification, obtains order ads tabulation in the classification according to said user's current behavior type;
Integrated ordered step sorts based on the clicking rate of order ads tabulation in said clicking rate predicted value and the said classification to all advertisements, obtains integrated ordered tabulation.
2. method according to claim 1 is characterized in that, said behavioral targeting prediction steps specifically may further comprise the steps:
Steps A is obtained the relevant user's characteristic information of behavior of clicking advertisement with said user;
Step B; Obtain user behavior information and fusion feature information respectively according to said user's characteristic information; Wherein said user behavior information is and the relevant information of user's interest advertisement classification that said fusion feature information is for to merge the information that obtains with each category feature in the said user's characteristic information;
Step C is based on said user behavior information and said fusion feature information training behavioral targeting model;
Step D utilizes the clicking rate of the said user of said behavioral targeting model prediction to different classes of advertisement, obtains said clicking rate predicted value.
3. method according to claim 1 and 2 is characterized in that, said classification internal sort step specifically may further comprise the steps:
Step e, according to said clicking rate predicted value obtain said user the advertising listing of interested each classification;
Step F, the behavior type current based on said user sorts to the advertisement in the advertising listing of said each classification respectively, obtains order ads tabulation in the said classification.
4. method according to claim 1 and 2 is characterized in that, said integrated ordered step further comprises:
According to said integrated ordered tabulation, show the highest advertisement of clicking rate of predetermined quantity to said user.
5. method according to claim 2 is characterized in that, comprises among the said step C showing the forecast model training and clicking the forecast model training, to obtain clicking the weight matrix and the weight matrix of showing forecast model of forecast model.
6. method according to claim 5 is characterized in that, said click forecast model training may further comprise the steps:
Step C11, other weight matrix of initialization commercial paper;
Step C12 calculates the number of clicks predicted value of said user to each classification advertisement based on said weight matrix;
Step C13 according to said number of clicks predicted value, upgrades said weight matrix;
Step C14, repeating step C12 and C13 carry out iterative processing, until process convergence or reach predetermined iterations, obtain the weight matrix of said click forecast model.
7. method according to claim 5 is characterized in that, said displaying forecast model training may further comprise the steps:
Step C21, other weight matrix of initialization commercial paper;
Step C22 calculates the displaying number of times predicted value of said user to each classification advertisement based on said weight matrix;
Step C23 according to said displaying number of times predicted value, upgrades said weight matrix;
Step C24, repeating step C22 and C23 carry out iterative processing, until process convergence or reach predetermined iterations, obtain the weight matrix of said displaying forecast model.
8. according to each described method of claim 5 to 7; It is characterized in that; In said step D; Weight matrix according to said click forecast model calculates click prediction number of times; According to the weight matrix calculations show prediction number of times of said displaying forecast model, obtain said clicking rate predicted value based on said click prediction number of times and said displaying prediction number of times.
9. method according to claim 3 is characterized in that, in said step F, said user's current behavior type is searched for or browsing page for utilizing search engine;
If the current behavior of said user is to utilize search engine search, then adopt the search trigger model that advertisement in the classification is sorted by clicking rate according to the query word that said user submitted to;
If the current behavior of said user is a browsing page,, select for use the content match model that advertisement in the classification is sorted by clicking rate then according to the web page contents that said user browsed.
10. method according to claim 1; It is characterized in that; In said integrated ordered step, the clicking rate value in said clicking rate predicted value and the tabulation of said order ads is carried out product, linear weighted function or nonlinear combination, obtain said integrated ordered tabulation.
11. the ad click rate prediction unit based on user behavior, said device comprises:
The behavioral targeting predicting unit is used for training the behavioral targeting model through the analysis user behavior, according to the clicking rate of the said user of said behavioral targeting model prediction to different classes of advertisement, obtains the clicking rate predicted value;
Classification internal sort unit is used for according to said user's current behavior type the order ads in each classification, obtains order ads tabulation in the classification;
Integrated ordered unit is used for sorting based on the clicking rate of order ads tabulation in said clicking rate predicted value and the said classification to all advertisements, obtains integrated ordered tabulation.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110302456XA CN102346899A (en) | 2011-10-08 | 2011-10-08 | Method and device for predicting advertisement click rate based on user behaviors |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201110302456XA CN102346899A (en) | 2011-10-08 | 2011-10-08 | Method and device for predicting advertisement click rate based on user behaviors |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102346899A true CN102346899A (en) | 2012-02-08 |
Family
ID=45545549
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201110302456XA Pending CN102346899A (en) | 2011-10-08 | 2011-10-08 | Method and device for predicting advertisement click rate based on user behaviors |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102346899A (en) |
Cited By (96)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102663617A (en) * | 2012-03-20 | 2012-09-12 | 亿赞普(北京)科技有限公司 | Method and system for prediction of advertisement clicking rate |
CN102750334A (en) * | 2012-06-01 | 2012-10-24 | 北京市农林科学院农业科技信息研究所 | Agricultural information accurate propelling method based on data mining (DM) |
CN102779193A (en) * | 2012-07-16 | 2012-11-14 | 哈尔滨工业大学 | Self-adaptive personalized information retrieval system and method |
CN102880688A (en) * | 2012-09-14 | 2013-01-16 | 北京百度网讯科技有限公司 | Method, device and equipment for evaluating webpage |
CN103235893A (en) * | 2013-05-06 | 2013-08-07 | 重庆大学 | User-commodity click rate self-adaptive prediction device and method |
CN103310003A (en) * | 2013-06-28 | 2013-09-18 | 华东师范大学 | Method and system for predicting click rate of new advertisement based on click log |
CN103345512A (en) * | 2013-07-06 | 2013-10-09 | 北京品友互动信息技术有限公司 | Online advertising click-through rate forecasting method and device based on user attribute |
CN103377435A (en) * | 2012-04-26 | 2013-10-30 | 飞拓无限信息技术(北京)有限公司 | Method and system for achieving mobile advertisement recommendation |
CN103489117A (en) * | 2012-06-12 | 2014-01-01 | 深圳市腾讯计算机系统有限公司 | Method and system for information releasing |
CN103514178A (en) * | 2012-06-18 | 2014-01-15 | 阿里巴巴集团控股有限公司 | Searching and sorting method and device based on click rate |
CN103714084A (en) * | 2012-10-08 | 2014-04-09 | 腾讯科技(深圳)有限公司 | Method and device for recommending information |
CN103810239A (en) * | 2013-11-19 | 2014-05-21 | 北京国双科技有限公司 | Statistical method and device for conversion data |
CN103853711A (en) * | 2012-11-28 | 2014-06-11 | 中国移动通信集团广西有限公司 | Text information processing method and device |
CN103955842A (en) * | 2014-04-14 | 2014-07-30 | 武汉烽火普天信息技术有限公司 | Online advertisement recommending system and method for large-scale medium data |
CN104090919A (en) * | 2014-06-16 | 2014-10-08 | 华为技术有限公司 | Advertisement recommending method and advertisement recommending server |
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 |
CN104281547A (en) * | 2013-07-02 | 2015-01-14 | 罗伯特·博世有限公司 | Method for operating a control unit and a control unit having a model calculation unit |
CN104424291A (en) * | 2013-09-02 | 2015-03-18 | 阿里巴巴集团控股有限公司 | Method and device for sorting search results |
CN104484372A (en) * | 2014-12-04 | 2015-04-01 | 北京奇虎科技有限公司 | Detecting method and device of business object sending information |
CN104503980A (en) * | 2014-11-26 | 2015-04-08 | 百度在线网络技术(北京)有限公司 | Method for determining comprehensive search information and for determining alternative search sequences to be pushed by comprehensive search information |
CN104537502A (en) * | 2015-01-15 | 2015-04-22 | 北京嘀嘀无限科技发展有限公司 | Method and device for processing orders |
CN104574160A (en) * | 2015-01-29 | 2015-04-29 | 深圳市英威诺科技有限公司 | Smooth advertisement traffic control method |
CN104679743A (en) * | 2013-11-26 | 2015-06-03 | 阿里巴巴集团控股有限公司 | Method and device for determining preference model of user |
WO2015081801A1 (en) * | 2013-12-06 | 2015-06-11 | Tencent Technology (Shenzhen) Company Limited | Method, server, and system for information push |
CN104750713A (en) * | 2013-12-27 | 2015-07-01 | 阿里巴巴集团控股有限公司 | Method and device for sorting search results |
CN104834641A (en) * | 2014-02-11 | 2015-08-12 | 腾讯科技(北京)有限公司 | Processing method of network medium information and relevant system |
CN104866474A (en) * | 2014-02-20 | 2015-08-26 | 阿里巴巴集团控股有限公司 | Personalized data searching method and device |
CN104883585A (en) * | 2014-02-28 | 2015-09-02 | 腾讯科技(北京)有限公司 | Method for displaying media data, equipment and system thereof |
CN104933603A (en) * | 2015-06-30 | 2015-09-23 | 北京奇艺世纪科技有限公司 | A method and an apparatus for advertisement displaying |
WO2015149307A1 (en) * | 2014-04-02 | 2015-10-08 | Google Inc. | Systems and methods for optimizing content layout using behavior metrics |
CN104992348A (en) * | 2015-06-24 | 2015-10-21 | 深圳市腾讯计算机系统有限公司 | Method and device for displaying information |
CN105117491A (en) * | 2015-09-22 | 2015-12-02 | 北京百度网讯科技有限公司 | Page pushing method and device |
CN105183772A (en) * | 2015-08-07 | 2015-12-23 | 百度在线网络技术(北京)有限公司 | Release information click rate estimation method and apparatus |
CN105184321A (en) * | 2015-09-10 | 2015-12-23 | 北京金山安全软件有限公司 | Data processing method and device for ftrl model |
CN105260477A (en) * | 2015-11-06 | 2016-01-20 | 北京金山安全软件有限公司 | Information pushing method and device |
CN105408928A (en) * | 2013-03-15 | 2016-03-16 | 万歌有限公司 | Methods and systems for determining relevance of advertising categories for devices |
CN105447045A (en) * | 2014-09-02 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Information ordering method and device and information providing method and system |
CN105631711A (en) * | 2015-12-30 | 2016-06-01 | 合一网络技术(北京)有限公司 | Advertisement putting method and apparatus |
CN105630185A (en) * | 2014-11-25 | 2016-06-01 | 奥多比公司 | Temporal dynamics in display advertising prediction |
CN105678587A (en) * | 2016-01-12 | 2016-06-15 | 腾讯科技(深圳)有限公司 | Recommendation feature determination method and information recommendation method and device |
CN105701191A (en) * | 2016-01-08 | 2016-06-22 | 腾讯科技(深圳)有限公司 | Push information click rate estimation method and device |
CN105812844A (en) * | 2014-12-29 | 2016-07-27 | 深圳市Tcl高新技术开发有限公司 | User advertisement push method and system of TV |
CN105956888A (en) * | 2016-05-31 | 2016-09-21 | 北京创意魔方广告有限公司 | Advertisement personalized display method |
CN105989004A (en) * | 2015-01-27 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Information releasing pretreatment method and device |
CN106156878A (en) * | 2015-04-21 | 2016-11-23 | 深圳市腾讯计算机系统有限公司 | Ad click rate antidote and device |
CN106203668A (en) * | 2015-05-07 | 2016-12-07 | 北京极迅智程网络科技股份有限公司 | A kind of system being predicted information index analyzing and showing |
CN106204122A (en) * | 2016-07-05 | 2016-12-07 | 北京京东尚科信息技术有限公司 | Contact measure of value method and apparatus |
CN106355442A (en) * | 2016-08-30 | 2017-01-25 | 杭州启冠网络技术有限公司 | Online precise advertising method and system based on big data driving |
WO2017028728A1 (en) * | 2015-08-18 | 2017-02-23 | 北京金山安全软件有限公司 | Determining method and device for click through rate (ctr) |
CN106504011A (en) * | 2015-09-06 | 2017-03-15 | 阿里巴巴集团控股有限公司 | A kind of methods of exhibiting of business object and device |
CN106557475A (en) * | 2015-09-24 | 2017-04-05 | 北京国双科技有限公司 | Path system of selection and device |
CN106688215A (en) * | 2014-06-27 | 2017-05-17 | 谷歌公司 | Automated click type selection for content performance optimization |
CN106780273A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | Passenger flight requirement analysis method and system |
CN106886915A (en) * | 2017-01-17 | 2017-06-23 | 华南理工大学 | A kind of ad click predictor method based on time decay sampling |
CN106897892A (en) * | 2015-12-18 | 2017-06-27 | 北京奇虎科技有限公司 | Advertisement placement method and device |
CN106933825A (en) * | 2015-12-29 | 2017-07-07 | 北京国双科技有限公司 | Website data processing method and processing device |
WO2017121258A1 (en) * | 2016-01-12 | 2017-07-20 | 腾讯科技(深圳)有限公司 | Information delivery method and apparatus, server, and storage medium |
CN107545301A (en) * | 2016-06-23 | 2018-01-05 | 阿里巴巴集团控股有限公司 | Page display method and device |
CN107613022A (en) * | 2017-10-20 | 2018-01-19 | 广州优视网络科技有限公司 | Content delivery method, device and computer equipment |
CN108062684A (en) * | 2017-12-12 | 2018-05-22 | 北京奇艺世纪科技有限公司 | The clicking rate Forecasting Methodology and device of a kind of advertisement |
CN108229994A (en) * | 2016-12-21 | 2018-06-29 | 北京金山云网络技术有限公司 | A kind of information-pushing method and device |
CN108229987A (en) * | 2016-12-14 | 2018-06-29 | 腾讯科技(深圳)有限公司 | A kind of advertisement inventory evaluation method and device |
CN108369433A (en) * | 2015-09-18 | 2018-08-03 | Mms美国控股有限公司 | Micro- moment analysis |
CN108388624A (en) * | 2018-02-12 | 2018-08-10 | 科大讯飞股份有限公司 | Multimedia messages recommend method and device |
CN108446374A (en) * | 2018-03-16 | 2018-08-24 | 北京三快在线科技有限公司 | User view prediction technique, device, electronic equipment, storage medium |
CN108647691A (en) * | 2018-03-12 | 2018-10-12 | 杭州电子科技大学 | A kind of image classification method based on click feature prediction |
CN108876422A (en) * | 2017-05-09 | 2018-11-23 | 北京京东尚科信息技术有限公司 | For the method, apparatus of information popularization, electronic equipment and computer-readable medium |
CN108965951A (en) * | 2018-07-12 | 2018-12-07 | 北京奇艺世纪科技有限公司 | The playback method and device of advertisement |
CN109101606A (en) * | 2018-08-02 | 2018-12-28 | 深圳市赛亚创想科技有限公司 | For the data processing method and device of industry information, server |
CN109214841A (en) * | 2017-06-30 | 2019-01-15 | 北京金山安全软件有限公司 | Method and device for obtaining advertisement predicted value and terminal |
CN109272360A (en) * | 2018-09-28 | 2019-01-25 | 有米科技股份有限公司 | A kind of advertisement intelligent recommended method, system and device |
CN109460513A (en) * | 2018-10-31 | 2019-03-12 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating clicking rate prediction model |
CN109474832A (en) * | 2018-11-28 | 2019-03-15 | 深圳市酷开网络科技有限公司 | A kind of information search sort method, intelligent terminal and storage medium |
CN109522506A (en) * | 2018-10-30 | 2019-03-26 | 广东原昇信息科技有限公司 | The dynamic prediction method of visitor's behavioral data conversion ratio |
CN109584020A (en) * | 2018-12-04 | 2019-04-05 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
CN109582862A (en) * | 2018-10-31 | 2019-04-05 | 网易传媒科技(北京)有限公司 | Clicking rate predictor method, medium, system and calculating equipment |
CN109871487A (en) * | 2019-02-22 | 2019-06-11 | 北京搜狐新媒体信息技术有限公司 | A kind of news recalls method and system |
CN110019163A (en) * | 2017-12-05 | 2019-07-16 | 北京京东尚科信息技术有限公司 | Method, system, equipment and the storage medium of prediction, the recommendation of characteristics of objects |
CN110070397A (en) * | 2019-04-24 | 2019-07-30 | 厦门美图之家科技有限公司 | Advertisement orientation method and electronic equipment |
CN110147496A (en) * | 2019-05-13 | 2019-08-20 | 百度在线网络技术(北京)有限公司 | Content delivery method and device |
CN110263959A (en) * | 2018-05-21 | 2019-09-20 | 腾讯科技(深圳)有限公司 | Clicking rate predictor method, device, machinery equipment and computer readable storage medium |
CN110322039A (en) * | 2018-03-29 | 2019-10-11 | 腾讯科技(深圳)有限公司 | A kind of clicking rate predictor method, server and computer readable storage medium |
CN110321517A (en) * | 2019-07-25 | 2019-10-11 | 秒针信息技术有限公司 | Browse detection method, detection device and readable storage medium storing program for executing that resource plays ratio |
CN110909069A (en) * | 2019-11-11 | 2020-03-24 | 口碑(上海)信息技术有限公司 | User hierarchical data display method and device |
CN110969482A (en) * | 2019-11-27 | 2020-04-07 | 杭州古点网络科技有限公司 | Accurate advertisement putting method |
CN111177575A (en) * | 2020-04-07 | 2020-05-19 | 腾讯科技(深圳)有限公司 | Content recommendation method and device, electronic equipment and storage medium |
CN111191165A (en) * | 2019-12-05 | 2020-05-22 | 腾讯科技(深圳)有限公司 | Data processing method, device, equipment and computer readable storage medium |
CN111242318A (en) * | 2020-01-13 | 2020-06-05 | 拉扎斯网络科技(上海)有限公司 | Business model training method and device based on heterogeneous feature library |
CN112241896A (en) * | 2019-07-18 | 2021-01-19 | 百度在线网络技术(北京)有限公司 | Information pushing method, device, equipment and computer readable medium |
CN112434184A (en) * | 2020-12-15 | 2021-03-02 | 四川长虹电器股份有限公司 | Deep interest network sequencing method based on historical movie posters |
CN112633914A (en) * | 2014-07-01 | 2021-04-09 | 谷歌有限责任公司 | System and method for suggesting creative types for online content items to advertisers |
CN112823369A (en) * | 2018-10-09 | 2021-05-18 | 电子湾有限公司 | Generation of personalized banner images using machine learning |
CN112948678A (en) * | 2021-02-26 | 2021-06-11 | 北京房江湖科技有限公司 | Article recalling method and system and article recommending method and system |
CN113129046A (en) * | 2019-12-31 | 2021-07-16 | 上海哔哩哔哩科技有限公司 | Click rate prediction method and device and computer equipment |
CN117132330A (en) * | 2023-10-23 | 2023-11-28 | 蓝色火焰科技成都有限公司 | Intelligent advertisement method, device, equipment and storage medium based on big data |
CN111242318B (en) * | 2020-01-13 | 2024-04-26 | 拉扎斯网络科技(上海)有限公司 | Service model training method and device based on heterogeneous feature library |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1816812A (en) * | 2003-06-02 | 2006-08-09 | Google公司 | Serving advertisements using user request information and user information |
CN101290631A (en) * | 2008-05-28 | 2008-10-22 | 北京百问百答网络技术有限公司 | Network advertisement automatic delivery method and its system |
-
2011
- 2011-10-08 CN CN201110302456XA patent/CN102346899A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1816812A (en) * | 2003-06-02 | 2006-08-09 | Google公司 | Serving advertisements using user request information and user information |
CN101290631A (en) * | 2008-05-28 | 2008-10-22 | 北京百问百答网络技术有限公司 | Network advertisement automatic delivery method and its system |
Cited By (156)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102663617A (en) * | 2012-03-20 | 2012-09-12 | 亿赞普(北京)科技有限公司 | Method and system for prediction of advertisement clicking rate |
CN103377435A (en) * | 2012-04-26 | 2013-10-30 | 飞拓无限信息技术(北京)有限公司 | Method and system for achieving mobile advertisement recommendation |
CN102750334A (en) * | 2012-06-01 | 2012-10-24 | 北京市农林科学院农业科技信息研究所 | Agricultural information accurate propelling method based on data mining (DM) |
CN102750334B (en) * | 2012-06-01 | 2016-12-21 | 北京市农林科学院农业科技信息研究所 | A kind of method that Agricultural Information based on DM precisely pushes |
CN103489117B (en) * | 2012-06-12 | 2015-07-01 | 深圳市腾讯计算机系统有限公司 | Method and system for information releasing |
CN103489117A (en) * | 2012-06-12 | 2014-01-01 | 深圳市腾讯计算机系统有限公司 | Method and system for information releasing |
CN103514178A (en) * | 2012-06-18 | 2014-01-15 | 阿里巴巴集团控股有限公司 | Searching and sorting method and device based on click rate |
CN102779193B (en) * | 2012-07-16 | 2015-05-13 | 哈尔滨工业大学 | Self-adaptive personalized information retrieval system and method |
CN102779193A (en) * | 2012-07-16 | 2012-11-14 | 哈尔滨工业大学 | Self-adaptive personalized information retrieval system and method |
CN102880688B (en) * | 2012-09-14 | 2016-07-27 | 北京百度网讯科技有限公司 | A kind of method for webpage is estimated, device and equipment |
CN102880688A (en) * | 2012-09-14 | 2013-01-16 | 北京百度网讯科技有限公司 | Method, device and equipment for evaluating webpage |
US11710054B2 (en) | 2012-10-08 | 2023-07-25 | Tencent Technology (Shenzhen) Company Limited | Information recommendation method, apparatus, and server based on user data in an online forum |
CN103714084A (en) * | 2012-10-08 | 2014-04-09 | 腾讯科技(深圳)有限公司 | Method and device for recommending information |
US10268960B2 (en) | 2012-10-08 | 2019-04-23 | Tencent Technology (Shenzhen) Company Limited | Information recommendation method, apparatus, and server based on user data in an online forum |
CN103714084B (en) * | 2012-10-08 | 2018-04-03 | 腾讯科技(深圳)有限公司 | The method and apparatus of recommendation information |
CN103853711A (en) * | 2012-11-28 | 2014-06-11 | 中国移动通信集团广西有限公司 | Text information processing method and device |
CN103853711B (en) * | 2012-11-28 | 2017-02-08 | 中国移动通信集团广西有限公司 | Text information processing method and device |
CN105408928A (en) * | 2013-03-15 | 2016-03-16 | 万歌有限公司 | Methods and systems for determining relevance of advertising categories for devices |
CN103235893B (en) * | 2013-05-06 | 2016-03-23 | 重庆大学 | A kind of user-commodity clicking rate adaptive prediction device and Forecasting Methodology |
CN103235893A (en) * | 2013-05-06 | 2013-08-07 | 重庆大学 | User-commodity click rate self-adaptive prediction device and method |
CN103310003A (en) * | 2013-06-28 | 2013-09-18 | 华东师范大学 | Method and system for predicting click rate of new advertisement based on click log |
CN104281547A (en) * | 2013-07-02 | 2015-01-14 | 罗伯特·博世有限公司 | Method for operating a control unit and a control unit having a model calculation unit |
CN104281547B (en) * | 2013-07-02 | 2019-03-22 | 罗伯特·博世有限公司 | For running the method for controller and the controller with model computing unit |
CN103345512A (en) * | 2013-07-06 | 2013-10-09 | 北京品友互动信息技术有限公司 | Online advertising click-through rate forecasting method and device based on user attribute |
CN104424291A (en) * | 2013-09-02 | 2015-03-18 | 阿里巴巴集团控股有限公司 | Method and device for sorting search results |
CN104424291B (en) * | 2013-09-02 | 2018-12-21 | 阿里巴巴集团控股有限公司 | The method and device that a kind of pair of search result is ranked up |
CN103810239A (en) * | 2013-11-19 | 2014-05-21 | 北京国双科技有限公司 | Statistical method and device for conversion data |
CN103810239B (en) * | 2013-11-19 | 2017-06-30 | 北京国双科技有限公司 | The statistical method and device of conversion data |
CN104679743A (en) * | 2013-11-26 | 2015-06-03 | 阿里巴巴集团控股有限公司 | Method and device for determining preference model of user |
CN104679743B (en) * | 2013-11-26 | 2019-10-29 | 阿里巴巴集团控股有限公司 | A kind of method and device of the preference pattern of determining user |
WO2015081801A1 (en) * | 2013-12-06 | 2015-06-11 | Tencent Technology (Shenzhen) Company Limited | Method, server, and system for information push |
CN104750713A (en) * | 2013-12-27 | 2015-07-01 | 阿里巴巴集团控股有限公司 | Method and device for sorting search results |
US10348550B2 (en) | 2014-02-11 | 2019-07-09 | Tencent Technology (Shenzhen) Company Limited | Method and system for processing network media information |
CN104834641B (en) * | 2014-02-11 | 2019-03-15 | 腾讯科技(北京)有限公司 | The processing method and related system of network media information |
WO2015120798A1 (en) * | 2014-02-11 | 2015-08-20 | Tencent Technology (Shenzhen) Company Limited | Method for processing network media information and related system |
CN104834641A (en) * | 2014-02-11 | 2015-08-12 | 腾讯科技(北京)有限公司 | Processing method of network medium information and relevant system |
CN104866474B (en) * | 2014-02-20 | 2018-10-09 | 阿里巴巴集团控股有限公司 | Individuation data searching method and device |
CN104866474A (en) * | 2014-02-20 | 2015-08-26 | 阿里巴巴集团控股有限公司 | Personalized data searching method and device |
CN104883585A (en) * | 2014-02-28 | 2015-09-02 | 腾讯科技(北京)有限公司 | Method for displaying media data, equipment and system thereof |
CN104883585B (en) * | 2014-02-28 | 2019-06-14 | 腾讯科技(北京)有限公司 | Show method, equipment and the system of media data |
WO2015149307A1 (en) * | 2014-04-02 | 2015-10-08 | Google Inc. | Systems and methods for optimizing content layout using behavior metrics |
US9465887B2 (en) | 2014-04-02 | 2016-10-11 | Google Inc. | Systems and methods for optimizing content layout using behavior metrics |
US10146743B2 (en) | 2014-04-02 | 2018-12-04 | Google Llc | Systems and methods for optimizing content layout using behavior metrics |
CN103955842B (en) * | 2014-04-14 | 2017-03-29 | 武汉烽火普天信息技术有限公司 | A kind of online advertisement commending system and method towards mass media data |
CN103955842A (en) * | 2014-04-14 | 2014-07-30 | 武汉烽火普天信息技术有限公司 | Online advertisement recommending system and method for large-scale medium data |
CN104090919A (en) * | 2014-06-16 | 2014-10-08 | 华为技术有限公司 | Advertisement recommending method and advertisement recommending server |
WO2015192667A1 (en) * | 2014-06-16 | 2015-12-23 | 华为技术有限公司 | Advertisement recommending method and advertisement recommending server |
CN104090919B (en) * | 2014-06-16 | 2017-04-19 | 华为技术有限公司 | Advertisement recommending method and advertisement recommending server |
CN106688215B (en) * | 2014-06-27 | 2020-07-14 | 谷歌有限责任公司 | Automatic click type selection for content performance optimization |
CN106688215B8 (en) * | 2014-06-27 | 2020-08-18 | 谷歌有限责任公司 | Automatic click type selection for content performance optimization |
CN106688215A (en) * | 2014-06-27 | 2017-05-17 | 谷歌公司 | Automated click type selection for content performance optimization |
CN112633914A (en) * | 2014-07-01 | 2021-04-09 | 谷歌有限责任公司 | System and method for suggesting creative types for online content items to advertisers |
CN105447045A (en) * | 2014-09-02 | 2016-03-30 | 阿里巴巴集团控股有限公司 | Information ordering method and device and information providing method and system |
CN105447045B (en) * | 2014-09-02 | 2019-06-07 | 阿里巴巴集团控股有限公司 | Information sorting method, apparatus and information providing method, 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 |
CN105630185A (en) * | 2014-11-25 | 2016-06-01 | 奥多比公司 | Temporal dynamics in display advertising prediction |
CN105630185B (en) * | 2014-11-25 | 2021-02-12 | 奥多比公司 | Temporal dynamics in display advertisement prediction |
CN104503980B (en) * | 2014-11-26 | 2020-06-05 | 百度在线网络技术(北京)有限公司 | Determining comprehensive search information and determining candidate search sequences to be pushed according to comprehensive search information |
CN104503980A (en) * | 2014-11-26 | 2015-04-08 | 百度在线网络技术(北京)有限公司 | Method for determining comprehensive search information and for determining alternative search sequences to be pushed by comprehensive search information |
CN104484372A (en) * | 2014-12-04 | 2015-04-01 | 北京奇虎科技有限公司 | Detecting method and device of business object sending information |
CN105812844A (en) * | 2014-12-29 | 2016-07-27 | 深圳市Tcl高新技术开发有限公司 | User advertisement push method and system of TV |
CN105812844B (en) * | 2014-12-29 | 2019-02-26 | 深圳市Tcl高新技术开发有限公司 | A kind of the user advertising method for pushing and system of TV |
CN104537502A (en) * | 2015-01-15 | 2015-04-22 | 北京嘀嘀无限科技发展有限公司 | Method and device for processing orders |
CN105989004A (en) * | 2015-01-27 | 2016-10-05 | 阿里巴巴集团控股有限公司 | Information releasing pretreatment method and device |
CN104574160A (en) * | 2015-01-29 | 2015-04-29 | 深圳市英威诺科技有限公司 | Smooth advertisement traffic control method |
CN106156878B (en) * | 2015-04-21 | 2020-09-15 | 深圳市腾讯计算机系统有限公司 | Advertisement click rate correction method and device |
CN106156878A (en) * | 2015-04-21 | 2016-11-23 | 深圳市腾讯计算机系统有限公司 | Ad click rate antidote and device |
CN106203668B (en) * | 2015-05-07 | 2021-04-20 | 北京极迅智程网络科技股份有限公司 | System for carrying out predictive analysis and display on information index |
CN106203668A (en) * | 2015-05-07 | 2016-12-07 | 北京极迅智程网络科技股份有限公司 | A kind of system being predicted information index analyzing and showing |
CN104992348A (en) * | 2015-06-24 | 2015-10-21 | 深圳市腾讯计算机系统有限公司 | Method and device for displaying information |
CN104992348B (en) * | 2015-06-24 | 2018-07-10 | 深圳市腾讯计算机系统有限公司 | A kind of method and apparatus of information displaying |
CN104933603A (en) * | 2015-06-30 | 2015-09-23 | 北京奇艺世纪科技有限公司 | A method and an apparatus for advertisement displaying |
CN105183772A (en) * | 2015-08-07 | 2015-12-23 | 百度在线网络技术(北京)有限公司 | Release information click rate estimation method and apparatus |
WO2017028728A1 (en) * | 2015-08-18 | 2017-02-23 | 北京金山安全软件有限公司 | Determining method and device for click through rate (ctr) |
CN106504011B (en) * | 2015-09-06 | 2019-11-05 | 阿里巴巴集团控股有限公司 | A kind of methods of exhibiting and device of business object |
CN106504011A (en) * | 2015-09-06 | 2017-03-15 | 阿里巴巴集团控股有限公司 | A kind of methods of exhibiting of business object and device |
CN105184321B (en) * | 2015-09-10 | 2018-12-14 | 北京金山安全软件有限公司 | Data processing method and device for ftrl model |
CN105184321A (en) * | 2015-09-10 | 2015-12-23 | 北京金山安全软件有限公司 | Data processing method and device for ftrl model |
CN108369433A (en) * | 2015-09-18 | 2018-08-03 | Mms美国控股有限公司 | Micro- moment analysis |
CN105117491A (en) * | 2015-09-22 | 2015-12-02 | 北京百度网讯科技有限公司 | Page pushing method and device |
CN105117491B (en) * | 2015-09-22 | 2018-12-25 | 北京百度网讯科技有限公司 | Page push method and apparatus |
CN106557475B (en) * | 2015-09-24 | 2021-01-26 | 北京国双科技有限公司 | Method and device for selecting new path |
CN106557475A (en) * | 2015-09-24 | 2017-04-05 | 北京国双科技有限公司 | Path system of selection and device |
CN105260477A (en) * | 2015-11-06 | 2016-01-20 | 北京金山安全软件有限公司 | Information pushing method and device |
CN106897892A (en) * | 2015-12-18 | 2017-06-27 | 北京奇虎科技有限公司 | Advertisement placement method and device |
CN106933825A (en) * | 2015-12-29 | 2017-07-07 | 北京国双科技有限公司 | Website data processing method and processing device |
CN105631711A (en) * | 2015-12-30 | 2016-06-01 | 合一网络技术(北京)有限公司 | Advertisement putting method and apparatus |
CN105701191B (en) * | 2016-01-08 | 2020-12-29 | 腾讯科技(深圳)有限公司 | Pushed information click rate estimation method and device |
CN105701191A (en) * | 2016-01-08 | 2016-06-22 | 腾讯科技(深圳)有限公司 | Push information click rate estimation method and device |
CN105678587A (en) * | 2016-01-12 | 2016-06-15 | 腾讯科技(深圳)有限公司 | Recommendation feature determination method and information recommendation method and device |
US11144950B2 (en) | 2016-01-12 | 2021-10-12 | Tencent Technology (Shenzhen) Company Limited | Information delivery method and apparatus, server, and storage medium |
US11301525B2 (en) | 2016-01-12 | 2022-04-12 | Tencent Technology (Shenzhen) Company Limited | Method and apparatus for processing information |
CN105678587B (en) * | 2016-01-12 | 2020-11-24 | 腾讯科技(深圳)有限公司 | Recommendation feature determination method, information recommendation method and device |
WO2017121258A1 (en) * | 2016-01-12 | 2017-07-20 | 腾讯科技(深圳)有限公司 | Information delivery method and apparatus, server, and storage medium |
CN105956888A (en) * | 2016-05-31 | 2016-09-21 | 北京创意魔方广告有限公司 | Advertisement personalized display method |
CN107545301B (en) * | 2016-06-23 | 2020-10-20 | 阿里巴巴集团控股有限公司 | Page display method and device |
CN107545301A (en) * | 2016-06-23 | 2018-01-05 | 阿里巴巴集团控股有限公司 | Page display method and device |
CN106204122B (en) * | 2016-07-05 | 2020-09-29 | 北京京东尚科信息技术有限公司 | Contact point value measurement method and device |
CN106204122A (en) * | 2016-07-05 | 2016-12-07 | 北京京东尚科信息技术有限公司 | Contact measure of value method and apparatus |
CN106355442A (en) * | 2016-08-30 | 2017-01-25 | 杭州启冠网络技术有限公司 | Online precise advertising method and system based on big data driving |
CN108229987A (en) * | 2016-12-14 | 2018-06-29 | 腾讯科技(深圳)有限公司 | A kind of advertisement inventory evaluation method and device |
CN108229994A (en) * | 2016-12-21 | 2018-06-29 | 北京金山云网络技术有限公司 | A kind of information-pushing method and device |
CN106780273A (en) * | 2016-12-30 | 2017-05-31 | 中国民航信息网络股份有限公司 | Passenger flight requirement analysis method and system |
CN106886915B (en) * | 2017-01-17 | 2020-07-28 | 华南理工大学 | Advertisement click estimation method based on time attenuation sampling |
CN106886915A (en) * | 2017-01-17 | 2017-06-23 | 华南理工大学 | A kind of ad click predictor method based on time decay sampling |
CN108876422A (en) * | 2017-05-09 | 2018-11-23 | 北京京东尚科信息技术有限公司 | For the method, apparatus of information popularization, electronic equipment and computer-readable medium |
CN109214841A (en) * | 2017-06-30 | 2019-01-15 | 北京金山安全软件有限公司 | Method and device for obtaining advertisement predicted value and terminal |
CN107613022A (en) * | 2017-10-20 | 2018-01-19 | 广州优视网络科技有限公司 | Content delivery method, device and computer equipment |
CN107613022B (en) * | 2017-10-20 | 2020-10-16 | 阿里巴巴(中国)有限公司 | Content pushing method and device and computer equipment |
CN110019163A (en) * | 2017-12-05 | 2019-07-16 | 北京京东尚科信息技术有限公司 | Method, system, equipment and the storage medium of prediction, the recommendation of characteristics of objects |
CN108062684B (en) * | 2017-12-12 | 2021-01-22 | 北京奇艺世纪科技有限公司 | Method and device for predicting click rate of advertisement |
CN108062684A (en) * | 2017-12-12 | 2018-05-22 | 北京奇艺世纪科技有限公司 | The clicking rate Forecasting Methodology and device of a kind of advertisement |
CN108388624A (en) * | 2018-02-12 | 2018-08-10 | 科大讯飞股份有限公司 | Multimedia messages recommend method and device |
CN108647691A (en) * | 2018-03-12 | 2018-10-12 | 杭州电子科技大学 | A kind of image classification method based on click feature prediction |
CN108446374B (en) * | 2018-03-16 | 2019-09-20 | 北京三快在线科技有限公司 | User's Intention Anticipation method, apparatus, electronic equipment, storage medium |
CN108446374A (en) * | 2018-03-16 | 2018-08-24 | 北京三快在线科技有限公司 | User view prediction technique, device, electronic equipment, storage medium |
CN110322039A (en) * | 2018-03-29 | 2019-10-11 | 腾讯科技(深圳)有限公司 | A kind of clicking rate predictor method, server and computer readable storage medium |
CN110322039B (en) * | 2018-03-29 | 2022-12-02 | 腾讯科技(深圳)有限公司 | Click rate estimation method, server and computer readable storage medium |
CN110263959A (en) * | 2018-05-21 | 2019-09-20 | 腾讯科技(深圳)有限公司 | Clicking rate predictor method, device, machinery equipment and computer readable storage medium |
CN110263959B (en) * | 2018-05-21 | 2024-01-09 | 腾讯科技(深圳)有限公司 | Click rate estimation method, click rate estimation device, machine equipment and computer readable storage medium |
CN108965951A (en) * | 2018-07-12 | 2018-12-07 | 北京奇艺世纪科技有限公司 | The playback method and device of advertisement |
CN108965951B (en) * | 2018-07-12 | 2020-10-02 | 北京奇艺世纪科技有限公司 | Advertisement playing method and device |
CN109101606A (en) * | 2018-08-02 | 2018-12-28 | 深圳市赛亚创想科技有限公司 | For the data processing method and device of industry information, server |
CN109272360A (en) * | 2018-09-28 | 2019-01-25 | 有米科技股份有限公司 | A kind of advertisement intelligent recommended method, system and device |
CN109272360B (en) * | 2018-09-28 | 2021-09-10 | 有米科技股份有限公司 | Intelligent advertisement recommendation method, system and device |
CN112823369A (en) * | 2018-10-09 | 2021-05-18 | 电子湾有限公司 | Generation of personalized banner images using machine learning |
CN109522506A (en) * | 2018-10-30 | 2019-03-26 | 广东原昇信息科技有限公司 | The dynamic prediction method of visitor's behavioral data conversion ratio |
CN109460513B (en) * | 2018-10-31 | 2021-01-08 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating click rate prediction model |
CN109460513A (en) * | 2018-10-31 | 2019-03-12 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating clicking rate prediction model |
CN109582862A (en) * | 2018-10-31 | 2019-04-05 | 网易传媒科技(北京)有限公司 | Clicking rate predictor method, medium, system and calculating equipment |
CN109474832A (en) * | 2018-11-28 | 2019-03-15 | 深圳市酷开网络科技有限公司 | A kind of information search sort method, intelligent terminal and storage medium |
CN109474832B (en) * | 2018-11-28 | 2021-02-02 | 深圳市酷开网络科技有限公司 | Information searching and sorting method, intelligent terminal and storage medium |
CN109584020A (en) * | 2018-12-04 | 2019-04-05 | 联想(北京)有限公司 | A kind of information processing method and electronic equipment |
CN109871487A (en) * | 2019-02-22 | 2019-06-11 | 北京搜狐新媒体信息技术有限公司 | A kind of news recalls method and system |
CN109871487B (en) * | 2019-02-22 | 2021-03-23 | 北京搜狐新媒体信息技术有限公司 | News recall method and system |
CN110070397A (en) * | 2019-04-24 | 2019-07-30 | 厦门美图之家科技有限公司 | Advertisement orientation method and electronic equipment |
CN110070397B (en) * | 2019-04-24 | 2021-08-20 | 厦门美图之家科技有限公司 | Advertisement targeting method and electronic equipment |
CN110147496B (en) * | 2019-05-13 | 2021-06-25 | 百度在线网络技术(北京)有限公司 | Content pushing method and device |
CN110147496A (en) * | 2019-05-13 | 2019-08-20 | 百度在线网络技术(北京)有限公司 | Content delivery method and device |
CN112241896A (en) * | 2019-07-18 | 2021-01-19 | 百度在线网络技术(北京)有限公司 | Information pushing method, device, equipment and computer readable medium |
CN110321517A (en) * | 2019-07-25 | 2019-10-11 | 秒针信息技术有限公司 | Browse detection method, detection device and readable storage medium storing program for executing that resource plays ratio |
CN110909069A (en) * | 2019-11-11 | 2020-03-24 | 口碑(上海)信息技术有限公司 | User hierarchical data display method and device |
CN110969482B (en) * | 2019-11-27 | 2023-08-01 | 杭州古点网络科技有限公司 | Accurate advertisement delivery method |
CN110969482A (en) * | 2019-11-27 | 2020-04-07 | 杭州古点网络科技有限公司 | Accurate advertisement putting method |
CN111191165A (en) * | 2019-12-05 | 2020-05-22 | 腾讯科技(深圳)有限公司 | Data processing method, device, equipment and computer readable storage medium |
CN113129046A (en) * | 2019-12-31 | 2021-07-16 | 上海哔哩哔哩科技有限公司 | Click rate prediction method and device and computer equipment |
CN111242318A (en) * | 2020-01-13 | 2020-06-05 | 拉扎斯网络科技(上海)有限公司 | Business model training method and device based on heterogeneous feature library |
CN111242318B (en) * | 2020-01-13 | 2024-04-26 | 拉扎斯网络科技(上海)有限公司 | Service model training method and device based on heterogeneous feature library |
CN111177575A (en) * | 2020-04-07 | 2020-05-19 | 腾讯科技(深圳)有限公司 | Content recommendation method and device, electronic equipment and storage medium |
CN111177575B (en) * | 2020-04-07 | 2020-07-24 | 腾讯科技(深圳)有限公司 | Content recommendation method and device, electronic equipment and storage medium |
US11893071B2 (en) | 2020-04-07 | 2024-02-06 | Tencent Technology (Shenzhen) Company Limited | Content recommendation method and apparatus, electronic device, and storage medium |
CN112434184A (en) * | 2020-12-15 | 2021-03-02 | 四川长虹电器股份有限公司 | Deep interest network sequencing method based on historical movie posters |
CN112948678A (en) * | 2021-02-26 | 2021-06-11 | 北京房江湖科技有限公司 | Article recalling method and system and article recommending method and system |
CN112948678B (en) * | 2021-02-26 | 2023-07-21 | 北京房江湖科技有限公司 | Article recall method and system and article recommendation method and system |
CN117132330A (en) * | 2023-10-23 | 2023-11-28 | 蓝色火焰科技成都有限公司 | Intelligent advertisement method, device, equipment and storage medium based on big data |
CN117132330B (en) * | 2023-10-23 | 2024-01-30 | 蓝色火焰科技成都有限公司 | Intelligent advertisement method, device, equipment and storage medium based on big data |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102346899A (en) | Method and device for predicting advertisement click rate based on user behaviors | |
CN109408703B (en) | Information recommendation method and system, device, electronic equipment and storage medium thereof | |
Zamani et al. | Situational context for ranking in personal search | |
US20170091805A1 (en) | Advertisement Recommendation Method and Advertisement Recommendation Server | |
US10061820B2 (en) | Generating a user-specific ranking model on a user electronic device | |
US8768954B2 (en) | Relevancy-based domain classification | |
US8442849B2 (en) | Emotional mapping | |
US20090187540A1 (en) | Prediction of informational interests | |
US20100017262A1 (en) | Predicting selection rates of a document using click-based translation dictionaries | |
US20110270678A1 (en) | System and method for using real-time keywords for targeting advertising in web search and social media | |
CN103838756A (en) | Method and device for determining pushed information | |
US8888497B2 (en) | Emotional web | |
CN103886090A (en) | Content recommendation method and device based on user favorites | |
CN108416649A (en) | Search result ordering method, device, electronic equipment and storage medium | |
US20110231256A1 (en) | Automated building of a model for behavioral targeting | |
CN105677780A (en) | Scalable user intent mining method and system thereof | |
CN103164521A (en) | Keyword calculation method and device based on user browse and search actions | |
WO2011112319A2 (en) | Emotional targeting | |
CN103348342A (en) | Personal content stream based on user-topic profile | |
CN109819015B (en) | Information pushing method, device and equipment based on user portrait and storage medium | |
EP1735720A2 (en) | System and method for responding to search requests in a computer network | |
CN104462594A (en) | Method and device for providing user personalized resource message pushing | |
WO2009029689A1 (en) | Distinguishing accessories from products for ranking search results | |
CN103412881A (en) | Method and system for providing search result | |
US20110015990A1 (en) | Advertising Based on a Dynamic Ad Taxonomy |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C12 | Rejection of a patent application after its publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120208 |