CN104298719B - Category division, advertisement placement method and the system of user is carried out based on Social behaviors - Google Patents
Category division, advertisement placement method and the system of user is carried out based on Social behaviors Download PDFInfo
- Publication number
- CN104298719B CN104298719B CN201410492126.5A CN201410492126A CN104298719B CN 104298719 B CN104298719 B CN 104298719B CN 201410492126 A CN201410492126 A CN 201410492126A CN 104298719 B CN104298719 B CN 104298719B
- Authority
- CN
- China
- Prior art keywords
- user
- attributive character
- social
- behavior
- group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/955—Retrieval from the web using information identifiers, e.g. uniform resource locators [URL]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Abstract
The invention discloses a kind of category division, advertisement placement method and system that user is carried out based on Social behaviors, this method includes:According to the UID of user to be sorted, obtain to should be in the setting cycle of UID records social daily record, and as the user social contact data of the user;Cookie ID according to corresponding to the UID, obtain to should be in the setting cycle of cookie ID records access log, and as the user behavior data of the user;User behavior attributive character is extracted from the user behavior data, user social contact attributive character is extracted from the user social contact data;According to the user behavior attributive character and user social contact attributive character of extraction, the user is divided into respective classes.Using the present invention, the degree of accuracy and the advertisement putting validity of the category division of user can be improved.
Description
Technical field
The present invention relates to internet arena, more particularly to a kind of category division, advertisement that user is carried out based on Social behaviors
Put-on method and system.
Background technology
With the continuous development of Internet technology, Internet advertising is taken rapidly with the advantage that its is quick, convenient, flexibility is strong
For traditional media advertisement.Internet advertising is exactly that advertisement is launched on network by web advertisement platform, using on website
Ad banner, text link, multimedia method, internet publication or releasing advertisements, pass through network delivery to internet
A kind of high-tech advertising campaign mode of user.With traditional four big communications media (newspaper, magazine, TV, broadcast) advertisement and
The outdoor advertising shown appreciation for somebody is enjoyed to compare, Internet advertising has advantageous advantage using internet as advertising media, is to implement
An important part for Modern Marketing media strategy.
Moreover, with the high speed development of internet advertisement technology, advertiser is to the needs of target audience also more and more higher.And
Traditional media advertisement is not that life cycle is shorter, it is complicated to make, and is exactly that flexibility and promptness are poor, at the same also easily by
To the interference of extraneous factor;The audient crowd that more mainly they can not be required for accurate lock oneself, therefore offer one is provided badly
Kind can towards crowd, lifting the advertisement mode that promotion effect greatly promotes.
In practical application, IGRP (Internet Gross Rating Points, the total rating point in internet) advertisement is exactly
The advertisement towards crowd that a kind of GRP (Gross Rating Point, total rating point) from traditional tv advertisement is developed
Sell scheme.It can go deep into excavating the features such as the age of audient user, sex, interest by internet big data advantage, then
Category division is carried out to user according to the feature of excavation, so that advertiser targetedly carries out advertisement putting, lifting is promoted
Effect.
Existing to provide a kind of class of subscriber division methods, it is mainly according to access day of the user in advertisement putting website
Will, user is divided into corresponding classification.Specifically, user can obtained after the access log of advertisement putting website, it is right
The page that user accesses is analyzed, and obtains the feature of the page of user's access;Then, the spy of the page accessed according to user
Sign, it is inferred to the classification of user.However, in fact, the feature of the page accessed according to one-side user pushes away user's come counter
Classification, its confidence level is not high, causes the degree of accuracy of the category division result of user not high, then causes according to category division result
Audient's less pertinence, the promotion effect for the advertisement launched are poor, reduce the validity of advertisement putting.
Therefore, it is necessary to provide a kind of class of subscriber division methods for improving the category division degree of accuracy.
The content of the invention
In view of the above-mentioned drawbacks of the prior art, the invention provides a kind of classification that user is carried out based on Social behaviors
Division, advertisement placement method and system, to improve the degree of accuracy of the category division of user and advertisement putting validity.
According to an aspect of the invention, there is provided a kind of category classification method that user is carried out based on Social behaviors, bag
Include:
Obtain the user behavior data of user to be sorted in the setting cycle, and user social contact data;
User behavior attributive character is extracted from the user behavior data, user is extracted from the user social contact data
Social attribute feature;
According to the user behavior attributive character and user social contact attributive character of extraction, the user is divided into respective class
Not.
It is preferred that the user behavior attributive character and user social contact attributive character according to extraction, the user is drawn
Respective classes are assigned to, are specifically included:
The user behavior category that will be recorded in the user behavior attributive character of the user, with the behavioural characteristic rule base of built in advance
Property feature is matched, the behavior group according to belonging to matching result judges the user;Wherein, the behavioural characteristic rule
Every kind of behavior group is directed in storehouse, have recorded user behavior attributive character possessed by behavior group;
The user social contact category that will be recorded in the user social contact attributive character of the user, with the social characteristics rule base of built in advance
Property feature is matched, the social group according to belonging to matching result judges the user;Wherein, the social characteristics rule
Every kind of social group is directed in storehouse, have recorded user social contact attributive character possessed by the social group;
Group common between the behavior group belonging to the user and social group is found out, and the user is divided
Into the common group found out.
It is preferred that the user behavior attributive character and user social contact attributive character according to extraction, the user is drawn
Respective classes are assigned to, are specifically included:
The user behavior attributive character of the user of extraction and user social contact attributive character are obtained by convergence described
The user characteristics of user;
The user characteristics of the obtained user is input in the category division model for category division;By described
The user is divided into respective classes by category division model.
It is preferred that the category division model is training in advance:
For each training user selected from the user of website, the user behavior attribute for counting the training user is special
After user social contact attributive character of seeking peace, the user characteristics of the training user is obtained by convergence;And according to the use of the training user
Family behavior property feature determines the behavior group belonging to the training user, according to the user social contact attributive character of the training user
The social group belonging to the training user is determined, and will be common between the behavior group belonging to the training user and social group
Affiliated group of the group as the training user;
Using the affiliated group of each training user, user characteristics as training data, using the training data with pre-
The multi-tag multi-classification algorithm put carries out model training, obtains category division model.
It is preferred that it is described obtain category division model before, in addition to:
For each test user selected from the user of website, the user behavior attribute for counting test user is special
After user social contact attributive character of seeking peace, the user characteristics of test user is obtained by convergence;And according to the use of test user
Family behavior property feature determines the behavior group belonging to test user, according to the user social contact attributive character of test user
The social group belonging to test user is determined, and will be common between the behavior group belonging to test user and social group
Affiliated group of the group as test user;
Using the affiliated group of each test user, user characteristics as test data;And
When carrying out model training with preset multi-tag multi-classification algorithm using the training data, also using described
Test data carries out model training with preset multi-tag multi-classification algorithm, obtains the category division model.
Present invention also offers a kind of advertisement placement method, including:
After website receives the access request of user, the identification number cookieID and correspondingly of the user is obtained
User identity prove UID;
According to the cookieID and UID of the user, using the method as described in claim 1-5 is any, obtain described
The category division result of user;
Using obtained category division result as search key, the advertisement to match with the search key is found out
Content;
The advertisement found out is played in the advertisement played column of the webpage returned for the access request to the user
Content.
According to another aspect of the present invention, a kind of category division system that user is carried out based on Social behaviors is additionally provided
System, including:
User data acquisition module, for obtaining user behavior data, the Yi Jiyong of user to be sorted in the setting cycle
Family social data simultaneously exports;
Attributive character abstraction module, used for being extracted in the user behavior data that is exported from the user data acquisition module
Family behavior property feature, it is special that user social contact attribute is extracted from the user social contact data of user data acquisition module output
Sign;
Class of subscriber division module, for the user behavior attributive character that is extracted according to the attributive character abstraction module and
User social contact attributive character, the user is divided into respective classes.
It is preferred that the class of subscriber division module specifically includes:First class of subscriber division unit or second user class
Other division unit;Wherein,
The first class of subscriber division unit specifically includes:
Behavioural characteristic coupling subelement, for the user behavior attributive character for extracting the attributive character abstraction module,
Matched with the user behavior attributive character recorded in the behavioural characteristic rule base of built in advance, according to being judged matching result
Behavior group belonging to user;Wherein, every kind of behavior group is directed in the behavioural characteristic rule base, have recorded behavior group
Possessed user behavior attributive character;
Social characteristics coupling subelement, for the user social contact attributive character for extracting the attributive character abstraction module,
Matched with the user social contact attributive character recorded in the social characteristics rule base of built in advance, according to being judged matching result
Social group belonging to user;Wherein, every kind of social group is directed in the social characteristics rule base, have recorded the social group
Possessed user social contact attributive character;
Category division subelement, the behavior group belonging to user exported for receiving the behavioural characteristic matching unit,
And the social group belonging to the user of the social characteristics matching unit output;Find out the behavior group belonging to the user
The common group between social group, and the user is divided into the common group found out;And
The second user category division unit is used for the use for the user for extracting the attributive character abstraction module
Family behavior property feature and user social contact attributive character, the user characteristics of the user is obtained by convergence;And the institute that will be obtained
The user characteristics for stating user is input in the category division model for category division;By the category division model by described in
User is divided into respective classes.
It is preferred that the category division system of user is carried out based on Social behaviors also to be included:
Partitioning model training module, for each training user for being selected from the user of website, count the instruction
After the user behavior attributive character and user social contact attributive character of practicing user, the user that the training user is obtained by convergence is special
Sign;And the behavior group according to belonging to the user behavior attributive character of the training user determines the training user, according to the instruction
The user social contact attributive character of white silk user determines the social group belonging to the training user, and the row belonging to by the training user
Affiliated group of the common group as the training user between group and social group;By the affiliated race of each training user
Group, user characteristics carry out model instruction using the training data as training data with preset multi-tag multi-classification algorithm
Practice, obtain category division model.
Present invention also offers a kind of advertisement delivery system, including:
User data acquisition module, attributive character in the above-mentioned category division system that user is carried out based on Social behaviors are taken out
Modulus block, class of subscriber division module;
Web-page requests receiving module, after the access request for receiving user, obtain the cookieID and correspondingly of user
UID, and send to the user data acquisition module;
Advertising inquiry module, for the category division result of user that exports the class of subscriber division module as searching
Rope keyword, find out the ad content to match with the search key;
Request processing module, the advertisement is inserted in the advertisement played column for the webpage asked in the access request
After the ad content that enquiry module is found out, the webpage is returned to the user;
The user data acquisition module is specifically used for cookieID and corresponding UID according to user, obtains setting
The user behavior data of user and user social contact data and exported in cycle.
In technical scheme, according to user behavior data of the user on website, the user under social platform
Social data, user behavior attributive character and user social contact attributive character are extracted respectively;Then, according to the use extracted
Family behavior property feature and user social contact attributive character carry out the category division of user.Compared to existing page is accessed according only to user
The information in face carrys out the anti-interest for pushing away user, technical scheme provided by the invention, considers user behavior attributive character and user
Social attribute feature carries out class of subscriber division, adds the feature rich degree of category division, and improves class of subscriber and draw
Divide the degree of accuracy, also can improves crowd's positional accuracy;So, based on higher crowd's positional accuracy, can improve wide
The specific aim of the audient of announcement, promotional videos etc., launch validity.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the category classification method that user is carried out based on Social behaviors of the embodiment of the present invention;
Fig. 2 is the schematic flow sheet of the advertisement placement method of the embodiment of the present invention;
Fig. 3 is the schematic flow sheet for extracting user behavior attributive character of the embodiment of the present invention;
Fig. 4 is the schematic flow sheet for extracting user social contact attributive character of the embodiment of the present invention;
Fig. 5 is the training method schematic flow sheet of the category division model of the embodiment of the present invention;
Fig. 6 is the structural representation of the category division system that user is carried out based on Social behaviors of the embodiment of the present invention;
Fig. 7 is the structural representation of the first class of subscriber division unit of the embodiment of the present invention;
Fig. 8 is the structural representation of the advertisement delivery system of the embodiment of the present invention.
Embodiment
Clear, complete description is carried out to technical scheme below with reference to accompanying drawing, it is clear that described implementation
Example is only the part of the embodiment of the present invention, rather than whole embodiments.It is general based on the embodiment in the present invention, this area
Logical technical staff all other embodiment resulting on the premise of creative work is not made, belongs to the present invention and is protected
The scope of shield.
The term such as " module " used in this application, " system " is intended to include the entity related to computer, such as but unlimited
In hardware, firmware, combination thereof, software or executory software.For example, module can be, it is not limited to:Processing
The process run on device, processor, object, executable program, thread, program and/or the computer performed.For example, count
It can be module to calculate the application program run in equipment and this computing device.One or more modules can be located at executory
In one process and/or thread, a module can also be located on a computer and/or be distributed in two or more platforms and calculate
Between machine.
, can be with except the situation of the accession page of user on the web site it was found by the inventors of the present invention that in website
Reflect the interest of user to a certain extent, it is that user issues in the social platform of the website log, forwarding, thumb up it is rich
Text can also react the interest of user to a certain extent.
Therefore, the present inventor is it is considered that can be using access log of the user on website as user behavior number
According to, using social daily record of the user under social platform as user social contact data, and from user behavior data, user social contact data
User behavior attributive character and user social contact attributive character are extracted respectively;Then, can be according to the user behavior extracted
Attributive character and user social contact attributive character carry out the category division of user.Compared to the existing letter according only to user to access pages
Breath carrys out the anti-interest for pushing away user, technical scheme provided by the invention, considers user behavior attributive character and user social contact category
Property feature carry out class of subscriber division, the feature rich degree of category division can be increased, and it is accurate to improve class of subscriber division
Spend, also can improves crowd's positional accuracy;Based on higher crowd's positional accuracy, advertisement, promotional videos can be improved
Deng audient specific aim, launch validity.
The technical scheme that the invention will now be described in detail with reference to the accompanying drawings.
The embodiments of the invention provide a kind of category classification method that user is carried out based on Social behaviors, its flow such as Fig. 1
It is shown, specifically it may include steps of:
S101:Obtain the user behavior data of user to be sorted in the setting cycle, and user social contact data.
In practical application, for user when sending access request to website, website would generally pass through cookie ID
(Identity, identification number) marks user, and records all access logs of the user.Correspondingly, if user exists
The social platform of the website is registered, and can obtain unique UID (User Identification, user identity prove);
So, when the user logs in social platform by UID every time, website can to should UID, record the cookie of the user
ID and all social daily records.Social platform in the embodiment of the present invention can be specifically microblogging, push away top grade.
Therefore, specifically, then after determining the UID of user to be sorted, be able to can be obtained pair according to the UID of the user
Social daily record that should be in the setting cycle of UID records, and as the user social contact data of the user.It is possible to further root
According to the cookie ID corresponding to the UID, obtain to should be in the setting cycle of cookie ID records access log, and conduct
The user behavior data of the user.
Wherein, access log of the user within the setting cycle can specifically include:User is within the setting cycle on website
Each page and its access times, each advertisement of click and its number of clicks for accessing etc..Society of the user within the setting cycle
Daily record is handed over specifically to include:The blog article of user, the log-on message of user, the concern information of user, the LBS of user
(Location Based Service, based on location-based service) positional information etc..
S102:User behavior attributive character is extracted from user behavior data, user society is extracted from user social contact data
Hand over attributive character.
Specifically, the user behavior attributive character of user, example according to the user behavior data of acquisition, can therefrom be extracted
Such as, the sex of user, age, interest etc..It is possible to further the user social contact data according to acquisition, user is therefrom extracted
User social contact attributive character, for example, sex, age, blog article feature, concern user characteristics, LBS features etc..In practical application,
The user behavior attributive character extracted from user behavior data is referred to as the feature based on cookie ID;From user
The user social contact attributive character extracted in social data is referred to as the feature based on UID.
On how to extract above-mentioned user behavior attributive character from user behavior data, and how from user social contact
Above-mentioned user social contact attributive character is extracted in data, will be subsequently discussed in detail.
S103:According to the user behavior attributive character and user social contact attributive character of extraction, user is divided into respective class
Not.
Specifically, the use that will can be recorded in the user behavior attributive character of user, with the behavioural characteristic rule base of built in advance
Family behavior property feature is matched, the behavior group according to belonging to matching result judges user.Wherein, behavioural characteristic rule
Every kind of behavior group is directed in storehouse, have recorded user behavior attributive character possessed by behavior group.Correspondingly, will can obtain
The user social contact attributive character recorded in the user social contact attributive character of the user taken, with the social characteristics rule base of built in advance is carried out
Matching, the social group according to belonging to matching result judges user.Wherein, every kind of social race is directed in social characteristics rule base
Group, have recorded user social contact attributive character possessed by the social group.
In the embodiment of the present invention, the social group in behavior group and social characteristics rule base in behavioural characteristic rule base
All pre-set, such as can include:Investment & Financing group, sport and body-building group, city white collar group, race of parent-offspring family
Group, youth campus group, video display group, photography group etc..
If interested in certain class things in view of user, its access navigation patterns on and Social behaviors on would generally table
Reveal similitude.Therefore, group common between the behavior group belonging to user and social group can be searched, if looking into
Find out group common between the behavior group belonging to user and social group, then user can be divided into find out it is common
Group in.Compared to the existing classification divided based on one-side behavior, in the solution of the present invention, by based on cookie
Common group is as user between behavior group and social group that ID feature and feature based on UID are judged respectively
Category division result, the degree of accuracy is higher.
, can be for every kind of behavior group in behavioural characteristic rule base in practical application, the user behavior that will be extracted
Attributive character is matched with user behavior attributive character possessed by behavior group, if the match is successful, can be determined that this
Behavior group is the behavior group belonging to the user;If mismatching, it can be determined that the user is not belonging to behavior group.Accordingly
Ground, the every kind of social group that can be directed in social characteristics rule base, by the user social contact attributive character extracted and the social activity
User social contact attributive character is matched possessed by group, if the match is successful, can be determined that the social group is the user
Affiliated social group;If mismatching, it can be determined that the user is not belonging to the social group.
Further, it is directed to every kind of behavior group in behavioural characteristic rule base, user's row possessed by behavior group
It can specifically include several prerequisite user behavior attributive character of user for belonging to behavior group for attributive character,
The user behavior attributive character that some users for belonging to behavior group generally possess can also be included.
Therefore, the user behavior attributive character extracted and user behavior attributive character possessed by behavior group are entered
During row matching, if the user behavior attributive character extracted includes the prerequisite user of user institute for belonging to behavior group
Behavior property feature, then it may determine that the match is successful;Otherwise, judge to match unsuccessful.Correspondingly, in social characteristics rule base
For every kind of social group, user social contact attributive character possessed by the social group specifically can belong to the society including several
The prerequisite user social contact attributive character of user institute of group is handed over, some user institutes for belonging to the social group can also be included
The user social contact attributive character generally possessed.Therefore, the user social contact attributive character extracted and the social group are had
User social contact attributive character when being matched, if the user social contact attributive character extracted, which includes, belongs to the social group
The prerequisite user social contact attributive character of user institute, then may determine that the match is successful;Otherwise, judge to match unsuccessful.
In addition, if common group is not present between behavior group and social group belonging to user, will can extract
User behavior attributive character and user social contact the attributive character fusion of the user gone out collects, and obtains the user characteristics of user.Then,
According to the user characteristics of obtained user and the category division model of training in advance, user is divided into respective classes.Specifically
Ground, multiple Fusion Features that can be nearly adopted are a user characteristics;By the feature of not near justice separately as a user spy
Sign;It is a user characteristics by multiple Fusion Features of contradiction according to default fusion rule.If for example, user extracted
Age characteristics in behavior property feature contradicts with the age characteristics in the user social contact attributive character extracted, then can be by
According to the default fusion rule on the age, by the age characteristics in user social contact attributive character directly as in user characteristics
Age characteristics.
Present invention also offers a kind of more excellent embodiment, is carrying out step S102:Extracted from user behavior data
User behavior attributive character, can be directly by the use of extraction after extracting user social contact attributive character from user social contact data
The user behavior attributive character and user social contact attributive character at family carry out fusion and collected, and obtain the user characteristics of user;Then, may be used
So that the user characteristics of obtained user is input in the category division model for category division of training in advance;Pass through classification
User is divided into respective classes by partitioning model.
In the embodiment of the present invention, on the training of category division model, will subsequently it be discussed in detail.
In practical application, in the user behavior attributive character and user social contact attributive character according to extraction, user is divided
To after respective classes, the classification that user is divided into can be stored as category division result.In view of practical application
In the user in different computer Website logins, website to should user record cookie ID it is different, that is to say, that it is right
Same user is answered to there may be multiple cookie ID;But the UID of same user's existence anduniquess;Therefore, further, may be used
With by the category division result of user it is corresponding with the UID of user storage, be easy to subsequent use.Certainly, in practical application, also may be used
With by the cookie ID of user and UID and the corresponding storage jointly of category division result.
Based on above-mentioned category classification method, present invention also offers a kind of advertisement placement method, its flow as shown in Fig. 2
Specifically it may include steps of:
S201:After website receives the access request of user, the cookie ID and corresponding UID of user are obtained.
Specifically, it is corresponding with access request except can normally obtain after website receives the access request of user
Webpage;Can be so that the cookie ID of the user and its corresponding UID can be obtained.
S202:According to the cookie ID and UID of acquisition, the user behavior data for setting user in the cycle, Yi Jiyong are obtained
Family social data.
S203:User behavior attributive character is extracted from the user behavior data of acquisition, from the user social contact data of acquisition
Middle extraction user social contact attributive character;And according to the user behavior attributive character and user social contact attributive character of extraction, by user
Respective classes are divided into, obtain the category division result of user.
More preferably, website receives the access request of user, after the cookie ID and UID that obtain user, can directly from
Category division result corresponding with the cookie ID and UID that obtain is found out in class of subscriber division result storehouse, as the user
Category division result.Wherein, each user is corresponded in class of subscriber division result storehouse, be stored with according to such as step S101-
The category division result and corresponding cookie ID and UID that S103 method marks off.
S204:Using obtained category division result as search key, find out match with search key it is wide
Accuse content;The ad content found out is played in the advertisement played column of the webpage returned for access request to user.
Specifically, method well-known to those skilled in the art can be used, can be from net after search key is obtained
Stand and find out the ad content to match with search key from the background;And the ad content of all acquisitions is added to for accessing
Ask into the advertisement played column of the webpage of user's return.So, when user is accessing the webpage of website, website can basis
The category division result of user pushes the possible advertisement interested of user to user, improves the precision of advertisement putting.
In the embodiment of the present invention, before category division is carried out to user, each page of website, root can be directed in advance
According to the content of the page and the characteristic of audient user, it is corresponding be stored with each page properties each page properties feature and its
Point.
Wherein, page properties can specifically include:Age, sex and hobby;There is difference under each page properties
Page properties feature.So, can be according to the characteristic of the audient user of the page for each page, the page is in difference
Page properties feature on there is corresponding score., can also be by page properties feature in the embodiment of the present invention for ease of describing
Score be referred to as page category feature point.
In practical application, the page properties feature (can also be referred to as age characteristics) at page properties-age specifically can be with
Including:10~20 years old or 21~30 years old or 31~40 years old or more than or equal to 41 years old.Such as a certain page for website, should
Page properties feature is scored at 0.1 in -10~20 years old corresponding to the page;Page properties feature -21~30 years old corresponding to the page
It is scored at 0.7;Page properties feature is scored at 0.1 in -31~40 years old corresponding to the page;Page properties corresponding to the page
Feature-more than or equal to 41 years old is scored at 0.1.
The page properties feature (can also be referred to as sex character) of page properties-sex can specifically include:Male, female
Property.Such as a certain page for website website, page properties feature-male is scored at 0.3 corresponding to the page;The page
Page properties feature-women is scored at 0.7 corresponding to face.
The page properties feature (can also be referred to as interest characteristics) of page properties-interest can specifically include;Reading, trip
Trip, house property, second-hand house, basketball, football, financing etc..Such as a certain page for website website, the page corresponding to the page
Attributive character-financing is scored at 0.5;Page properties feature-reading is scored at 0.3 corresponding to the page;The page is corresponding
Page properties feature-house property be scored at 0.2;And other page properties under page properties-interest corresponding to the page are special
The score of sign is 0.
Based on each page properties feature and its score corresponding to the above-mentioned each page prestored in advance, on above-mentioned steps
How what is referred in S102 extracts user behavior attributive character from user behavior data, as shown in figure 3, can specifically pass through
Following steps extract:
S301:From the user behavior data of acquisition, each page that user accesses within the setting cycle is determined.
Specifically, each page accessed from user behavior data except user can be determined within the setting cycle,
The access times of each page can also be determined.
S302:The each page accessed for user within the setting cycle, it is each under each page properties to obtain the page
Page feature and its score.
S303:According to the score of all page properties features of each page under each page properties, determine that user exists
Page feature under each page properties, and using the page feature determined as the user behavior attributive character extracted.
Specifically, for page properties-age, for each page properties feature under the page properties, (age is special
Sign), the score for the page properties feature that can be stored according to corresponding to each page of acquisition, calculate the page properties feature
The average value of score;Page feature of the average value highest page properties feature as user under the page properties is chosen, and
As the user behavior attributive character extracted.Correspondingly, for page properties-sex, for each under the page properties
Page properties feature (sex character), the score for the page properties feature that can be stored according to corresponding to each page of acquisition,
Calculate the average value of the score of the page properties feature;Average value highest page properties feature is chosen as user in the page
Page feature under attribute, and as the user behavior attributive character extracted.
Further, for page properties-interest, stored corresponding to each page that acquisition user accesses in the page
After each page properties feature and its score under attribute-interest, for each page properties feature under page properties-interest,
The score of the page properties feature of the corresponding storage of each page is added up, if the cumulative score of the page properties feature is more than
The page properties score threshold of setting, then the page using the page properties feature as the user under page properties-interest is special
Sign, and as the user behavior attributive character extracted.Wherein, page properties score threshold can be specifically configured to 0.
In practical application, relative to the interest of user, the sex of user and age belong to the base attribute of user, therefore,
In the embodiment of the present invention, the user behavior attributive character of user can be divided into:The base attribute feature of Behavior-based control, it is based on
The interest characteristics of behavior.So, page feature of the user under page properties-age and the page under page properties-sex
Region feature particularly belongs to the base attribute feature of the Behavior-based control in the user behavior attributive character of user;User is in page category
Page feature under property-interest particularly belongs to the interest characteristics of Behavior-based control in the user behavior attributive character of user.
In practical application, if user is interested in a certain channel of the advance classification in website, its page under the channel
Access times will be relatively more;Therefore, more preferably, in the embodiment of the present invention, while step S303 is carried out, can also pass through
Following steps extract user behavior attributive character:
S304:The channel belonging to each page accessed according to user within the setting cycle, counts user on website
Each access channel, and using each access channel counted as the user behavior attributive character extracted.
Specifically, each page accessed for user within the setting cycle, determines the channel belonging to the page;According to
The channel belonging to each page determined, count user and set cycle interior all access channels on website, and will statistics
The access channel gone out is as one kind in the interest characteristics of Behavior-based control in the user behavior attributive character of user.Wherein, website
The subordinate channel of upper each page divides in advance.
Further, for each access channel counted, it can be set according to user in the cycle and belong to the access
The access times of each page of channel determine the feature score of the access channel.So, more preferably, can be from the behavior of user
Access channel of the feature score not less than the access channels feature score threshold of setting is rejected in attributive character, guarantee is based on user
The degree of accuracy for the category division result that behavior property feature is divided.
For example, accessing channel for each, user can be set in the cycle and belong to each page of the access channel
Feature score of the summation of access times as the channels feature.Or each access channel can also be directed to, user is set
Belonging to visitation frequency of the access times summation of each page of the access channel as the access channel in cycle;And by this frequently
Spy of the ratio of the visitation frequency summations of the visitation frequency of road feature and all access channels counted as the channels feature
Obtain point.
In practical application, except that can determine that user is setting each page accessed in the cycle from user behavior data
Face and its access times, it can also determine that user is setting each advertisement clicked in the cycle and its number of clicks.
Therefore, more preferably, in the embodiment of the present invention, while step S303 is carried out, can also take out as follows
Take out user behavior attributive character:
S305:The industry belonging to each advertisement clicked on according to user within the setting cycle on website, counts user
Each click industry on website, and using the click industry counted as the user behavior attributive character extracted.
Specifically, each advertisement clicked on for user within the setting cycle on website, is determined belonging to the advertisement
Industry;According to the industry belonging to each advertisement determined, count user and set cycle interior all click industries on website,
And using the click industry counted as one kind in the interest characteristics of Behavior-based control in the user behavior attributive character of user.Its
In, the industry on website belonging to each advertisement divides in advance.
Further, for each click industry counted, it can be set according to user in the cycle and belong to the click
The number of clicks of each advertisement of industry, determine the feature score of the click industry.So, more preferably, can be from the use of user
Click industry of the feature score not less than the click industrial characteristic score threshold of setting is rejected in the behavior property feature of family, ensures to use
The degree of accuracy for the category division result that family behavior property feature is divided.
For example, clicking on industry for each, user can be set in the cycle and belong to each advertisement of the click industry
Feature score of the number of clicks summation as the click industry;Or each click industry can also be directed to, user is set into week
Belonging to the click frequency of the number of clicks summation as the click industry of each advertisement of the click industry in phase, and this is being clicked on
Feature score of the ratio of the click frequency summations of the click frequency of industry and all click industries as the click industry.
How user social contact attributive character is extracted from user social contact data on what is referred in above-mentioned steps S102, such as
Shown in Fig. 4, it can specifically extract as follows:
S401:From the user social contact data of acquisition, user's every blog article within the setting cycle is determined.
Wherein, the blog article of user can specifically include:The blog article that user initiates, forwarded, commenting on, thumbing up.
S402:For the every blog article determined, the blog article descriptor of the blog article is extracted;It is and all rich by what is extracted
User social contact attributive character of the blog article descriptor of text as user.
In the embodiment of the present invention, the user social contact attributive character of user can be divided into:Base attribute based on social activity
Feature, the interest characteristics based on social activity.Wherein, the blog article descriptor extracted particularly belongs to the user social contact attributive character of user
In based on social activity interest characteristics in one kind.On how to extract the blog article descriptor in blog article, this area skill can be used
The common technology means of art personnel, will not be described in detail herein.
Further, can also be according to the blog article descriptor in all blog articles for each blog article descriptor extracted
In occurrence number, determine the feature score of the blog article descriptor., can be with for each blog article descriptor in practical application
Feature score by the occurrence number of the blog article descriptor in all blog articles counted directly as the blog article descriptor.Or
Person, can also be by the occurrence number of the blog article descriptor in all blog articles and appearance of each blog article descriptor in all blog articles
Feature score of the ratio of number summation as the blog article descriptor.So, more preferably, can be from the user social contact attribute of user
Blog article descriptor of the feature score not less than the blog article descriptor feature score threshold value of setting is rejected in feature, guarantee is based on user
The degree of accuracy for the category division result that social attribute feature is divided.
In practical application, social platform of the user in website needs to register before logging in, and the log-on message of user leads to
It can often include:Age of user, sex, educational background, occupational information etc..
Therefore, more preferably, in the embodiment of the present invention, when carrying out step S401, S402, can also take out as follows
Take family social attribute feature:
S403:The log-on message of user is determined from the user social contact data of acquisition;And from the log-on message of user
Extract the user social contact attributive character of user.
Specifically, can be direct by information such as the age of the user extracted from the log-on message of user, sex, educational backgrounds
As the base attribute feature based on social activity in the user social contact attributive character of user.
Further, user can select customized label according to the situation of oneself in social platform.In view of making by oneself
Adopted label can reflect the hobby of user to a certain extent.Therefore, further, in the embodiment of the present invention, carrying out
Step S401, while S402, user social contact attributive character can also be extracted as follows:
S404:The customized label of user is determined from the user social contact data of acquisition, and it is self-defined by what is determined
Label is as the user social contact attributive character extracted.
Specifically, can be using the customized label of the user determined from the user social contact data of acquisition as user's
Interest characteristics based on social activity in user social contact attributive character.
In practical application, user would generally pay close attention to other users in social platform, and user it is of interest object it is usual
The interest of user can be reflected to a certain extent.Therefore, more preferably, further, in the embodiment of the present invention, walked
While rapid S401, S402, the user social contact attributive character of user can also be extracted according to the concern information of user, specifically
User social contact attributive character can be extracted as follows:
Social daily record of the user within the setting cycle can specifically include:The blog article of user, the log-on message of user, user
Concern information, LBS (Location Based Service, based on location-based service) positional information of user etc..
S405:Determine the concern information of user from the user social contact data of acquisition, and from the concern information determined
In extract the user certification user for belonging to specific industry of interest;For each certification user extracted, by the certification
Industry belonging to user is as the user social contact attributive character extracted.
Specifically, the affiliated industry for each certification user that can be paid close attention to user is special as the user social contact attribute of the user
Interest characteristics based on social activity in sign.
For example, if user has paid close attention to the certification that@appoints will strong ,@Sinas house property ,@Beijing self-housing etc. to belong to real estate industry
User, then can be using real estate industry as the user social contact attributive character extracted.
Further,, can basis after determining the industry belonging to certification user for each certification user extracted
The affiliated industry of each certification user, counts user's industry of interest;The industry paid close attention to for each user counted,
The quantity of the certification user of the sector can also be belonged in concern information according to user, calculate the feature score of the sector.
So, more preferably, feature score can be rejected from the user social contact attributive character of user to obtain not less than the industrial characteristic of setting
Divide the industry of threshold value, ensure the degree of accuracy of category division result divided based on user social contact attributive character.
In practical application, the feature of some geographic areas is obvious (characteristic area), for example, National Stadium, tourism
Sight spot region, commerce and trade area etc.;The place if user often comes in and goes out, the interest of user can be reflected to a certain extent.Therefore,
Region division can be carried out in advance, these are had into the region of obvious characteristic as characteristic area, and corresponding this feature region is deposited
The associated interest of storage.
The present inventor it is considered that the user of website be often based upon certain factor (such as:Work, interest etc.) often
Social platform is logged in a certain position;Therefore, it may be considered that according to the LBS positions of user and the characteristic area divided in advance, take out
Take out user social contact attributive character.
Specifically, therefore, can be from social day because the LBS service of social platform can record the login position of user
The LBS positions in user's setting cycle to be positioned are directly obtained in will.And then for each LBS positions of acquisition, according to this
Characteristic area belonging to LBS positions, using the interest associated by this feature region as base in the user social contact attributive character of user
In the interest characteristics of social activity.
In the embodiment of the present invention, the category division model on being referred in step S103 is training in advance, such as Fig. 5 institutes
Show, can specifically be trained as follows:
S501:For each training user selected from the user of website, the user behavior of the training user is counted
After attributive character and user social contact attributive character, the user characteristics of the training user is obtained by convergence.
S502:For each training user, determine that the training is used according to the user behavior attributive character of the training user
Behavior group belonging to family.
S503:For each training user, determine that the training is used according to the user social contact attributive character of the training user
Social group belonging to family.
S504:, will be common between the behavior group belonging to the training user and social group for each training user
Affiliated group of the group as the training user.
S505:Using the affiliated group of each training user, user characteristics as training data, using training data with pre-
The multi-tag multi-classification algorithm put carries out model training, obtains category division model.
Wherein, preset multi-tag multi-classification algorithm is specifically as follows M3l algorithms or Multiboost algorithms.
In the embodiment of the present invention, step S501-S504 specifically may be referred to the above-mentioned class that user is carried out based on Social behaviors
Step S101-S103 in other division methods, will not be described in detail herein.
More preferably, in order to ensure to obtain the higher category division model of accuracy, the category division result of user is improved
The degree of accuracy, it can also enter one when carrying out model training with preset multi-tag multi-classification algorithm using training data
Step carries out model training with preset multi-tag multi-classification algorithm using test data and obtains category division model.Practical application
In, it can be assessed using division effect (such as accuracy and recall rate) of the test data to category division model, with this
Obtain the optimal category division model of effect.Wherein, test data can specifically obtain in the following way:
For each test user selected from the user of website, the user behavior attribute for counting test user is special
After user social contact attributive character of seeking peace, the user characteristics of test user is obtained by convergence;And according to the use of test user
Family behavior property feature determines the behavior group belonging to test user, according to the user social contact attributive character of test user
The social group belonging to test user is determined, and will be common between the behavior group belonging to test user and social group
Affiliated group of the group as test user;Using the affiliated group of each test user, user characteristics as test data.
In practical application, in order to improve model training efficiency, the training data that can be directed to all training users is carried out in advance
Processing, can specifically include:Coding, normalization, dimensionality reduction etc. operate.
According to the above-mentioned category classification method that user is carried out based on Social behaviors, the embodiment of the present invention additionally provides a kind of base
The category division system of user is carried out in Social behaviors, as shown in fig. 6, can specifically include:User data acquisition module 601,
Attributive character abstraction module 602 and class of subscriber division module 603.
Wherein, user data acquisition module 601 be used to obtain the user behavior data of user to be sorted in the setting cycle,
And user social contact data and export.
Specifically, user data acquisition module 601 can according to the UID of user to be sorted, obtain to should UID record
The setting cycle in social daily record, and as the user social contact data of the user.It is possible to further right according to UID institutes
The cookie ID answered, obtain to should be in the setting cycle of cookie ID records access log, and as the use of the user
Family behavioral data.
Extracted in the user behavior data that attributive character abstraction module 602 is used to export from user data acquisition module 601
User behavior attributive character, it is special that user social contact attribute is extracted from the user social contact data of the output of user data acquisition module 601
Sign.
Class of subscriber division module 603 is used for the user behavior attributive character extracted according to attributive character abstraction module 602
With user social contact attributive character, user is divided into respective classes.
More preferably, in the embodiment of the present invention, the category division system that user is carried out based on Social behaviors is still further comprised:
Partitioning model training module 604.
Partitioning model training module 604 is used for each training user for being selected from the user of website, and statistics should
After the user behavior attributive character and user social contact attributive character of training user, the user that the training user is obtained by convergence is special
Sign;And the behavior group according to belonging to the user behavior attributive character of the training user determines the training user, according to the instruction
The user social contact attributive character of white silk user determines the social group belonging to the training user, and the row belonging to by the training user
Affiliated group of the common group as the training user between group and social group;By the affiliated race of each training user
Group, user characteristics carry out model training with preset multi-tag multi-classification algorithm using training data, obtained as training data
To category division model.
In the embodiment of the present invention, class of subscriber division module 603 can specifically include:First class of subscriber division unit or
Second user category division unit.
Wherein, the first class of subscriber division unit is used for the user behavior category extracted according to attributive character abstraction module 602
Property feature, user social contact attributive character, the behavioural characteristic rule base and social characteristics rule base of built in advance, user is divided into accordingly
Classification.
Second user category division unit is used for the user behavior attribute for the user for extracting attributive character abstraction module 602
Feature and user social contact attributive character, the user characteristics of user is obtained by convergence;It is and the user characteristics of obtained user is defeated
Enter into the category division model for category division;User is divided into by respective classes by category division model.
Further, as shown in fig. 7, the first class of subscriber division unit can specifically include:Behavioural characteristic matching is single
Member 701, social characteristics coupling subelement 702 and category division subelement 703.
Behavioural characteristic coupling subelement 701 is used for the user behavior attributive character for extracting attributive character abstraction module 602,
Matched with the user behavior attributive character recorded in the behavioural characteristic rule base of built in advance, according to being judged matching result
Behavior group belonging to user.Wherein, every kind of behavior group is directed in behavioural characteristic rule base, behavior group is have recorded and is had
Some user behavior attributive character.
Social characteristics coupling subelement 702 is used for the user social contact attributive character for extracting attributive character abstraction module 602,
Matched with the user social contact attributive character recorded in the social characteristics rule base of built in advance, according to being judged matching result
Social group belonging to user.Wherein, every kind of social group is directed in social characteristics rule base, the social group is have recorded and is had
Some user social contact attributive character.
Category division subelement 703 is used for the behavior race belonging to the user that reception behavior features characteristic matching subelement 701 exports
The social group belonging to user that group and social characteristics coupling subelement 702 export;Find out the behavior group belonging to user
The common group between social group, and user is divided into the common group found out.
In the embodiment of the present invention, based on above-mentioned advertisement placement method and class of subscriber dividing system, the embodiment of the present invention is also
A kind of advertisement delivery system is provided, as shown in figure 8, can specifically include:The category division of user is carried out based on Social behaviors
User data acquisition module 601, attributive character abstraction module 602 in system, class of subscriber division module 603, web-page requests
Receiving module 801, advertising inquiry module 802 and request processing module 803.
Wherein, web-page requests receiving module 801 is used for after receiving the access request of user, obtain the cookieID of user with
And corresponding UID, and send to user data acquisition module 601.User data acquisition module 601 can be according to user's
CookieID and corresponding UID, obtain the user behavior data and user social contact data and defeated of user in the setting cycle
Go out.
Advertising inquiry module 802 be used for the category division result of user that exports class of subscriber division module 603 as
Search key, find out the ad content to match with search key.
Request processing module 803 is used to insert advertising inquiry mould in the advertisement played column for the webpage that access request is asked
After the ad content that block 802 is found out, the webpage that access request is asked is returned to user.
In the embodiment of the present invention, each module, unit in the category division system based on Social behaviors progress user, son are single
The concrete function realization of member, the step S101- being referred in the above-mentioned category classification method that user is carried out based on Social behaviors
103rd, step S301-S305 and step S401-S405 detailed process;The specific work(of each module in advertisement delivery system
It can realize, the step S201-204 being referred in above-mentioned advertisement placement method, will not be described in detail herein.
In technical scheme, according to user behavior data of the user on website, the user under social platform
Social data, user behavior attributive character and user social contact attributive character are extracted respectively;Then, according to the use extracted
Family behavior property feature and user social contact attributive character carry out the category division of user.Compared to existing page is accessed according only to user
The information in face carrys out the anti-interest for pushing away user, technical scheme provided by the invention, considers user behavior attributive character and user
Social attribute feature carries out class of subscriber division, can increase the feature rich degree of category division, and improves class of subscriber and draw
Divide the degree of accuracy, also can improves crowd's positional accuracy;Based on higher crowd's positional accuracy, advertisement, a surname can be improved
Pass the specific aim of the audient of video etc., launch validity.
Described above is only the preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
For member, under the premise without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications also should
It is considered as protection scope of the present invention.
Claims (5)
- A kind of 1. category classification method that user is carried out based on Social behaviors, it is characterised in that including:Obtain the user behavior data of user to be sorted in the setting cycle, and user social contact data;User behavior attributive character is extracted from the user behavior data, user social contact is extracted from the user social contact data Attributive character;According to the user behavior attributive character and user social contact attributive character of extraction, the user is divided into respective classes;Institute The user behavior attributive character and user social contact attributive character according to extraction are stated, the user is divided into respective classes, specifically Including:The user behavior attribute recorded in the user behavior attributive character of the user, with the behavioural characteristic rule base of built in advance is special Sign is matched, the behavior group according to belonging to matching result judges the user;Wherein, in the behavioural characteristic rule base For every kind of behavior group, user behavior attributive character possessed by behavior group have recorded;The user social contact attribute recorded in the user social contact attributive character of the user, with the social characteristics rule base of built in advance is special Sign is matched, the social group according to belonging to matching result judges the user;Wherein, in the social characteristics rule base For every kind of social group, user social contact attributive character possessed by the social group have recorded;Group common between the behavior group belonging to the user and social group is found out, and the user is divided into and looked into In the common group found out;Or the user behavior attributive character and user social contact attributive character according to extraction, by described in User is divided into respective classes, specifically includes:The user is obtained by convergence in the user behavior attributive character of the user of extraction and user social contact attributive character User characteristics;The user characteristics of the obtained user is input in the category division model for category division;Pass through the classification The user is divided into respective classes by partitioning model;Wherein, the category division model is training in advance:For each training user selected from the user of website, count the training user user behavior attributive character and After user social contact attributive character, the user characteristics of the training user is obtained by convergence;And according to user's row of the training user The behavior group belonging to the training user is determined for attributive character, is determined according to the user social contact attributive character of the training user Go out the social group belonging to the training user, and by race common between the behavior group belonging to the training user and social group Affiliated group of the group as the training user;Using the affiliated group of each training user, user characteristics as training data, using the training data with preset Multi-tag multi-classification algorithm carries out model training, obtains category division model.
- 2. the method as described in claim 1, it is characterised in that it is described obtain category division model before, in addition to:For each test user selected from the user of website, count test user user behavior attributive character and After user social contact attributive character, the user characteristics of test user is obtained by convergence;And according to user's row of test user The behavior group belonging to test user is determined for attributive character, is determined according to the user social contact attributive character of test user Go out the social group belonging to test user, and by race common between the behavior group belonging to test user and social group Affiliated group of the group as test user;Using the affiliated group of each test user, user characteristics as test data;AndWhen carrying out model training with preset multi-tag multi-classification algorithm using the training data, also using the test The preset multi-tag multi-classification algorithm of data application carries out model training, obtains the category division model.
- A kind of 3. advertisement placement method, it is characterised in that including:After website receives the access request of user, the identification number cookieID of the user and corresponding use are obtained Family proof of identification UID;According to the cookieID and UID of the user, using the method as described in claim 1-2 is any, the user is obtained Category division result;Using obtained category division result as search key, find out in the advertisement to match with the search key Hold;The ad content found out is played in the advertisement played column of the webpage returned for the access request to the user.
- A kind of 4. category division system that user is carried out based on Social behaviors, it is characterised in that including:User data acquisition module, for obtaining user behavior data and the user society of user to be sorted in the setting cycle Intersection number evidence simultaneously exports;Attributive character abstraction module, for extracting user's row in the user behavior data that is exported from the user data acquisition module For attributive character, user social contact attributive character is extracted from the user social contact data of user data acquisition module output;Class of subscriber division module, for the user behavior attributive character extracted according to the attributive character abstraction module and user Social attribute feature, the user is divided into respective classes;The class of subscriber division module specifically includes:First user class Other division unit or second user category division unit;Partitioning model training module, for each training user for being selected from the user of website, count training use After the user behavior attributive character and user social contact attributive character at family, the user characteristics of the training user is obtained by convergence;And Behavior group according to belonging to the user behavior attributive character of the training user determines the training user, according to the training user User social contact attributive character determine social group belonging to the training user, and by the behavior group belonging to the training user Affiliated group of the common group as the training user between social group;By the affiliated group of each training user, use Family feature carries out model training with preset multi-tag multi-classification algorithm, obtains category division model as training data;Wherein, the first class of subscriber division unit specifically includes:Behavioural characteristic coupling subelement, it is and pre- for the user behavior attributive character for extracting the attributive character abstraction module The user behavior attributive character recorded in the behavioural characteristic rule base built is matched, and judges the user according to matching result Affiliated behavior group;Wherein, every kind of behavior group is directed in the behavioural characteristic rule base, behavior group is have recorded and is had Some user behavior attributive character;Social characteristics coupling subelement, it is and pre- for the user social contact attributive character for extracting the attributive character abstraction module The user social contact attributive character recorded in the social characteristics rule base built is matched, and judges the user according to matching result Affiliated social group;Wherein, every kind of social group is directed in the social characteristics rule base, the social group is have recorded and is had Some user social contact attributive character;Category division subelement, for receive behavior group belonging to the user of behavioural characteristic matching unit output and Social group belonging to the user of the social characteristics matching unit output;Find out behavior group and the society belonging to the user Common group between friendship group, and the user is divided into the common group found out;AndThe second user category division unit is used for the user's row for the user for extracting the attributive character abstraction module For attributive character and user social contact attributive character, the user characteristics of the user is obtained by convergence;And the use that will be obtained The user characteristics at family is input in the category division model for category division;By the category division model by the user It is divided into respective classes.
- A kind of 5. advertisement delivery system, it is characterised in that including:User data acquisition module, attributive character abstraction module, class of subscriber division module as claimed in claim 4;Web-page requests receiving module, after the access request for receiving user, obtain the cookieID of user and corresponding UID, and send to the user data acquisition module;Advertising inquiry module, the category division result of the user for the class of subscriber division module to be exported are closed as search Keyword, find out the ad content to match with the search key;Request processing module, the advertising inquiry is inserted in the advertisement played column for the webpage asked in the access request After the ad content that module is found out, the webpage is returned to the user;The user data acquisition module is specifically used for cookieID and corresponding UID according to user, obtains the setting cycle The user behavior data and user social contact data of interior user simultaneously export.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410492126.5A CN104298719B (en) | 2014-09-23 | 2014-09-23 | Category division, advertisement placement method and the system of user is carried out based on Social behaviors |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410492126.5A CN104298719B (en) | 2014-09-23 | 2014-09-23 | Category division, advertisement placement method and the system of user is carried out based on Social behaviors |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104298719A CN104298719A (en) | 2015-01-21 |
CN104298719B true CN104298719B (en) | 2018-02-27 |
Family
ID=52318444
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410492126.5A Active CN104298719B (en) | 2014-09-23 | 2014-09-23 | Category division, advertisement placement method and the system of user is carried out based on Social behaviors |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104298719B (en) |
Families Citing this family (31)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104683122A (en) * | 2015-02-12 | 2015-06-03 | 北京集奥聚合科技有限公司 | Multi-screen-linkage-based information transmitting method and system |
CN104750832A (en) * | 2015-04-02 | 2015-07-01 | 百度在线网络技术(北京)有限公司 | Information releasing method, device and system |
CN106204083B (en) * | 2015-04-30 | 2020-02-18 | 中国移动通信集团山东有限公司 | Target user classification method, device and system |
CN104992342B (en) * | 2015-05-11 | 2019-08-13 | 腾讯科技(北京)有限公司 | Promotion message delivery effectiveness determines method, monitoring server and terminal |
CN106294508B (en) * | 2015-06-10 | 2020-02-11 | 深圳市腾讯计算机系统有限公司 | Brushing amount tool detection method and device |
CN104951544A (en) * | 2015-06-19 | 2015-09-30 | 百度在线网络技术(北京)有限公司 | User data processing method and system and method and system for providing user data |
CN106487825B (en) * | 2015-08-25 | 2019-12-24 | 阿里巴巴集团控股有限公司 | Information association method and device |
CN105447376B (en) * | 2015-10-30 | 2018-08-03 | 广州市汇助惠电子商务有限公司 | A kind of Subscriber Management System |
CN106817384A (en) * | 2015-11-27 | 2017-06-09 | 亿阳信通股份有限公司 | A kind of analysis method and system that behavior is accessed based on user's telecommunications |
CN105516928A (en) * | 2016-01-15 | 2016-04-20 | 中国联合网络通信有限公司广东省分公司 | Position recommending method and system based on position crowd characteristics |
CN107205011A (en) * | 2016-03-17 | 2017-09-26 | 阿里巴巴集团控股有限公司 | Business datum method for pushing and its system |
CN106780064B (en) * | 2016-12-02 | 2021-01-05 | 腾讯科技(深圳)有限公司 | Region division method, device and network equipment |
CN106600323A (en) * | 2016-12-13 | 2017-04-26 | 合肥华耀广告传媒有限公司 | Advertisement sharing system based on social network |
CN108280458B (en) * | 2017-01-05 | 2022-01-14 | 腾讯科技(深圳)有限公司 | Group relation type identification method and device |
CN107330001A (en) * | 2017-06-09 | 2017-11-07 | 国政通科技股份有限公司 | The creation method and system of a kind of diversification label |
CN107392648A (en) * | 2017-06-29 | 2017-11-24 | 上海精数信息科技有限公司 | Assess the method and device of directory audient distribution |
CN107451247B (en) * | 2017-07-28 | 2021-03-30 | 北京小米移动软件有限公司 | User identification method and device |
TWI670662B (en) * | 2017-11-09 | 2019-09-01 | 財團法人資訊工業策進會 | Inference system for data relation, method and system for generating marketing targets |
CN108182255B (en) * | 2017-12-29 | 2020-07-28 | 重庆金融资产交易所有限责任公司 | Title item information recommendation method and device, storage medium and computer equipment |
CN110197386B (en) * | 2018-04-12 | 2022-02-08 | 腾讯科技(深圳)有限公司 | Media resource pushing method and device, storage medium and electronic device |
CN109033118B (en) * | 2018-05-23 | 2021-06-29 | 国政通科技股份有限公司 | Dynamic data judgment method and device based on object |
CN109345075A (en) * | 2018-08-31 | 2019-02-15 | 深圳市轱辘汽车维修技术有限公司 | A kind of professional person, which authenticates, investigates method, apparatus and terminal device |
CN110917629A (en) * | 2018-09-20 | 2020-03-27 | 北京默契破冰科技有限公司 | Method, apparatus, and computer storage medium for managing user behavior |
CN110917628A (en) * | 2018-09-20 | 2020-03-27 | 北京默契破冰科技有限公司 | Method, apparatus, and computer storage medium for determining user grouping |
CN109409943A (en) * | 2018-10-11 | 2019-03-01 | 温州你创我帮网络科技有限公司 | A kind of advertisement promotion system Internet-based |
CN111353001B (en) * | 2018-12-24 | 2023-08-18 | 杭州海康威视数字技术股份有限公司 | Method and device for classifying users |
CN110059249B (en) | 2019-04-03 | 2022-11-25 | 华为技术有限公司 | Personalized recommendation method, terminal device and system |
TWI757854B (en) * | 2020-08-28 | 2022-03-11 | 中國信託商業銀行股份有限公司 | Business recommendation system and method |
CN112435067A (en) * | 2020-11-30 | 2021-03-02 | 翼果(深圳)科技有限公司 | Intelligent advertisement putting method and system for cross-e-commerce platform and social platform |
CN113630336A (en) * | 2021-07-19 | 2021-11-09 | 上海德衡数据科技有限公司 | Data distribution method and system based on optical interconnection |
CN113672777B (en) * | 2021-08-30 | 2023-09-08 | 上海飞旗网络技术股份有限公司 | User intention exploration method and system based on flow correlation analysis |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101685521A (en) * | 2008-09-23 | 2010-03-31 | 北京搜狗科技发展有限公司 | Method for showing advertisements in webpage and system |
US8170971B1 (en) * | 2011-09-28 | 2012-05-01 | Ava, Inc. | Systems and methods for providing recommendations based on collaborative and/or content-based nodal interrelationships |
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 |
CN103366020A (en) * | 2013-08-06 | 2013-10-23 | 刘临 | System and method for analyzing user behaviors |
CN103458042A (en) * | 2013-09-10 | 2013-12-18 | 上海交通大学 | Microblog advertisement user detection method |
CN103778555A (en) * | 2014-01-21 | 2014-05-07 | 北京集奥聚合科技有限公司 | User attribute mining method and system based on user tags |
-
2014
- 2014-09-23 CN CN201410492126.5A patent/CN104298719B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101685521A (en) * | 2008-09-23 | 2010-03-31 | 北京搜狗科技发展有限公司 | Method for showing advertisements in webpage and system |
US8170971B1 (en) * | 2011-09-28 | 2012-05-01 | Ava, Inc. | Systems and methods for providing recommendations based on collaborative and/or content-based nodal interrelationships |
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 |
CN103366020A (en) * | 2013-08-06 | 2013-10-23 | 刘临 | System and method for analyzing user behaviors |
CN103458042A (en) * | 2013-09-10 | 2013-12-18 | 上海交通大学 | Microblog advertisement user detection method |
CN103778555A (en) * | 2014-01-21 | 2014-05-07 | 北京集奥聚合科技有限公司 | User attribute mining method and system based on user tags |
Also Published As
Publication number | Publication date |
---|---|
CN104298719A (en) | 2015-01-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104298719B (en) | Category division, advertisement placement method and the system of user is carried out based on Social behaviors | |
EP3779841B1 (en) | Method, apparatus and system for sending information, and computer-readable storage medium | |
CN108197330B (en) | Data digging method and device based on social platform | |
US20180144367A1 (en) | Method and system for creating user based summaries for content distribution | |
WO2017096832A1 (en) | Media data recommendation method and server | |
CN1917463B (en) | Method for associating user information based on user's operating characters | |
CN109155136A (en) | The computerized system and method for highlight are detected and rendered automatically from video | |
CN110035314A (en) | Methods of exhibiting and device, storage medium, the electronic device of information | |
US11070851B2 (en) | System and method for providing image-based video service | |
CN109409928A (en) | A kind of material recommended method, device, storage medium, terminal | |
CN101510856A (en) | Method and apparatus for extracting member relation loop in SNS network | |
JP7197930B2 (en) | Methods and systems for providing location-based personalized content | |
CN106204146A (en) | A kind of Internet advertising put-on method, device | |
CN108229994A (en) | A kind of information-pushing method and device | |
CN107004179A (en) | The system and method for dynamic select communication identifier | |
CN107590267A (en) | Information-pushing method and device, terminal and readable storage medium storing program for executing based on picture | |
CN103955480B (en) | A kind of method and apparatus for determining the target object information corresponding to user | |
CN102722832A (en) | Online video advertisement refinement targeting delivery method | |
CN105871585A (en) | Terminal association method and device | |
CN106202260A (en) | searching method, device and search engine | |
CN103412880B (en) | Method and device for determining implicit associated information between multimedia resources | |
CN107729438B (en) | User behavior data establishing and analyzing method | |
CN109688217A (en) | A kind of information push method, device and electronic equipment | |
CN110097388A (en) | A kind of web advertisement data analysing method | |
US20220309279A1 (en) | Computerized system and method for fine-grained event detection and content hosting therefrom |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right | ||
TR01 | Transfer of patent right |
Effective date of registration: 20230417 Address after: Room 501-502, 5/F, Sina Headquarters Scientific Research Building, Block N-1 and N-2, Zhongguancun Software Park, Dongbei Wangxi Road, Haidian District, Beijing, 100193 Patentee after: Sina Technology (China) Co.,Ltd. Address before: 100080, International Building, No. 58 West Fourth Ring Road, Haidian District, Beijing, 20 floor Patentee before: Sina.com Technology (China) Co.,Ltd. |