CN103345512A - Online advertising click-through rate forecasting method and device based on user attribute - Google Patents
Online advertising click-through rate forecasting method and device based on user attribute Download PDFInfo
- Publication number
- CN103345512A CN103345512A CN2013102831999A CN201310283199A CN103345512A CN 103345512 A CN103345512 A CN 103345512A CN 2013102831999 A CN2013102831999 A CN 2013102831999A CN 201310283199 A CN201310283199 A CN 201310283199A CN 103345512 A CN103345512 A CN 103345512A
- Authority
- CN
- China
- Prior art keywords
- user
- interest
- clicking rate
- advertisement
- training sample
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an online advertising click-through rate forecasting method based on user attribute, and further provides an online advertising click-through rate forecasting device based on the user attribute correspondingly. The method comprises the followings steps that a network user interest attribute category system is built, wherein the category system comprises a plurality of interest categories; user behavior data are collected, behavior classification models are built, the user behavior data are divided according to the classification models, and corresponding behavior data are used as a training sample of the interest categories according to classification; a click-through rate forecasting model corresponding to the interest categories is built to conduct click-through rate forecasting according to the interest categories and the training sample. The online advertising click-through rate forecasting method based on the user attribute and the online advertising click-through rate forecasting device based on the user attribute can improve the advertising click-through rate forecasting precision and improve network browsing experience of a user.
Description
Technical field
The present invention relates to the advertisement putting field on the internet, relate in particular to a kind of web advertisement clicking rate Forecasting Methodology and device based on user property.
Background technology
The internet is compared with traditional mass medium as typical Focus medium, and its unique demand and feature is arranged.For example, the user is interested in the information that satisfies its demand of showing on the internet, and the advertising message of passive displaying is often entertained the psychology of conflicting.In order to show the required advertising message of user as far as possible, in the prior art usually based on the audient of the statistics and analysis advertisement putting of user's clicking rate, with better with advertisement putting to the user platform that Related product or demand for services are arranged, with the lifting advertisement delivery effect.
Therefore, how the probability of more accurately user being clicked the web advertisement is predicted and is become the major issue that the advertisement putting field need solve, although the factor of assessment advertisement delivery effect is not limited to clicking rate, also comprise such as conversion ratio, rate of return on investment (ROI) etc., but clicking rate is the most common, the most important factor of assessment advertisement delivery effect, if can promote the assessment accuracy of clicking rate, other assessment factors also can follow the method so, assess in the lump.
Summary of the invention
The purpose of this invention is to provide a kind of web advertisement clicking rate Forecasting Methodology and device based on user property, can effectively promote web advertisement clicking rate accuracy for predicting.
According to an aspect of the present invention, provide a kind of web advertisement clicking rate Forecasting Methodology based on user property, this method may further comprise the steps:
Set up network user's interest attribute classification system, wherein, described classification system comprises a plurality of category of interest;
Collect user behavior data, set up the behavior disaggregated model, based on described disaggregated model described user behavior data is divided, and according to classifying the training sample of respective behavior data as described category of interest;
Based on category of interest and the described training sample of described user's correspondence, set up the clicking rate forecast model corresponding with described category of interest and carry out the clicking rate prediction.
According to another aspect of the present invention, also provide a kind of web advertisement clicking rate prediction unit based on user property, having comprised:
The classification system is set up the unit, is used for setting up network user's interest attribute classification system, and wherein, described classification system comprises a plurality of category of interest;
The user behavior analysis unit is used for collecting user behavior data, sets up the behavior disaggregated model, based on described disaggregated model described user behavior data is divided, and according to classifying the training sample of respective behavior data as described category of interest;
The clicking rate predicting unit based on category of interest and the described training sample of described user's correspondence, is set up the clicking rate forecast model corresponding with described category of interest and is carried out the clicking rate prediction.
Compared with prior art, the present invention has the following advantages:
1) ad click rate prediction scheme provided by the invention can be according to the variation of active user's point of interest, and the advertisement that real-time exhibition and user interest mate most improves user's viewing experience;
2) the present invention is by extracting user's network crowd attribute, ad click rate is predicted, has high generality, especially in the advertisement putting field, based on institute's prediction result, can be by catching the variation of user interest point, throw in real time and show the advertisement of suitable this user interest, thereby promote advertising results.
Description of drawings
By reading the detailed description of doing with reference to the following drawings that non-limiting example is done, it is more obvious that other features, objects and advantages of the present invention will become:
Fig. 1 is the web advertisement clicking rate Forecasting Methodology process flow diagram based on user property in accordance with a preferred embodiment of the present invention;
Fig. 2 is DAAT tree structure synoptic diagram in accordance with a preferred embodiment of the present invention;
Fig. 3 is the schematic block diagram based on the web advertisement clicking rate prediction unit of user property in accordance with a preferred embodiment of the present invention.
Embodiment
Below in conjunction with accompanying drawing the present invention is described in further detail.
According to an aspect of the present invention, provide a kind of web advertisement clicking rate method based on user property.Need to prove, though present embodiment just is example explanation web advertisement clicking rate Forecasting Methodology and device with the clicking rate, in fact, the present invention can also be applicable to advertising results assessment factors such as web advertisement conversion ratio, rate of return on investment in interior prediction, and clicking rate is predicted as preferred embodiment wherein.
Please refer to Fig. 1, Fig. 1 is the web advertisement clicking rate Forecasting Methodology process flow diagram based on user property in accordance with a preferred embodiment of the present invention.
As shown in Figure 1, the web advertisement clicking rate Forecasting Methodology based on user property provided by the present invention may further comprise the steps:
Step S101 sets up network user's interest attribute classification system, and wherein, described classification system comprises a plurality of category of interest.
Particularly, when the resource focus in the Internet advertising trade mode is converted to the people from advertisement position, understand exactly " people ", become the important prerequisite condition that the web advertisement is thrown in, the division of crowd's attribute of science, be the basis of correct effectively identification commercial audience, therefore, set up network user's interest attribute classification system (being called for short DAAT) and have vital role for division and the classification of user interest attribute.
More specifically, described classification system is tree-shaped data structure, and in this system, the user interest attribute is divided into a plurality of levels, and each level comprises a plurality of category of interest attributes.Please refer to Fig. 2, Fig. 2 is DAAT tree structure synoptic diagram in accordance with a preferred embodiment of the present invention.As shown in Figure 2, typically, crowd's interest attribute is divided into Regional Distribution, the ascribed characteristics of population, individual's concern and purchase intention four dimensions, and this four dimensions is in the one-level dimension of described tree structure; Continue down to divide, for example, the ascribed characteristics of population is divided into sex, age, occupation, monthly income, educational background, six dimensions of crucial division of life span, individual's concern is divided into information, news, automobile, house properties etc. wherein, are further divided, sex is divided into man, woman ... and the like, each level comprises the data of a plurality of dimensions, and each dimension down can continue to divide, until the leaf node end of this tree structure.
Above-mentioned DAAT system can disclose the inherent correlativity of user interest attribute multi-faceted with tree structure, at many levels, it is attached in the present embodiment, can carry out predicting to clicking rate more accurately, for example, for a people who likes outdoor travel, can reasonably infer this user's interested hotel lodging, traffic and outdoor products etc. in conjunction with DAAT, namely described DAAT can help advertiser localizing objects user better.
Preferably, can also upgrade network user's interest attribute classification system, for example, increase interest label and respective labels content.
Step S102 collects user behavior data, sets up the behavior disaggregated model, based on described disaggregated model described user behavior data is divided, and according to classifying the training sample of respective behavior data as described category of interest.
Particularly, collected user behavior data comprises all kinds of behavioral datas such as user's visit, click, purchase, search, these behavioral datas can also be classified according to time, region, medium and client, the mode of collecting data will not limit at this, obtains user's diverse network behavioral data as grasping by webpage.
Wherein, described behavior disaggregated model is set up based on processes such as the extraction of the collection of behavioral data, sample data, cleaning, filtrations and in conjunction with described DAAT, and the behavior disaggregated model of setting up comprises corresponding parameter.More specifically, for ad content to be put, mate the user behavior data corresponding with it, carry out modeling based on the behavioral data of described correspondence.
According to the behavior disaggregated model of setting up, can with behavioral data according to respective standard divide, processing such as weighted sum normalization, for example, whether produce corresponding advertising results based on described behavioral data, it is divided into positive sample and negative sample, and can be based on the characteristic distributions of sample, align that negative sample is weighted or normalized etc.
After based on the behavior disaggregated model collected user behavior data being classified, set up the category of interest corresponding relation of described grouped data and described classification system, namely according to the distribute training sample of corresponding behavioral data of described category of interest.
Step S103 based on category of interest and the described training sample of described user's correspondence, sets up the clicking rate forecast model corresponding with described category of interest and carries out the clicking rate prediction.
Particularly, obtain user behavior data training sample and with described user's category of interest distribution situation after, according to the data characteristics of each category of interest residing classification system level and training sample, set up the clicking rate forecast model corresponding with each category of interest.Further, for advertisement to be put, based on described clicking rate forecast model, predict the probability that advertisement to be put is clicked by the user.
Wherein, the basic algorithm that described forecast model is set up does not limit at this, comprises such as logistic regression, random forest scheduling algorithm.
Preferably, in conjunction with the distribution situation of described training sample and described user interest classification, described basic algorithm is carried out accommodation, to different classes of training sample, according to its importance degree corresponding weight is set, comprises the weight of behavioral datas such as visit, click, registration or purchase.For example, the user is clicked the corresponding training sample of behavior higher training weight is set, be click data with greater than the leaf node of setting threshold its output valve being set in this training sample perhaps, in other words, in the corresponding training sample of click behavior institute, if exist most of with having certain common leaf node attribute per family, and these users to account for all users' ratio very high (for example, be higher than certain threshold value), determine that then the user with this attribute can click corresponding advertisement.Wherein, importance degree can be divided into the user and throws in importance influence degree, unit price influence degree, prediction clicking rate influence degree etc.
Preferably, further combined with advertisement attributes, the advertisement position attribute of advertisement to be put, based on described clicking rate forecast model, predict the probability that advertisement to be put is clicked by the active user.Wherein, described advertisement attributes comprises ad content, ad content layout, advertising rates etc., and described advertisement position attribute comprises area, length ratio, the page location of waiting to show advertisement, represents form etc.
Preferably, further combined with described user's present located medium and page info attribute, comprise medium type, the corresponding user interest attribute of content of pages, based on described clicking rate forecast model, predict the probability that advertisement to be put is clicked by the active user.
Compared with prior art, clicking rate Forecasting Methodology provided by the present invention has the following advantages:
1) this method can be based on advertising creative and medium property, interest attribute in conjunction with the user is predicted ad click rate, wherein integrated user's long-term interest and short-term hobby, for important impetus has been played in ad click rate prediction and advertisement putting optimization;
2) Forecasting Methodology provided by the invention can be seen clearly each ad traffic value behind preferably, further optimizes advertisement delivery effect;
3) the present invention is based on user interest behavioral data is divided into a plurality of levels, and set up the clicking rate forecast model of each category of interest correspondence according to the contact between each layer data, can improve accuracy for predicting.
According to another aspect of the present invention, also provide a kind of web advertisement clicking rate prediction unit based on user property.Please refer to shown in Figure 3ly, Fig. 3 is the schematic block diagram based on the web advertisement clicking rate prediction unit of user property according to another preferred embodiment of the present invention.According to Fig. 3, described device comprises:
The classification system is set up unit 301, is used for setting up network user's interest attribute classification system, and wherein, described classification system comprises a plurality of category of interest;
User behavior analysis unit 302 is used for collecting user behavior data, sets up the behavior disaggregated model, based on described disaggregated model described user behavior data is divided, and according to classifying the training sample of respective behavior data as described category of interest;
Clicking rate predicting unit 303 based on category of interest and the described training sample of described user's correspondence, is set up the clicking rate forecast model corresponding with described category of interest and is carried out the clicking rate prediction.
Below, will the course of work of each unit of clicking rate prediction unit provided by the present invention be specifically described.
Particularly, described classification system is set up the classification system of setting up unit 301 and is tree-shaped data structure, and in this system, the user interest attribute is divided into a plurality of levels, and each level comprises a plurality of category of interest attributes.Typically, for example, crowd's interest attribute is divided into Regional Distribution, the ascribed characteristics of population, individual's concern and purchase intention four dimensions, and this four dimensions is in the one-level dimension of described tree structure; Continue down to divide, for example, the ascribed characteristics of population is divided into sex, age, occupation, monthly income, educational background, six dimensions of crucial division of life span, individual's concern is divided into information, news, automobile, house properties etc. wherein, are further divided, sex is divided into man, woman ... and the like, each level comprises the data of a plurality of dimensions, and each dimension down can continue to divide, until the leaf node end of this tree structure.
Above-mentioned DAAT system can disclose the inherent correlativity of user interest attribute multi-faceted with tree structure, at many levels, it is attached in the present embodiment, can carry out predicting to clicking rate more accurately, for example, for a people who likes outdoor travel, can reasonably infer this user's interested hotel lodging, traffic and outdoor products etc. in conjunction with DAAT, namely described DAAT can help advertiser localizing objects user better.
Described user behavior analysis unit 302 carries out the foundation of behavior disaggregated model and the division of behavioral data by the behavioral data of collecting the user.Particularly, collected user behavior data comprises all kinds of behavioral datas such as user's visit, click, purchase, search, these behavioral datas can also be classified according to time, region, medium and client, the mode of collecting data for described user behavior analysis unit 302 will not limit at this, obtains user's diverse network behavioral data as grasping by webpage.
Wherein, described user behavior analysis unit 302 is set up the behavior disaggregated model and is formed correlation parameter in the described disaggregated model based on processes such as the extraction of the collection of behavioral data, sample data, cleaning, filtrations and in conjunction with described DAAT.More specifically, for ad content to be put, described user behavior analysis unit 302 is by the coupling user behavior data corresponding with this ad content, carries out modeling based on the behavioral data of described correspondence.
Further, described user behavior analysis unit 302 is according to the behavior disaggregated model of setting up, can with behavioral data according to respective standard divide, processing such as weighted sum normalization, for example, whether produce corresponding advertising results based on described behavioral data, it is divided into positive sample and negative sample, and can be based on the characteristic distributions of sample, aligns that negative sample is weighted or normalized etc.
After based on the behavior disaggregated model collected user behavior data being classified, the category of interest corresponding relation of described grouped data and described classification system is set up in described user behavior analysis unit 302, namely according to the distribute training sample of corresponding behavioral data of described category of interest.
By described user behavior analysis unit 302 form described user behavior data training sample and with described user's category of interest distribution situation after, by the data characteristics of described clicking rate predicting unit 303 according to each category of interest residing classification system level and training sample, set up the clicking rate forecast model corresponding with each category of interest.Further, for advertisement to be put, described clicking rate predicting unit 303 is predicted the probability that advertisement to be put is clicked by the user based on described clicking rate forecast model.
Wherein, the basic algorithm that described forecast model is set up does not limit at this, comprises such as logistic regression, random forest scheduling algorithm.
Preferably, distribution situation in conjunction with described training sample and described user interest classification, described basic algorithm is carried out accommodation, to different classes of training sample, according to its importance degree corresponding weight is set, for example, the user is clicked the corresponding training sample of behavior higher training weight is set, be click data with greater than the leaf node of setting threshold its output valve being set in this training sample perhaps, in other words, in the corresponding training sample of click behavior institute, if exist most of with having certain common leaf node attribute per family, and the ratio that these users account for all users very high (for example, being higher than certain threshold value) determines that then the user with this attribute can click corresponding advertisement.
Preferably, described clicking rate predicting unit 303 based on described clicking rate forecast model, is predicted the probability that advertisement to be put is clicked by the active user further combined with advertisement attributes, the advertisement position attribute of advertisement to be put.Wherein, described advertisement attributes comprises ad content, ad content layout, advertising rates etc., and described advertisement position attribute comprises area, length ratio, the page location of waiting to show advertisement, represents form etc.
Preferably, described clicking rate predicting unit 303 is passed through in conjunction with described user's present located medium and page info attribute, comprise medium type, the corresponding user interest attribute of content of pages, based on described clicking rate forecast model, predict the probability that advertisement to be put is clicked by the active user.
Clicking rate prediction unit provided by the present invention has the following advantages: by behavioral data and the described DAAT system in conjunction with the user, set up the ad click rate forecast model corresponding with each category of interest, clicking rate to advertisement to be put is predicted, contain all types of user each series advertisements has been clicked the data predicted model, promoted web advertisement clicking rate accuracy for predicting.
Above disclosed only is preferred embodiment of the present invention, certainly can not limit the present invention's interest field with this, therefore the equivalent variations of doing according to claim of the present invention, in the general idea that does not change in the technical program, and just adopt the similar design scheme according to technical thought of the present invention, just replace some algoritic modules and details parameter, change under the situation of the building mode of general frame and order, be equivalent to just provide for simple replacement of implementation, and do not change inherent algorithm logic, still belong to the scope that the present invention is contained.
Claims (12)
1. web advertisement clicking rate Forecasting Methodology based on user property, this method comprises:
A) set up network user's interest attribute classification system, wherein, described classification system comprises a plurality of category of interest;
B) collect user behavior data, set up the behavior disaggregated model, based on described disaggregated model described user behavior data is divided, and according to classifying the training sample of respective behavior data as described category of interest;
C) based on category of interest and the described training sample of described user's correspondence, set up the clicking rate forecast model corresponding with described category of interest and carry out the clicking rate prediction.
2. method according to claim 1 is characterized in that, described step b) specifically also comprises:
Whether produce corresponding advertising results based on described user behavior data, it is divided into positive sample and negative sample.
3. method according to claim 1 and 2 is characterized in that, described step c) specifically also comprises:
Described training sample is click behavior corresponding sample data, sets the threshold value of training sample, is click data with greater than the leaf node of setting threshold its output valve being set in the described training sample.
4. method according to claim 1 and 2 is characterized in that, described step c) further comprises:
In conjunction with advertisement attributes, the advertisement position attribute of advertisement to be put, based on described clicking rate forecast model, predict the probability that advertisement to be put is clicked by the active user.
5. method according to claim 1 and 2 is characterized in that, described step c) further comprises:
In conjunction with described user's present located medium and page info attribute, based on described clicking rate forecast model, predict the probability that advertisement to be put is clicked by the active user.
6. according to each described method of claim 1-5, it is characterized in that the data structure of described network user's interest attribute classification system is the tree structure that comprises at least one level, at least one dimension.
7. web advertisement clicking rate prediction unit based on user property comprises:
The classification system is set up the unit, is used for setting up network user's interest attribute classification system, and wherein, described classification system comprises a plurality of category of interest;
The user behavior analysis unit is used for collecting user behavior data, sets up the behavior disaggregated model, based on described disaggregated model described user behavior data is divided, and according to classifying the training sample of respective behavior data as described category of interest;
The clicking rate predicting unit based on category of interest and the described training sample of described user's correspondence, is set up the clicking rate forecast model corresponding with described category of interest and is carried out the clicking rate prediction.
8. device according to claim 7 is characterized in that, described user behavior analysis unit also is used for whether producing corresponding advertising results based on described user behavior data, and it is divided into positive sample and negative sample.
9. according to claim 7 or 8 described devices, it is characterized in that described clicking rate predicting unit also is used for following operation:
When described training sample is click behavior corresponding sample data, setting the threshold value of training sample, is click data with greater than the leaf node of setting threshold its output valve being set in the described training sample.
10. according to claim 7 or 8 described devices, it is characterized in that described clicking rate predicting unit further also is used for following operation:
In conjunction with advertisement attributes, the advertisement position attribute of advertisement to be put, based on described clicking rate forecast model, predict the probability that advertisement to be put is clicked by the active user.
11., it is characterized in that described clicking rate predicting unit further also is used for following operation according to claim 7 or 8 described devices:
In conjunction with described user's present located medium and page info attribute, based on described clicking rate forecast model, predict the probability that advertisement to be put is clicked by the active user.
12., it is characterized in that the data structure of described network user's interest attribute classification system is the tree structure that comprises at least one level, at least one dimension according to each described device of claim 7-11.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013102831999A CN103345512A (en) | 2013-07-06 | 2013-07-06 | Online advertising click-through rate forecasting method and device based on user attribute |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013102831999A CN103345512A (en) | 2013-07-06 | 2013-07-06 | Online advertising click-through rate forecasting method and device based on user attribute |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103345512A true CN103345512A (en) | 2013-10-09 |
Family
ID=49280307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013102831999A Pending CN103345512A (en) | 2013-07-06 | 2013-07-06 | Online advertising click-through rate forecasting method and device based on user attribute |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103345512A (en) |
Cited By (39)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103746898A (en) * | 2013-12-25 | 2014-04-23 | 新浪网技术(中国)有限公司 | Sampling analysis based e-mail sending method and system |
| CN103996088A (en) * | 2014-06-10 | 2014-08-20 | 苏州工业职业技术学院 | Advertisement click-through rate prediction method based on multi-dimensional feature combination logical regression |
| CN104035994A (en) * | 2014-06-11 | 2014-09-10 | 华东师范大学 | Prediction method of television play on-demand amount based on network data |
| 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 |
| CN104281634A (en) * | 2014-03-13 | 2015-01-14 | 电子科技大学 | Neighborhood-based mobile subscriber basic attribute forecasting method |
| CN104574124A (en) * | 2013-10-10 | 2015-04-29 | 深圳市腾讯计算机系统有限公司 | Method and device for determining exhibition effect of advertisement data |
| CN104794631A (en) * | 2015-03-31 | 2015-07-22 | 北京奇艺世纪科技有限公司 | Verification method and device for advertisement putting effect |
| CN105023170A (en) * | 2015-06-26 | 2015-11-04 | 深圳市腾讯计算机系统有限公司 | Processing method and device of click stream data |
| 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 |
| CN105162643A (en) * | 2015-06-30 | 2015-12-16 | 天津车之家科技有限公司 | Flow estimation method and device, and calculating equipment |
| CN105590240A (en) * | 2015-12-30 | 2016-05-18 | 合一网络技术(北京)有限公司 | Discrete calculating method of brand advertisement effect optimization |
| CN105608604A (en) * | 2015-12-30 | 2016-05-25 | 合一网络技术(北京)有限公司 | Continuous calculation method of brand advertisement effectiveness optimization |
| CN105760530A (en) * | 2016-03-03 | 2016-07-13 | 北京百度网讯科技有限公司 | Information card sequencing method and device based on artificial intelligence |
| CN105956888A (en) * | 2016-05-31 | 2016-09-21 | 北京创意魔方广告有限公司 | Advertisement personalized display method |
| CN106295832A (en) * | 2015-05-12 | 2017-01-04 | 阿里巴巴集团控股有限公司 | Product information method for pushing and device |
| CN106445954A (en) * | 2015-08-07 | 2017-02-22 | 北京奇虎科技有限公司 | Business object display method and apparatus |
| CN106688215A (en) * | 2014-06-27 | 2017-05-17 | 谷歌公司 | Automated click type selection for content performance optimization |
| CN107423992A (en) * | 2016-05-23 | 2017-12-01 | 北京易车互联信息技术有限公司 | Determine the method and device of the prediction model of ad click rate |
| CN104102819B (en) * | 2014-06-27 | 2017-12-19 | 北京奇艺世纪科技有限公司 | A kind of determination method and apparatus of user's natural quality |
| CN107516235A (en) * | 2016-06-17 | 2017-12-26 | 阿里巴巴集团控股有限公司 | Commodity preference predictor method and device |
| CN107578294A (en) * | 2017-09-28 | 2018-01-12 | 北京小度信息科技有限公司 | User's behavior prediction method, apparatus and electronic equipment |
| 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 |
| WO2018018279A1 (en) * | 2016-07-24 | 2018-02-01 | 金蕾 | Method for pushing information when feeding back ad serving object and data collecting device |
| WO2018018277A1 (en) * | 2016-07-24 | 2018-02-01 | 金蕾 | Method for feeding back hotel advertisement data and data collecting device |
| WO2018018278A1 (en) * | 2016-07-24 | 2018-02-01 | 金蕾 | Advertisement feedback technology usage count statistical method and a data acquisition device |
| CN107808295A (en) * | 2016-09-09 | 2018-03-16 | 腾讯科技（深圳）有限公司 | Multi-medium data put-on method and device |
| CN108073794A (en) * | 2016-11-18 | 2018-05-25 | 百度在线网络技术（北京）有限公司 | Method for authenticating and device |
| CN108369433A (en) * | 2015-09-18 | 2018-08-03 | Mms美国控股有限公司 | Micro- moment analysis |
| CN108573392A (en) * | 2017-03-09 | 2018-09-25 | 合网络技术（北京）有限公司 | A kind of advertisement placement method and device |
| CN108629608A (en) * | 2017-03-22 | 2018-10-09 | 腾讯科技（深圳）有限公司 | User data processing method and processing device |
| CN109033219A (en) * | 2018-06-29 | 2018-12-18 | 北京奇虎科技有限公司 | Point of interest POI classification method and device |
| CN110070391A (en) * | 2019-04-17 | 2019-07-30 | 同盾控股有限公司 | Data processing method, device, computer-readable medium and electronic equipment |
| 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 |
| CN110634006A (en) * | 2018-06-22 | 2019-12-31 | 广州优视网络科技有限公司 | Method, device and equipment for predicting advertisement click rate and readable storage medium |
| CN111461795A (en) * | 2020-05-02 | 2020-07-28 | 上海佳投互联网技术集团有限公司 | Advertisement click effect prediction method and system |
| CN111523202A (en) * | 2020-03-20 | 2020-08-11 | 北京国电通网络技术有限公司 | Analysis report generation method and device based on full index analysis scene pool |
| CN112533028A (en) * | 2020-12-10 | 2021-03-19 | 杭州次元岛科技有限公司 | Publishing system for live broadcast advertisement |
| CN113034167A (en) * | 2019-12-24 | 2021-06-25 | 上海佳投互联网技术集团有限公司 | User interest analysis method and advertisement delivery method based on user behaviors |
| US11887164B2 (en) | 2015-05-26 | 2024-01-30 | Microsoft Technology Licensing, Llc | Personalized information from venues of interest |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101192235A (en) * | 2007-04-11 | 2008-06-04 | 腾讯科技（深圳）有限公司 | Method, system and equipment for delivering advertisement based on user feature |
| US20090119177A1 (en) * | 2007-11-05 | 2009-05-07 | John Thomas K | Outdoor and out of home advertising method and system |
| CN102346899A (en) * | 2011-10-08 | 2012-02-08 | 亿赞普(北京)科技有限公司 | Method and device for predicting advertisement click rate based on user behaviors |
| CN102592235A (en) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | Internet advertisement serving system |
| CN102592236A (en) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | Internet advertising crowd analysis system and analysis method |
| CN102663026A (en) * | 2012-03-22 | 2012-09-12 | 浙江盘石信息技术有限公司 | Implementation method for directionally running internet advertisements |
| 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 |
| CN102722832A (en) * | 2012-01-05 | 2012-10-10 | 合一网络技术(北京)有限公司 | Online video advertisement refinement targeting delivery method |
-
2013
- 2013-07-06 CN CN2013102831999A patent/CN103345512A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101192235A (en) * | 2007-04-11 | 2008-06-04 | 腾讯科技（深圳）有限公司 | Method, system and equipment for delivering advertisement based on user feature |
| US20090119177A1 (en) * | 2007-11-05 | 2009-05-07 | John Thomas K | Outdoor and out of home advertising method and system |
| CN102346899A (en) * | 2011-10-08 | 2012-02-08 | 亿赞普(北京)科技有限公司 | Method and device for predicting advertisement click rate based on user behaviors |
| CN102592235A (en) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | Internet advertisement serving system |
| CN102592236A (en) * | 2011-12-28 | 2012-07-18 | 北京品友互动信息技术有限公司 | Internet advertising crowd analysis system and analysis method |
| CN102722832A (en) * | 2012-01-05 | 2012-10-10 | 合一网络技术(北京)有限公司 | Online video advertisement refinement targeting delivery method |
| 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 |
| CN102663026A (en) * | 2012-03-22 | 2012-09-12 | 浙江盘石信息技术有限公司 | Implementation method for directionally running internet advertisements |
Non-Patent Citations (1)
| Title |
|---|
| 王海雷等: "搜索引擎广告用户行为预测与特征分析", 《计算机应用研究》 * |
Cited By (54)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104574124A (en) * | 2013-10-10 | 2015-04-29 | 深圳市腾讯计算机系统有限公司 | Method and device for determining exhibition effect of advertisement data |
| CN104574124B (en) * | 2013-10-10 | 2018-07-17 | 深圳市腾讯计算机系统有限公司 | Determine the method and device of the bandwagon effect of ad data |
| CN103746898A (en) * | 2013-12-25 | 2014-04-23 | 新浪网技术(中国)有限公司 | Sampling analysis based e-mail sending method and system |
| CN104281634A (en) * | 2014-03-13 | 2015-01-14 | 电子科技大学 | Neighborhood-based mobile subscriber basic attribute forecasting method |
| CN104281634B (en) * | 2014-03-13 | 2018-04-20 | 电子科技大学 | A kind of mobile subscriber's primary attribute Forecasting Methodology based on neighborhood |
| CN103996088A (en) * | 2014-06-10 | 2014-08-20 | 苏州工业职业技术学院 | Advertisement click-through rate prediction method based on multi-dimensional feature combination logical regression |
| CN104035994A (en) * | 2014-06-11 | 2014-09-10 | 华东师范大学 | Prediction method of television play on-demand amount based on network data |
| CN104035994B (en) * | 2014-06-11 | 2017-04-12 | 华东师范大学 | Prediction method of television play on-demand amount based on network data |
| CN104102819B (en) * | 2014-06-27 | 2017-12-19 | 北京奇艺世纪科技有限公司 | A kind of determination method and apparatus of user's natural quality |
| CN106688215A (en) * | 2014-06-27 | 2017-05-17 | 谷歌公司 | Automated click type selection for content performance optimization |
| 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 |
| 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 |
| CN104794631A (en) * | 2015-03-31 | 2015-07-22 | 北京奇艺世纪科技有限公司 | Verification method and device for advertisement putting effect |
| CN106295832A (en) * | 2015-05-12 | 2017-01-04 | 阿里巴巴集团控股有限公司 | Product information method for pushing and device |
| CN106295832B (en) * | 2015-05-12 | 2020-05-19 | 阿里巴巴集团控股有限公司 | Product information pushing method and device |
| US11887164B2 (en) | 2015-05-26 | 2024-01-30 | Microsoft Technology Licensing, Llc | Personalized information from venues of interest |
| CN105023170A (en) * | 2015-06-26 | 2015-11-04 | 深圳市腾讯计算机系统有限公司 | Processing method and device of click stream data |
| CN105162643A (en) * | 2015-06-30 | 2015-12-16 | 天津车之家科技有限公司 | Flow estimation method and device, and calculating equipment |
| CN105162643B (en) * | 2015-06-30 | 2018-04-27 | 天津车之家科技有限公司 | The method, apparatus and computing device that flow is estimated |
| CN106445954A (en) * | 2015-08-07 | 2017-02-22 | 北京奇虎科技有限公司 | Business object display method and apparatus |
| CN108369433A (en) * | 2015-09-18 | 2018-08-03 | Mms美国控股有限公司 | Micro- moment analysis |
| 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 |
| 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 |
| CN105590240A (en) * | 2015-12-30 | 2016-05-18 | 合一网络技术(北京)有限公司 | Discrete calculating method of brand advertisement effect optimization |
| CN105608604A (en) * | 2015-12-30 | 2016-05-25 | 合一网络技术(北京)有限公司 | Continuous calculation method of brand advertisement effectiveness optimization |
| CN105760530A (en) * | 2016-03-03 | 2016-07-13 | 北京百度网讯科技有限公司 | Information card sequencing method and device based on artificial intelligence |
| CN107423992A (en) * | 2016-05-23 | 2017-12-01 | 北京易车互联信息技术有限公司 | Determine the method and device of the prediction model of ad click rate |
| CN105956888A (en) * | 2016-05-31 | 2016-09-21 | 北京创意魔方广告有限公司 | Advertisement personalized display method |
| CN107516235A (en) * | 2016-06-17 | 2017-12-26 | 阿里巴巴集团控股有限公司 | Commodity preference predictor method and device |
| WO2018018278A1 (en) * | 2016-07-24 | 2018-02-01 | 金蕾 | Advertisement feedback technology usage count statistical method and a data acquisition device |
| WO2018018277A1 (en) * | 2016-07-24 | 2018-02-01 | 金蕾 | Method for feeding back hotel advertisement data and data collecting device |
| WO2018018279A1 (en) * | 2016-07-24 | 2018-02-01 | 金蕾 | Method for pushing information when feeding back ad serving object and data collecting device |
| CN107808295A (en) * | 2016-09-09 | 2018-03-16 | 腾讯科技（深圳）有限公司 | Multi-medium data put-on method and device |
| CN107808295B (en) * | 2016-09-09 | 2021-06-11 | 腾讯科技（深圳）有限公司 | Multimedia data delivery method and device |
| CN108073794A (en) * | 2016-11-18 | 2018-05-25 | 百度在线网络技术（北京）有限公司 | Method for authenticating and device |
| CN108573392A (en) * | 2017-03-09 | 2018-09-25 | 合网络技术（北京）有限公司 | A kind of advertisement placement method and device |
| CN108629608B (en) * | 2017-03-22 | 2023-02-24 | 腾讯科技（深圳）有限公司 | User data processing method and device |
| CN108629608A (en) * | 2017-03-22 | 2018-10-09 | 腾讯科技（深圳）有限公司 | User data processing method and processing device |
| CN107578294A (en) * | 2017-09-28 | 2018-01-12 | 北京小度信息科技有限公司 | User's behavior prediction method, apparatus and electronic 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 |
| 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 |
| 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 |
| CN110634006A (en) * | 2018-06-22 | 2019-12-31 | 广州优视网络科技有限公司 | Method, device and equipment for predicting advertisement click rate and readable storage medium |
| CN109033219A (en) * | 2018-06-29 | 2018-12-18 | 北京奇虎科技有限公司 | Point of interest POI classification method and device |
| CN109033219B (en) * | 2018-06-29 | 2022-03-11 | 北京奇虎科技有限公司 | Point of interest (POI) classification method and device |
| CN110070391A (en) * | 2019-04-17 | 2019-07-30 | 同盾控股有限公司 | Data processing method, device, computer-readable medium and electronic equipment |
| CN110070391B (en) * | 2019-04-17 | 2020-06-19 | 同盾控股有限公司 | Data processing method and device, computer readable medium and electronic equipment |
| CN113034167A (en) * | 2019-12-24 | 2021-06-25 | 上海佳投互联网技术集团有限公司 | User interest analysis method and advertisement delivery method based on user behaviors |
| CN111523202A (en) * | 2020-03-20 | 2020-08-11 | 北京国电通网络技术有限公司 | Analysis report generation method and device based on full index analysis scene pool |
| CN111523202B (en) * | 2020-03-20 | 2024-01-09 | 北京国电通网络技术有限公司 | Analysis report generation method and device based on full-scale index analysis scene pool |
| CN111461795A (en) * | 2020-05-02 | 2020-07-28 | 上海佳投互联网技术集团有限公司 | Advertisement click effect prediction method and system |
| CN112533028A (en) * | 2020-12-10 | 2021-03-19 | 杭州次元岛科技有限公司 | Publishing system for live broadcast advertisement |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103345512A (en) | Online advertising click-through rate forecasting method and device based on user attribute | |
| CN103714139B (en) | Parallel data mining method for identifying a mass of mobile client bases | |
| CN102708131B (en) | By consumer's automatic classification in fine point | |
| CN102737334B (en) | Micro-segment definition system | |
| US8655695B1 (en) | Systems and methods for generating expanded user segments | |
| JP6271345B2 (en) | Extraction apparatus, extraction method, and extraction program | |
| CN104834641B (en) | The processing method and related system of network media information | |
| CN110942337A (en) | Accurate tourism marketing method based on internet big data | |
| CN110222267A (en) | A kind of gaming platform information-pushing method, system, storage medium and equipment | |
| CN102110265A (en) | Network advertisement effect estimating method and network advertisement effect estimating system | |
| US20140122245A1 (en) | Method for audience profiling and audience analytics | |
| CN102592235A (en) | Internet advertisement serving system | |
| CN106355442A (en) | Online precise advertising method and system based on big data driving | |
| US10949889B2 (en) | Methods and apparatus for managing models for classification of online users | |
| CN103177129B (en) | Internet real-time information recommendation prognoses system | |
| JP2016517094A (en) | Systems and methods for audience targeting | |
| CN109949089B (en) | Method, device and terminal for determining display rate | |
| CN102592236A (en) | Internet advertising crowd analysis system and analysis method | |
| CN103034718A (en) | Target data sequencing method and target data sequencing device | |
| JP2012088994A (en) | Advertising effect analysis system and method of the same | |
| JP6182478B2 (en) | Analysis apparatus and analysis method | |
| CN103198098A (en) | Network information transfer method and device | |
| JP2019053668A (en) | Device, method, and program for granting privilege | |
| CN111489190A (en) | Anti-cheating method and system based on user relationship | |
| TWI657395B (en) | Opinion leader related network-based trading system, method, and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20131009 |