CN104008184A - Method and device for pushing information - Google Patents

Method and device for pushing information Download PDF

Info

Publication number
CN104008184A
CN104008184A CN201410256923.3A CN201410256923A CN104008184A CN 104008184 A CN104008184 A CN 104008184A CN 201410256923 A CN201410256923 A CN 201410256923A CN 104008184 A CN104008184 A CN 104008184A
Authority
CN
China
Prior art keywords
user
interest
category
crowd
context data
Prior art date
Application number
CN201410256923.3A
Other languages
Chinese (zh)
Inventor
刘阳
汪冠春
李静
林加新
向伟
徐倩
黄硕
Original Assignee
百度在线网络技术(北京)有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 百度在线网络技术(北京)有限公司 filed Critical 百度在线网络技术(北京)有限公司
Priority to CN201410256923.3A priority Critical patent/CN104008184A/en
Publication of CN104008184A publication Critical patent/CN104008184A/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation

Abstract

The embodiment of the invention provides a method and device for pushing information. The method includes the steps that user identification and currently-used scene data are obtained, wherein the scene data comprise at least one of time, the place, the network access means and the terminal type; according to a user interest model which is established in advance, a recommended interest type correlated to a user and the currently-used scene data is determined; according to the recommended interest type, information is pushed to the user. According to the method and device for pushing the information, the recommended interest type correlated to the user and the currently-used scene data is determined through the user interest model which is established in advance, the recommended interest type is pushed to the user, so that currently-required content is pushed to the user with the combination of the user and the currently-used scene data, and accuracy of the information pushed to the user is improved.

Description

The method for pushing of information and device

Technical field

The embodiment of the present invention relates to Information Filtering Technology and data mining technology, relates in particular to a kind of method for pushing and device of information.

Background technology

Along with the development of Internet technology, each terminal is presented to user's information sharp increase.The information of magnanimity makes user find and obtain fast, easily own needed information and becomes difficulty.Recommended technology, as the important means of information filtering, can automatically be found the interested information of user, effectively for user provides personalized recommendation service.

Existing personalized recommendation technology mainly comprises personalized recommendation technology based on user interest modeling and the collaborative filtering recommending technology based on content recommendation.

Personalized recommendation technology based on user interest modeling, is generally the category of interest according to user, recommends corresponding information to user, thereby reaches the object of personalized recommendation.

Collaborative filtering recommending technology based on content recommendation, is generally the user crowd of knowing under user, and according to similar user crowd, to targeted customer's recommendation information, thereby reaches the object of personalized recommendation.

All there is following defect in above-mentioned personalized recommendation technology: the information pushing for user is not the current required content of user, cause the accuracy rate of propelling movement low, and personalized push degree is low.

Summary of the invention

The embodiment of the present invention provides a kind of method for pushing and device of information, to improve the accuracy rate of the information pushing to user.

First aspect, the embodiment of the present invention provides a kind of method for pushing of information, comprising:

Obtain user ID and current use context data, wherein, described context data comprises following at least one: time, place, networking means and terminal type;

According to the user interest model of setting up in advance, determine the recommendation category of interest being associated with described user and current use context data;

According to described recommendation category of interest to user's pushed information.

Second aspect, the embodiment of the present invention also provides a kind of pusher of information, comprising:

User ID and current use context data module, for obtaining user ID and current use context data, wherein, described context data comprises following at least one: time, place, networking means and terminal type;

Recommend category of interest determination module, for according to the user interest model of setting up in advance, determine the recommendation category of interest being associated with described user and current use context data;

Pushing module, for according to described recommendation category of interest to user's pushed information.

The method for pushing of the information that the embodiment of the present invention provides and device, by the user interest model of setting up in advance, determine the recommendation category of interest being associated with user and current use context data, and push to user, therefore can be in conjunction with user itself, and in conjunction with current use context data, push current required content to user, and improve the accuracy rate of the information pushing to user.

Brief description of the drawings

In order to be illustrated more clearly in the present invention, introduce simply the accompanying drawing of required use in the present invention being done to one below, apparently, accompanying drawing in the following describes is some embodiments of the present invention, for those of ordinary skill in the art, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

The process flow diagram of the method for pushing of a kind of information that Fig. 1 a provides for the embodiment of the present invention one;

The pushed information schematic diagram of Fig. 1 b for adopting the method for pushing of the information that provides of the embodiment of the present invention one to present;

Fig. 1 c another pushed information schematic diagram for adopting the method for pushing of the information that provides of the embodiment of the present invention one to present;

The process flow diagram of the method for pushing of a kind of information that Fig. 2 a provides for the embodiment of the present invention two;

Fig. 2 b is the schematic diagram of the method for pushing example of information applicable in the embodiment of the present invention two;

The process flow diagram of the method for pushing of a kind of information that Fig. 3 provides for the embodiment of the present invention three;

The structural representation of the pusher of a kind of information that Fig. 4 provides for the embodiment of the present invention four.

Embodiment

For making the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing, the technical scheme in the embodiment of the present invention is described in further detail, obviously, described embodiment is the present invention's part embodiment, instead of whole embodiment.Be understandable that; specific embodiment described herein is only for explaining the present invention; but not limitation of the invention; based on the embodiment in the present invention; those of ordinary skill in the art, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.It also should be noted that, for convenience of description, in accompanying drawing, only show part related to the present invention but not full content.

Embodiment mono-

Refer to Fig. 1 a, the process flow diagram of the method for pushing of a kind of information providing for the embodiment of the present invention one.The method of the embodiment of the present invention can be carried out by the pusher of the information realizing with hardware and/or software, and this device is typically and is disposed in the server that Information Push Service can be provided.

As shown in Figure 1a, described method comprises:

110, obtain user ID and current use context data, wherein, described context data comprises following at least one: time, place, networking means and terminal type;

User ID and the current use context data with described user are specifically obtained in this operation.User ID is for unique identification user, described user ID can comprise: MAC (MediumAccess Control, medium access control) address, mobile terminal are (for example, smart mobile phone) SIM (Subscriber Identity Module, client identification module) card information, and/or user logins the user account of internet or other marks, the present embodiment does not limit this, as long as can unique identification user.

Wherein, the user account that user logins internet is specially user logins the user account of each service application, both can comprise that user passed through PC (Personal Computer, PC) user account of the each service application of end login, also can comprise that user logins the user account of each service application by mobile terminal, described mobile terminal comprises panel computer, personal digital assistant and smart mobile phone.Each service application comprises each Baidu service application, particularly, can comprise: Baidu's search, Baidu's navigation Service, Baidu's community service are (for example, Baidu library, Baidupedia, Baidu's mhkc and Baidu are known etc.), Baidu's Entertainment application, Baidu's Software tool (for example, Baidu's browser, the bodyguard of Baidu and Baidu's virus killing etc.) and the each mobile class application of Baidu.

Can know by the user profile relevant with user ID to described user's current use context data.User can be logined to the system time of terminal of each Baidu service application as current use context data, user can be logined to current location that the position transducer of the terminal of each Baidu service application obtains as current use context data.Can know networking means and terminal type by the user profile relevant to user ID.

Networking means refer to the mode of user Internet access, comprise access in radio or wireless access etc., and wherein access in radio can comprise: ordinary telephone line access, copper cable access, optical fiber access and the access of LAN (Local Area Network) netting twine; Wireless access mainly comprises: WiFi (Wireless Fidelity, Wireless Fidelity) network access, CDMA (Code Division Multiple Access, CDMA) network access, GPRS (General Packet Radio Service, general packet radio service technology) network access, EDGE (Enhanced Data Rate for GSM Evolution, enhanced data rates for gsm evolution technology) network access and 3G (3rd-Generation, 3G (Third Generation) Moblie technology) wireless Internet access mode etc.Terminal type comprises: PC end and mobile terminal, described mobile terminal comprises panel computer, personal digital assistant and smart mobile phone.

For example, the current use context data that the user ID of obtaining is xiaoS comprises: the time is 18:56, and place is that coffee-house, network access mode are that WiFi and terminal type are smart mobile phone.

Described context data can also comprise: the mode of operation of terminal, the information about power of for example quiet or meeting, and/or terminal etc.

120,, according to the user interest model of setting up in advance, determine the recommendation category of interest being associated with described user and current use context data;

The user interest model of setting up in advance in this operation can be for including the statistic of classification result of the incidence relation between user crowd, use context data and category of interest, therefore this operation, specifically according to described statistic of classification result, is determined the recommendation category of interest being associated with described user and current use context data.

130, according to described recommendation category of interest to user's pushed information.

The technical scheme of the embodiment of the present invention, by the user interest model of setting up in advance, determine the recommendation category of interest being associated with user and current use context data, and push to user, therefore can be in conjunction with user itself, and in conjunction with current use context data, push current required content to user, and improve the accuracy rate of the information pushing to user.

On the basis of above-described embodiment, after operation 130, can also comprise:

140, obtain the click feedback data of user to category of interest, the use context data during according to user's clicking operation and click feedback data correction user interest model.

It should be noted that, user interest model is set up in advance, what represent is the user behavior tendency within the scope of the history samples of setting, for example, the user interest model of setting up in advance according to the user's history log in each Baidu service application on April 30,1 day to 2014 January in 2014, click feedback data when user's clicking operation has reflected the up-to-date behavior disposition of active user in the situation of presence.Click the numerical value of feedback data larger, represent interested in recommendation category of interest corresponding to this click feedback data in the situation of presence of active user.Therefore, can improve the degree of accuracy of user interest model by clicking feedback data correction user interest model, can push current required content to user according to revised user model, thereby further improve the accuracy rate of the information pushing to user.

Refer to Fig. 1 b and Fig. 1 c.Fig. 1 b is the category of interest that 9:00 pushed to active user A on May 1st, 2014 of determining and showing at mobile terminal according to operation 130, wherein news, music, purchase by group with the sortord of NBA and be news, music, purchase by group with the recommendation probability of NBA and successively decrease successively, Fig. 1 c is that active user A is the click feedback data of news and music at 9:00 to recommending category of interest, show, current time, user A is to recommending the classification that category of interest is music interested.Therefore, the user interest model of setting up in advance by the correction of described click feedback data, according to revised user interest model, time after 1 day May in 2014, for example on May 3rd, 2014 9:00, the order of the category of interest pushing to active user A showing at mobile terminal can be music, news, purchase by group and NBA, thereby has improved the accuracy rate of the information pushing to user A.

The method for pushing of the information that the embodiment of the present invention provides, can be applied to several scenes.For example, in the personalized homepage of Baidu, all kinds of contents can be opened with user different change the in time, place of homepage, realization such as the work hours do not push long video or novel, and push some shorter news category contents, novel or the TV play class content of coming off duty and just recommending some user habits to see in this time period while being in.And for example, subscribe on page at the individualized content of the App of mobile phone terminal Baidu, in different time places, subscribe to content ordering can change thereupon, realize such as mealtime outside, by push, near making, food and drink cuisines are discharged to more awake destination locations, when on and off duty by bus, push more newsflash, or amusement class content.

Embodiment bis-

Refer to Fig. 2 a, the process flow diagram of the method for pushing of a kind of information providing for the embodiment of the present invention two.The present embodiment, on the basis of above-described embodiment, provides the preferred version of setting up in advance user interest model, before obtaining the operation of user ID and current use context data, can comprise:

210, obtain user's history log, wherein, described user's history log at least comprises: user ID, the user's historical behavior data that are associated with user ID and use context data;

220, by user's history log according to category of interest, user crowd with use context data, carry out statistic of classification, obtain described user interest model.

Wherein, user's history log refers to that user logins after each service application at PC end or mobile terminal, the set of the information (for example, click, search, interpolation and/or deletion etc.) that the user that terminal or server record get off operates by service application interface and the context information while carrying out corresponding operating.

It should be noted that, be to be mutually related between the user ID in described user's history log, the user's historical behavior data that are associated with user ID and use context data.

In other words, what the user's historical behavior data that are associated with user ID and use context data reflected is that user passes through service application interface in service application, and the information of a series of concrete operations of carrying out under concrete use situation; What the use context data being associated with user ID and user's historical behavior data reflected is that user passes through service application interface, the context information while carrying out a series of concrete operations in service application; And the user ID reflection being associated with user's historical behavior data and use context data is to pass through service application interface the corresponding user's of a series of concrete operations who carries out under concrete use situation information in service application.

Illustrate by an example.The user ID of obtaining is that certain customers' history log of wei is as shown in table 1, and wherein the rise time of user's history log is on March 20th, 2014.

Table 1

Described in table 1, use the classification of context data to comprise: go out, company and be in three kinds, can be by the user's who obtains particular geographic location classification being obtained according to Preset Time granularity, described Preset Time granularity can comprise: 1 month, 5 months or 3 weeks etc.Wherein, company and the corresponding particular geographic location of being in are more fixing, difference is that company is generally 8:00-18:00 the corresponding time, and the corresponding time of being in is generally before 7:00 and 19:00 after, and the corresponding particular geographic location of going out variation is larger relatively.

By table 1 show this user's user ID, user's historical behavior data of being associated with user ID and use the interrelated relation between context data.For example, this user is at 8:30 with while going out, and this user is by the GPRS wireless mode online of smart mobile phone, and clicks Baidu's news.

It should be noted that, certain customers' history log of table 1 is certain customers' historical behavior data of user some day and corresponding use context data, similarly, in Preset Time granularity (for example can obtain a large number of users, half a year, 4 months, or 2 months etc.) whole user's historical behavior data and corresponding use context data, realize and obtain user's history log.

Due to user's behavioural habits, can obtain user by a large amount of user's history logs and carry out specific behavior specific use under situation, obtain the user interest preference relevant to given use sight, in addition, according to the behavioral similarity between different user, and user similar behavior is divided into specific user crowd, can obtain the user behavior data associated with specific user crowd and use context data, in other words, by user's history log according to category of interest, user crowd and use context data, carry out statistic of classification, obtain described user interest model.Accordingly, according to targeted customer and this user's current use context data, utilize described user interest model, can determine the recommendation category of interest being associated with this user and current use context data.

Also it should be noted that, Preset Time granularity is with the difference that uses the time in context data, choosing of Preset Time granularity affects the precision of user interest model and sets up complexity, and time in use context data only represents a dimension that uses context data, representative of consumer is in the behavior of specific time period or time point.Particularly, Preset Time granularity is larger, and the precision of user interest model is higher, user interest model to set up complexity larger; Preset Time granularity is less, and the precision of user interest model is lower, user interest model to set up complexity lower.

Use the choosing of dimension of time in context data, place, networking means and terminal type, on the impact of user interest model precision and complexity, similar on the impact of user interest model precision and complexity with Preset Time granularity, repeat no more herein.

The technical scheme of the present embodiment, by by the user's historical behavior data that include user ID, be associated with user ID with use user's history log of context data, carry out statistic of classification according to category of interest, user crowd and use context data, can obtain user interest model.

On the basis of this enforcement, after setting up in advance user interest model and obtaining user ID and current use context data, carry out the user interest model of setting up according in advance, determine the operation of the recommendation category of interest being associated with described user and current use context data.

According to the user interest model of setting up in advance, determine the preferred version of this operation of recommendation category of interest being associated with described user and current use context data below by instantiation explanation.For clarity sake, the user interest model that below, paper obtains by operation 210 and operation 220.

Refer to Fig. 2 b, be depicted as the statistic of classification that category of interest is novel.Horizontal ordinate representative is used the time in context data, unit is hour, the number of this category of interest is paid close attention in ordinate representative, and user crowd is divided into " under-18s ", " 18-24 year ", " 24-36 year ", " 36-50 year " and " more than 50 years old " five intervals, wherein Fig. 2 b only shows " under-18s ", " 18-24 year " and " 24-36 year " three different user crowds pay close attention to the statistic of classification result that category of interest is novel, and this statistic of classification result is included in described user interest model.

Statistic of classification result thus, can obtain the custom of different user crowd on this class behavior of novel, the students in middle and primary schools crowd that is generally of under-18s, and when paying close attention to novel, whole day keeps average, and daytime is relatively high; 18-24 year be mainly university student crowd, there are three obvious crests the time period of paying close attention to novel, corresponds respectively to 7-8 point, 11-13 point and 17-18 point; 24-36 year be mainly young working clan, 20 demands of paying close attention to novels later than other times Duan Yaogao.

Therefore, when active user is 21 years old, the time in current use context data is during for 17:05, according to described user interest model, and can be using novel as the recommendation category of interest being associated with described user and current use context data.

It should be noted that, above-mentioned example has only considered that category of interest is novel, and in use context data, only consider the time, and in user crowd, only considered the age dimension in " under-18s ", " 18-24 year ", " 24-36 year ", " 36-50 year " and " more than 50 years old ".Similarly, user crowd be can obtain including, (for example context data and other category of interest used, news, music, video etc.) between the statistic of classification result of incidence relation, based on this, can obtain user interest model, thereby utilize user interest model can determine the recommendation category of interest being associated with described user and current use context data.

The user interest model that described basis is set up in advance, determines the recommendation category of interest being associated with described user and current use context data, preferably includes:

Calculate the recommendation probability w of category of interest according to following formula:

W=P (crowd | user) × P (category of interest | crowd)

Wherein, P (crowd | user) belongs to a certain crowd's the ratio that enlivens number of days and this user and always enliven number of days for this user this user, for example, user is as 18 years old age bracket crowd's the number of days that enlivens, and the ratio between the total active number of days of this user, the crowd relatively static for age etc. divides foundation, and P (crowd user) is generally discrete 0 or 1; For example, crowd is by age divided into 18 years old crowd and 19 years old crowd, if user's age is 18 years old, so this user to belong to 18 years old crowd's probability P (18 years old crowd | user) be 1, the probability P (19 years old crowd | user) that this user belongs to 19 years old crowd is 0.P (category of interest | crowd) be in the crowd under this user, in setting a period of time, in the user number of this category of interest of concern and this crowd, pay close attention to the ratio of total user number of this category of interest.A period of time of this setting is for example 1 day, 1 week etc.P (category of interest | crowd) is not discrete 0 or 1 conventionally, but arbitrary value between 0 and 1.Taking the user crowd under the user of 18 years old as 18 years old crowd, and the category of interest paid close attention to of this 18 years old crowd comprises that music and game describe for example, the user number of this 18 years old crowd concern music in a week that sets is 600, this total user number of paying close attention to music for 18 years old in crowd is 1000, based on this, can obtain P (music | 18 years old crowd) is 60%; User number that this 18 years old crowd paid close attention to game in one week that sets is 500, and in this 18 years old crowd, paying close attention to total user number of playing is 2000, and based on this, can obtain P (play | 18 years old crowd) is 25%.

Also, in the time calculating the recommendation probability of category of interest, calculated respectively two kinds of probability: a kind of P of being (crowd | user), represent that active user belongs to corresponding user crowd's probability; Another kind is P (category of interest | crowd), represents the probability of the category of interest of the concern correspondence of current time in the user crowd under active user.The category of interest being calculated by above-mentioned formula recommends probability to reflect certain class crowd's recommendation probability, if user crowd is divided with two or more dimensions, can calculate respectively the recommendation probability under all kinds of crowds, and then tabulate statistics.

For example, in user interest model, age is the user crowd that 18-35 year, occupation are financial class, use sight for 8:00-9:00 and the use sight of going out under, tend to the wireless Internet access mode by mobile terminal, check the finance and economic news in Baidu's news and listen to the popular song in Baidu's music; In described user interest model, also comprise, age is 25-35 year, the user crowd that condition is pregnant woman, use sight for 8:00-10:00 and the use sight of going out under, tend to check the address, pregnant baby shop in Baidu's map, also tend to check pregnant baby's picture by mobile terminal.

Therefore, when active user is 25 years old, occupation is financial class, and condition is pregnant woman, and under the sight of going out, and when current time is 8:35, to belong to the age be 18-35 year, the occupation user crowd that is financial class and belong to the probability that the age is 25-35 year, the condition user crowd that is pregnant woman need to calculate this user; Also need to calculate specific user crowd and pay close attention to the probability of corresponding category of interest in current time, thereby obtain the recommendation probability of each category of interest, that is, check the recommendation probability w of the finance and economic news in Baidu's news 1, listen to the recommendation probability w of the popular song in Baidu's music 2, check the recommendation probability w of the address, pregnant baby shop in Baidu's map 3with the recommendation probability w that checks pregnant baby's picture 4.

By recommend probability the highest or recommend probability to be greater than category of interest predetermined threshold value or that recommend the higher predetermined number of probability, be defined as the recommendation category of interest of this user under this sight.

For example,, by relatively checking the recommendation probability w of the finance and economic news in Baidu's news 1, listen to the recommendation probability w of the popular song in Baidu's music 2, check the recommendation probability w of the address, pregnant baby shop in Baidu's map 3with the recommendation probability w that checks pregnant baby's picture 4, can will recommend the highest recommendation probability w that checks pregnant baby's picture of probability 4recommendation category of interest as active user under the situation of presence, pushes thereby carry out pregnant baby's picture.

This is preferred embodiment determined and is recommended category of interest by the recommendation probability of category of interest, has improved the accuracy rate of the information pushing to user.

Embodiment tri-

Refer to Fig. 3, the process flow diagram of the method for pushing of a kind of information providing for the embodiment of the present invention three.The present embodiment, on the basis of above-described embodiment, provides user's history log according to category of interest, user crowd and has used context data, carries out statistic of classification, obtains the preferred version of described this operation of user interest model.As shown in Figure 3, described method for optimizing comprises:

310, user's history log is classified by category of interest;

This operation specifically will be filtered with the part of special interests classification in user's history log.Various informative due to category of interest, both comprised type of theme (for example, music class, sport category, with news category etc.), also comprise hot part of speech type (for example, Liu Dehua, 36 kryptons, MH370 etc.), therefore this operation can have numerous embodiments, for example, comprises at least one in following embodiment:

Utilize subject classification device, described user's history log is carried out to the classification of category of interest; Or

Using station address feature as default category of interest, station address corresponding to user's historical behavior data in user's history log carried out to filtering classification; Or

Using keyword bag as default category of interest, utilize keyword bag matching technique, user's history log is classified by default category of interest.

Wherein, utilize subject classification device, using the type of theme of subject classification device as category of interest, thereby described user's history log can be carried out to the classification of category of interest.The subject categories of described subject classification device comprises existing common theme and category of employment.Particularly, can utilize search key (query) subject classification device, the subject categories providing by subject classification device by search key is classified, thereby realize, the user's history log obtaining is in advance classified; Also can utilize log subject classification device, utilize the type of user's history log that described user's history log is carried out to the classification of category of interest, for example, user's history log of the forms such as mp3, mp4, WMV and FLV is divided into music class.

Wherein, the website interest of station address feature representative of consumer, can pass through particularly URL (Uniform Resource Locator, uniform resource locator) adress analysis and obtain.

Wherein, keyword handbag, containing hot part of speech type, using keyword bag as default category of interest, utilizes keyword bag matching technique, can be by user's history log by classification.

Utilizing the mode that keyword bag is classified is supplementing of above-mentioned two kinds of mode classifications, because user's history log of the hot word that comprises MH370 and so on does not have corresponding subject classification or corresponding network address feature, therefore cannot adopt the means of subject classification device or station address filtering classification to classify, and utilize keyword bag can realize such other user's history log classification.

320, by the user history log corresponding with category of interest, classify by user crowd; Wherein, described user crowd's dimension comprises following at least one: the enlivening the time period of sex, age, occupation, condition, consumer behavior, point of interest and described point of interest;

Wherein, sex comprises man and female; The user history log corresponding to different category of interest, can adopt the different ages to choose mode, for example, in the time that category of interest is novel, the mode of choosing at age is " under-18s ", " 18-24 year ", " 24-36 year ", " 36-50 year " and " more than 50 years old "; When category of interest is when giving birth to children, the mode of choosing at age is " 20-30 year ", " 30-40 year " and " more than 40 years old "; Occupation can adopt existing category of employment classification; Condition can comprise pregnant baby and deformity; Consumption habit can comprise the classifications such as household electrical appliances number, house property and cosmetics; Point of interest refers to the concrete content of paying close attention to of user under specific category of interest, and for example, under the category of interest of fashion, point of interest can comprise: star, luxury goods and clothing matching etc.; The time period of enlivening of described point of interest can comprise: morning, afternoon and evening.

It should be noted that, the user history log corresponding to different category of interest, can classify according to different user crowds' dimension.

In addition, user crowd's dimension choose the impact on user interest model precision and complexity, similar on the impact of user interest model precision and complexity with Preset Time granularity, repeat no more herein.

330, the use context data in user's history log corresponding with category of interest and user crowd is added up, as described user interest model.

This step, specifically by statistic of classification, obtains user crowd, uses the incidence relation between context data and category of interest.

The technical scheme of the present embodiment, by user's history log is carried out to statistic of classification by category of interest, user crowd and use context data successively, obtain user interest model, because the user interest model obtaining includes user crowd, uses the incidence relation between context data and category of interest, therefore the precision of user interest model is high, and then can to improve according to this user interest model be user carries out information pushing accuracy rate in the situation of presence.

Embodiment tetra-

Refer to Fig. 4, the structural representation of the pusher of a kind of information providing for the embodiment of the present invention four.This device comprises: user ID and current use context data module 410, recommendation category of interest determination module 420 and pushing module 430.

Wherein, user ID and current use context data module 410 are for obtaining user ID and current use context data, and wherein, described context data comprises following at least one: time, place, networking means and terminal type; Recommend category of interest determination module 420 for according to the user interest model of setting up in advance, determine the recommendation category of interest being associated with described user and current use context data; Pushing module 430 for according to described recommendation category of interest to user's pushed information.

The technical scheme of the present embodiment, by the user interest model of setting up in advance, determine the recommendation category of interest being associated with user and current use context data, and push to user, therefore can be in conjunction with user itself, and in conjunction with current use context data, push current required content to user, and improve the accuracy rate of the information pushing to user.

In such scheme, described device can also comprise: user's history log acquisition module 401 and user interest model acquisition module 402.

Wherein, user's history log acquisition module 401 is for before obtaining user profile and current use context data, obtain user's history log, wherein, described user's history log at least comprises: user ID, the user's historical behavior data that are associated with user ID and use context data; User interest model acquisition module 402 for by user's history log according to category of interest, user crowd with use context data, carry out statistic of classification, obtain described user interest model.

User interest model acquisition module 402 preferably includes: the first taxon 4021, the second taxon 4022 and statistic unit 4023.

Wherein, the first taxon 4021 is for classifying user's history log by category of interest; The second taxon 4022, for by the user history log corresponding with category of interest, is classified by user crowd; Wherein, described user crowd's dimension comprises following at least one: the enlivening the time period of sex, age, occupation, condition, consumer behavior, point of interest and described point of interest; Statistic unit 4023 is added up for the use context data of the user's history log to corresponding with category of interest and user crowd, as described user interest model.

Wherein, the implementation that user's history log is classified by category of interest has multiple, and described the first taxon 4021 comprises: subject classification device classification subelement 4021a, station address tagsort subelement 4021b and/or keyword bag classification subelement 4021c.

Wherein, subject classification device classification subelement 4021a is used for utilizing subject classification device, and described user's history log is carried out to the classification of category of interest; Station address tagsort subelement 4021b, for using station address feature as default category of interest, carries out filtering classification to station address corresponding to user's historical behavior data in user's history log; Keyword bag classification subelement 4021c, for using keyword bag as default category of interest, utilizes keyword bag matching technique, and user's history log is classified by default category of interest.

In such scheme, recommend category of interest determination module 420 to preferably include: to recommend probability calculation unit 421 and recommend category of interest determining unit 422.

Wherein, recommend probability calculation unit 421 for calculate the recommendation probability w of category of interest according to following formula:

W=P (category of interest | crowd) × P (crowd | user)

Wherein, P (crowd | user) belongs to a certain crowd's the ratio that enlivens number of days and this user and always enliven number of days for this user this user; P (category of interest | crowd) be in the crowd under this user, in setting a period of time, in the user number of this category of interest of concern and this crowd, pay close attention to the ratio of total user number of this category of interest.Recommend category of interest determining unit 422 for by recommend probability the highest or recommend probability to be greater than category of interest predetermined threshold value or that recommend the predetermined number that probability is higher, be defined as the recommendation category of interest of this user under this sight.

In such scheme, preferably, described device also comprises: for interest model correcting module, and for obtaining the click feedback data of user to category of interest, the use context data during according to user's clicking operation and click feedback data correction user interest model.

The pusher of the information that the embodiment of the present invention provides can be carried out the method for pushing of the information that any embodiment of the present invention provides, and possesses the corresponding functional module of manner of execution and beneficial effect.

Finally it should be noted that: above each embodiment is only for technical scheme of the present invention is described, but not be limited; In embodiment, preferred embodiment, be not limited, to those skilled in the art, the present invention can have various changes and variation.All any amendments of doing, be equal to replacement, improvement etc., within protection scope of the present invention all should be included within spirit of the present invention and principle.

Claims (12)

1. a method for pushing for information, is characterized in that, comprising:
Obtain user ID and current use context data, wherein, described context data comprises following at least one: time, place, networking means and terminal type;
According to the user interest model of setting up in advance, determine the recommendation category of interest being associated with described user and current use context data;
According to described recommendation category of interest to user's pushed information.
2. method according to claim 1, is characterized in that, before obtaining user profile and current use context data, also comprises:
Obtain user's history log, wherein, described user's history log at least comprises: user ID, the user's historical behavior data that are associated with user ID and use context data;
User's history log, according to category of interest, user crowd and use context data, is carried out to statistic of classification, obtain described user interest model.
3. method according to claim 2, is characterized in that, user's history log, according to category of interest, user crowd and use context data, is carried out to statistic of classification, obtains described user interest model, comprising:
User's history log is classified by category of interest;
By the user history log corresponding with category of interest, classify by user crowd; Wherein, described user crowd's dimension comprises following at least one: the enlivening the time period of sex, age, occupation, condition, consumer behavior, point of interest and described point of interest;
Use context data in user's history log corresponding with category of interest and user crowd is added up, as described user interest model.
4. method according to claim 3, is characterized in that, user's history log, by category of interest classification, being comprised:
Utilize subject classification device, described user's history log is carried out to the classification of category of interest; And/or
Using station address feature as default category of interest, station address corresponding to user's historical behavior data in user's history log carried out to filtering classification; And/or
Using keyword bag as default category of interest, utilize keyword bag matching technique, user's history log is classified by default category of interest.
5. method according to claim 2, is characterized in that, according to the user interest model of setting up in advance, determines the recommendation category of interest being associated with described user and current use context data, comprising:
Calculate the recommendation probability w of category of interest according to following formula:
W=P (crowd | user) × P (category of interest | crowd)
Wherein, P (crowd | user) belongs to a certain crowd's the ratio that enlivens number of days and this user and always enliven number of days for this user this user; P (category of interest | crowd) be in the crowd under this user, in setting a period of time, in the user number of this category of interest of concern and this crowd, pay close attention to the ratio of total user number of this category of interest;
By recommend probability the highest or recommend probability to be greater than category of interest predetermined threshold value or that recommend the higher predetermined number of probability, be defined as the recommendation category of interest of this user under this sight.
6. method according to claim 1, is characterized in that, also comprises:
Obtain the click feedback data of user to category of interest, the use context data during according to user's clicking operation and click feedback data correction user interest model.
7. a pusher for information, is characterized in that, comprising:
User ID and current use context data module, for obtaining user ID and current use context data, wherein, described context data comprises following at least one: time, place, networking means and terminal type;
Recommend category of interest determination module, for according to the user interest model of setting up in advance, determine the recommendation category of interest being associated with described user and current use context data;
Pushing module, for according to described recommendation category of interest to user's pushed information.
8. device according to claim 7, is characterized in that, described device also comprises:
User's history log acquisition module, for before obtaining user profile and current use context data, obtain user's history log, wherein, described user's history log at least comprises: user ID, the user's historical behavior data that are associated with user ID and use context data;
User interest model acquisition module, for by user's history log according to category of interest, user crowd with use context data, carry out statistic of classification, obtain described user interest model.
9. device according to claim 8, is characterized in that, user interest model acquisition module comprises:
The first taxon, for classifying user's history log by category of interest;
The second taxon, for by the user history log corresponding with category of interest, classifies by user crowd; Wherein, described user crowd's dimension comprises following at least one: the enlivening the time period of sex, age, occupation, condition, consumer behavior, point of interest and described point of interest;
Statistic unit, adds up for the use context data of the user's history log to corresponding with category of interest and user crowd, as described user interest model.
10. device according to claim 9, is characterized in that, the first taxon, comprising:
Subject classification device classification subelement, for utilizing subject classification device, carries out the classification of category of interest by described user's history log; And/or
Station address tagsort subelement, for using station address feature as default category of interest, carries out filtering classification to station address corresponding to user's historical behavior data in user's history log; And/or
Keyword bag classification subelement, for using keyword bag as default category of interest, utilizes keyword bag matching technique, and user's history log is classified by default category of interest.
11. devices according to claim 8, is characterized in that, recommend category of interest determination module, comprising:
Recommend probability calculation unit, for calculate the recommendation probability w of category of interest according to following formula:
W=P (crowd | user) × P (category of interest | crowd)
Wherein, P (crowd | user) belongs to a certain crowd's the ratio that enlivens number of days and this user and always enliven number of days for this user this user; P (category of interest | crowd) be in the crowd under this user, in setting a period of time, in the user number of this category of interest of concern and this crowd, pay close attention to the ratio of total user number of this category of interest;
Recommend category of interest determining unit, for by recommend probability the highest or recommend probability to be greater than category of interest predetermined threshold value or that recommend the predetermined number that probability is higher, be defined as the recommendation category of interest of this user under this sight.
12. devices according to claim 7, is characterized in that, also comprise:
For interest model correcting module, for obtaining the click feedback data of user to category of interest, the use context data during according to user's clicking operation and click feedback data correction user interest model.
CN201410256923.3A 2014-06-10 2014-06-10 Method and device for pushing information CN104008184A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410256923.3A CN104008184A (en) 2014-06-10 2014-06-10 Method and device for pushing information

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410256923.3A CN104008184A (en) 2014-06-10 2014-06-10 Method and device for pushing information

Publications (1)

Publication Number Publication Date
CN104008184A true CN104008184A (en) 2014-08-27

Family

ID=51368841

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410256923.3A CN104008184A (en) 2014-06-10 2014-06-10 Method and device for pushing information

Country Status (1)

Country Link
CN (1) CN104008184A (en)

Cited By (38)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104239466A (en) * 2014-09-02 2014-12-24 百度在线网络技术(北京)有限公司 Method and device for recommending user item and equipment
CN104317973A (en) * 2014-11-20 2015-01-28 武汉图歌信息技术有限责任公司 Self-adaptation active service method on basis of geographic space information
CN104363261A (en) * 2014-10-21 2015-02-18 上海动盟网络技术有限公司 Information push method, device and server
CN104462024A (en) * 2014-10-29 2015-03-25 百度在线网络技术(北京)有限公司 Method and device for generating dialogue action strategy model
CN104486422A (en) * 2014-12-17 2015-04-01 百度在线网络技术(北京)有限公司 Information push method and device
CN104574146A (en) * 2014-12-31 2015-04-29 浙江大学 Service recommending method and system
CN104598529A (en) * 2014-12-26 2015-05-06 安徽寰智信息科技股份有限公司 Interactive science popularizing method based on basic user information
CN104731954A (en) * 2015-04-01 2015-06-24 天翼爱音乐文化科技有限公司 Music recommendation method and system based on group perspective
CN104735125A (en) * 2014-12-20 2015-06-24 骆小妮 Entertainment data synchronous method of in-car entertainment system
CN104820863A (en) * 2015-03-27 2015-08-05 北京智慧图科技有限责任公司 Consumer portrait generation method and device
CN104869439A (en) * 2015-05-14 2015-08-26 无锡天脉聚源传媒科技有限公司 Video push method and device
CN105095410A (en) * 2015-07-09 2015-11-25 广东欧珀移动通信有限公司 Setting method, mobile terminal, server and system of song menu
CN105159910A (en) * 2015-07-03 2015-12-16 安一恒通(北京)科技有限公司 Information recommendation method and device
CN105245609A (en) * 2015-10-23 2016-01-13 小米科技有限责任公司 Information pushing method, device, equipment and system
CN105243103A (en) * 2015-09-19 2016-01-13 杭州电子科技大学 Content based push time determination method
CN105338054A (en) * 2015-09-21 2016-02-17 百度在线网络技术(北京)有限公司 Method and device for pushing voice information
CN105373591A (en) * 2015-10-23 2016-03-02 同济大学 Construction method of user interest model based on temporal-spatial behavior pattern in mobile social network
CN105391801A (en) * 2015-12-14 2016-03-09 新华瑞德(北京)网络科技有限公司 Information pushing method and device for application program
CN105550312A (en) * 2015-12-14 2016-05-04 东软集团股份有限公司 Context information processing method and device
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
WO2016127906A1 (en) * 2015-02-15 2016-08-18 阿里巴巴集团控股有限公司 Information pushing method, apparatus, and system
TWI556121B (en) * 2015-08-27 2016-11-01 優像數位媒體科技股份有限公司 Gender prediction method by using webpage surfing behavior
WO2016201631A1 (en) * 2015-06-17 2016-12-22 Yahoo! Inc. Systems and methods for online content recommendation
CN106294775A (en) * 2016-08-11 2017-01-04 合智能科技(深圳)有限公司 Content delivery method and device
CN106354797A (en) * 2016-08-26 2017-01-25 腾讯科技(深圳)有限公司 Data recommendation method and device
CN106445977A (en) * 2015-08-12 2017-02-22 阿里巴巴集团控股有限公司 Picture pushing method and device
WO2017071251A1 (en) * 2015-10-28 2017-05-04 百度在线网络技术(北京)有限公司 Information pushing method and device
CN106649733A (en) * 2016-12-23 2017-05-10 中山大学 Online video recommendation method based on wireless access point situation classification and perception
WO2017084344A1 (en) * 2015-11-16 2017-05-26 北京百度网讯科技有限公司 Information recommendation method and apparatus
CN106909620A (en) * 2017-01-17 2017-06-30 广东小天才科技有限公司 One kind teaches feature music assorting method and system for early
CN107241215A (en) * 2017-05-10 2017-10-10 百度在线网络技术(北京)有限公司 User's behavior prediction method and apparatus
CN107291755A (en) * 2016-04-01 2017-10-24 中国移动通信有限公司研究院 A kind of terminal method for pushing and device
CN107391511A (en) * 2016-05-16 2017-11-24 中国移动通信集团内蒙古有限公司 A kind of information-pushing method and device
CN107465754A (en) * 2017-08-23 2017-12-12 北京搜狐新媒体信息技术有限公司 A kind of news recommends method and apparatus
CN107491474A (en) * 2017-06-26 2017-12-19 北京三快在线科技有限公司 Information recommendation method and device
CN107608792A (en) * 2017-09-12 2018-01-19 中国联合网络通信集团有限公司 Resource regulating method and device
CN108416616A (en) * 2018-02-05 2018-08-17 阿里巴巴集团控股有限公司 The sort method and device of complaints and denunciation classification
CN109829108A (en) * 2019-01-28 2019-05-31 北京三快在线科技有限公司 Information recommendation method, device, electronic equipment and readable storage medium storing program for executing

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080222119A1 (en) * 2007-03-08 2008-09-11 Microsoft Corporation Detecting a user's location, local intent and travel intent from search queries
CN102147903A (en) * 2010-02-05 2011-08-10 李久进 Geographic position-based internet information aggregating, pushing and interacting method
CN102385636A (en) * 2011-12-22 2012-03-21 陈伟 Intelligent searching method and device
CN103617547A (en) * 2013-12-04 2014-03-05 中国联合网络通信集团有限公司 Service recommendation method and system
CN103631954A (en) * 2013-12-13 2014-03-12 百度在线网络技术(北京)有限公司 Individualized recommendation method and device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080222119A1 (en) * 2007-03-08 2008-09-11 Microsoft Corporation Detecting a user's location, local intent and travel intent from search queries
CN102147903A (en) * 2010-02-05 2011-08-10 李久进 Geographic position-based internet information aggregating, pushing and interacting method
CN102385636A (en) * 2011-12-22 2012-03-21 陈伟 Intelligent searching method and device
CN103617547A (en) * 2013-12-04 2014-03-05 中国联合网络通信集团有限公司 Service recommendation method and system
CN103631954A (en) * 2013-12-13 2014-03-12 百度在线网络技术(北京)有限公司 Individualized recommendation method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
郑晓宇: "基于用户兴趣挖掘的个性化信息推送服务", 《情报探索》 *

Cited By (50)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104239466A (en) * 2014-09-02 2014-12-24 百度在线网络技术(北京)有限公司 Method and device for recommending user item and equipment
CN104363261A (en) * 2014-10-21 2015-02-18 上海动盟网络技术有限公司 Information push method, device and server
CN104363261B (en) * 2014-10-21 2018-05-01 上海重盟信息技术有限公司 Information-pushing method, device and server
CN104462024A (en) * 2014-10-29 2015-03-25 百度在线网络技术(北京)有限公司 Method and device for generating dialogue action strategy model
CN104317973A (en) * 2014-11-20 2015-01-28 武汉图歌信息技术有限责任公司 Self-adaptation active service method on basis of geographic space information
CN104317973B (en) * 2014-11-20 2018-08-03 武汉图歌信息技术有限责任公司 A kind of adaptive active service method of geospatial information
CN104486422A (en) * 2014-12-17 2015-04-01 百度在线网络技术(北京)有限公司 Information push method and device
CN104486422B (en) * 2014-12-17 2019-04-02 百度在线网络技术(北京)有限公司 Information-pushing method and device
CN104735125B (en) * 2014-12-20 2019-01-25 南京矛盾防辐射科技有限公司 A kind of recreation data synchronous method of vehicle entertainment system
CN104735125A (en) * 2014-12-20 2015-06-24 骆小妮 Entertainment data synchronous method of in-car entertainment system
CN104598529A (en) * 2014-12-26 2015-05-06 安徽寰智信息科技股份有限公司 Interactive science popularizing method based on basic user information
CN104574146A (en) * 2014-12-31 2015-04-29 浙江大学 Service recommending method and system
WO2016127906A1 (en) * 2015-02-15 2016-08-18 阿里巴巴集团控股有限公司 Information pushing method, apparatus, and system
CN105989127A (en) * 2015-02-15 2016-10-05 阿里巴巴集团控股有限公司 Information pushing method, device and system
CN104820863A (en) * 2015-03-27 2015-08-05 北京智慧图科技有限责任公司 Consumer portrait generation method and device
CN104731954A (en) * 2015-04-01 2015-06-24 天翼爱音乐文化科技有限公司 Music recommendation method and system based on group perspective
CN104731954B (en) * 2015-04-01 2018-01-26 天翼爱音乐文化科技有限公司 Music is had an X-rayed based on group and recommends method and system
CN104869439A (en) * 2015-05-14 2015-08-26 无锡天脉聚源传媒科技有限公司 Video push method and device
CN104869439B (en) * 2015-05-14 2018-03-09 无锡天脉聚源传媒科技有限公司 A kind of video pushing method and device
WO2016201631A1 (en) * 2015-06-17 2016-12-22 Yahoo! Inc. Systems and methods for online content recommendation
CN105159910A (en) * 2015-07-03 2015-12-16 安一恒通(北京)科技有限公司 Information recommendation method and device
CN105095410A (en) * 2015-07-09 2015-11-25 广东欧珀移动通信有限公司 Setting method, mobile terminal, server and system of song menu
CN106445977A (en) * 2015-08-12 2017-02-22 阿里巴巴集团控股有限公司 Picture pushing method and device
TWI556121B (en) * 2015-08-27 2016-11-01 優像數位媒體科技股份有限公司 Gender prediction method by using webpage surfing behavior
CN105243103A (en) * 2015-09-19 2016-01-13 杭州电子科技大学 Content based push time determination method
CN105338054A (en) * 2015-09-21 2016-02-17 百度在线网络技术(北京)有限公司 Method and device for pushing voice information
CN105338054B (en) * 2015-09-21 2019-04-16 百度在线网络技术(北京)有限公司 A kind of method for pushing and device of voice messaging
CN105373591B (en) * 2015-10-23 2019-05-14 同济大学 Based on the user interest model construction method of time-space behavior mode in mobile social networking
CN105373591A (en) * 2015-10-23 2016-03-02 同济大学 Construction method of user interest model based on temporal-spatial behavior pattern in mobile social network
CN105245609A (en) * 2015-10-23 2016-01-13 小米科技有限责任公司 Information pushing method, device, equipment and system
WO2017071251A1 (en) * 2015-10-28 2017-05-04 百度在线网络技术(北京)有限公司 Information pushing method and device
WO2017084344A1 (en) * 2015-11-16 2017-05-26 北京百度网讯科技有限公司 Information recommendation method and apparatus
CN105391801A (en) * 2015-12-14 2016-03-09 新华瑞德(北京)网络科技有限公司 Information pushing method and device for application program
CN105550312A (en) * 2015-12-14 2016-05-04 东软集团股份有限公司 Context information processing method and device
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
CN107291755A (en) * 2016-04-01 2017-10-24 中国移动通信有限公司研究院 A kind of terminal method for pushing and device
CN107391511A (en) * 2016-05-16 2017-11-24 中国移动通信集团内蒙古有限公司 A kind of information-pushing method and device
CN106294775A (en) * 2016-08-11 2017-01-04 合智能科技(深圳)有限公司 Content delivery method and device
CN106354797B (en) * 2016-08-26 2018-09-11 腾讯科技(深圳)有限公司 Data recommendation method and device
CN106354797A (en) * 2016-08-26 2017-01-25 腾讯科技(深圳)有限公司 Data recommendation method and device
CN106649733A (en) * 2016-12-23 2017-05-10 中山大学 Online video recommendation method based on wireless access point situation classification and perception
CN106649733B (en) * 2016-12-23 2020-04-10 中山大学 Online video recommendation method based on wireless access point context classification and perception
CN106909620A (en) * 2017-01-17 2017-06-30 广东小天才科技有限公司 One kind teaches feature music assorting method and system for early
CN107241215A (en) * 2017-05-10 2017-10-10 百度在线网络技术(北京)有限公司 User's behavior prediction method and apparatus
CN107491474A (en) * 2017-06-26 2017-12-19 北京三快在线科技有限公司 Information recommendation method and device
CN107465754A (en) * 2017-08-23 2017-12-12 北京搜狐新媒体信息技术有限公司 A kind of news recommends method and apparatus
CN107608792A (en) * 2017-09-12 2018-01-19 中国联合网络通信集团有限公司 Resource regulating method and device
CN108416616A (en) * 2018-02-05 2018-08-17 阿里巴巴集团控股有限公司 The sort method and device of complaints and denunciation classification
WO2019149145A1 (en) * 2018-02-05 2019-08-08 阿里巴巴集团控股有限公司 Compliant report class sorting method and apparatus
CN109829108A (en) * 2019-01-28 2019-05-31 北京三快在线科技有限公司 Information recommendation method, device, electronic equipment and readable storage medium storing program for executing

Similar Documents

Publication Publication Date Title
US20190095528A1 (en) Override of Automatically Shared Meta-Data of Media
JP2018113049A (en) Generation of recommended retrieval query on online social network
US20170076353A1 (en) Context-influenced application recommendations
US9881042B2 (en) Internet based method and system for ranking individuals using a popularity profile
US20160371709A1 (en) Determining information inter-relationships from distributed group discussions
US20160171103A1 (en) Systems and Methods for Gathering, Merging, and Returning Data Describing Entities Based Upon Identifying Information
US9892431B1 (en) Temporal features in a messaging platform
Garcin et al. Offline and online evaluation of news recommender systems at swissinfo. ch
US8909569B2 (en) System and method for revealing correlations between data streams
US9712588B1 (en) Generating a stream of content for a channel
US9645797B2 (en) Development system and method for providing external functionality
US20170286539A1 (en) User profile stitching
JP6511024B2 (en) Consumer-driven advertising system
US10491694B2 (en) Method and system for measuring user engagement using click/skip in content stream using a probability model
US9734258B2 (en) Method and system of displaying cross-website information
US9374396B2 (en) Recommended content for an endorsement user interface
US20160117696A1 (en) Method and system for determining on-line influence in social media
CA2848605C (en) Structured objects and actions on a social networking system
US20160275127A1 (en) Systems and methods for presenting content
CN104281961B (en) For the advertisement in on-line system and the quality score system of content
US9405734B2 (en) Image manipulation for web content
US20160094646A1 (en) Trending of aggregated personalized information streams and multi-dimensional graphical depiction thereof
US9626443B2 (en) Searching and accessing application functionality
US8775429B2 (en) Methods and systems for analyzing data of an online social network
Gretzel et al. Differences in consumer-generated media adoption and use: A cross-national perspective

Legal Events

Date Code Title Description
PB01 Publication
C06 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20140827

RJ01 Rejection of invention patent application after publication