CN105279206A - Intelligent recommendation method and system - Google Patents
Intelligent recommendation method and system Download PDFInfo
- Publication number
- CN105279206A CN105279206A CN201410360843.2A CN201410360843A CN105279206A CN 105279206 A CN105279206 A CN 105279206A CN 201410360843 A CN201410360843 A CN 201410360843A CN 105279206 A CN105279206 A CN 105279206A
- Authority
- CN
- China
- Prior art keywords
- user behavior
- behavior data
- data
- described user
- statistics
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides an intelligent recommendation method and system. The method comprises the steps of: S1, receiving and storing user behavior data; S2, performing multi-dimensional analysis and statistics on the user behavior data to obtain a statistic result of the user behavior data; and S3, according to the statistic result, sorting the user behavior data, and pushing data with the same type as the user behavior data to a user according to the sorting. Compared with a user-oriented intelligent recommendation method and system which directly analyzing user behaviors and preferences insufficiently, the intelligent recommendation method and system provided by the invention have the characteristics that the user behavior data are subjected to the multi-dimensional analysis and statistics through the fusion of powerful analytic and statistic capability of a server, the user behavior data is analyzed more deeply and carefully, and the understanding of user behaviors and habits is enhanced, so that it is ensured that the user behaviors and habits are better satisfied in subsequent intelligent recommendation.
Description
Technical field
The present invention relates to development of Mobile Internet technology application, particularly relate to a kind of intelligent recommendation method and system.
Background technology
Along with the development of Internet technology, the scale of internet and coverage rate are more and more wider, bring the swift and violent growth of information thereupon.The information that user will read interested hobby will waste time and energy, and intelligent recommendation system, as a kind of important means of information filtering, is the effective ways of current solution information overload problem.Intelligent recommendation be by system dominate user browse order, guide user find need information.High-quality intelligent recommendation system can make user produce dependence to this system.
At present, the user-defined categorizing selection of most intelligent recommendation system general and key word screening come to user's recommendation information, such as focus information, although this system also analyzes user behavior and hobby, but be more behavior and the hobby by analyzing user to the analysis of the information that internet is propagated, instead of directly the behavior of user and hobby are analyzed, therefore the information pushed is the focus information on internet mostly, more for masses, and for intelligent recommendation of low quality of specific user, user cannot be found fast on the interface of intelligent recommendation system and meet the information oneself be accustomed to and liked.
Summary of the invention
Object of the present invention is to provide a kind of intelligent recommendation method on the one hand, another object of the present invention is to provide a kind of intelligent recommendation system, thus solves the foregoing problems existed in prior art.
To achieve these goals, the technical solution used in the present invention is as follows:
A kind of intelligent recommendation method, comprises the steps:
S1, receives and stores user behavior data;
S2, carries out analysis and the statistics of various dimensions, obtains the statistics of described user behavior data to described user behavior data;
S3, according to described statistics, sorts to described user behavior data, then according to described sequence by the data-pushing with described user behavior data identical category to user.
Preferably, step S3, comprises further:
S31, according to described statistics, sets weight and the coefficient of described user behavior data;
S32, described weight is multiplied with described coefficient, calculates the weight score value of described user behavior data;
S33, according to the weight score value of described user behavior data, sorts to described user behavior data;
S34, according to described sequence by the data-pushing with described user behavior data identical category to user.
Preferably, before step S1, also comprise the steps:
Described user behavior data is verified, gathers effective described user behavior data;
Queue storage is carried out to described effective described user behavior data, obtains the described user behavior data that queue stores;
The described user behavior data that described queue stores is compressed, obtains the described user behavior data compressed;
Send the described user behavior data of described compression to service end.
Preferably, in step s3, described described user behavior data is sorted after, described again according to described sequence by the data-pushing with described user behavior data identical category to before user, also comprise step, according to the priority principle of setting, second time sequence is carried out to described user behavior data.
Particularly, described priority principle comprises: editor selects priority principle and ageing priority principle.
Particularly, described various dimensions comprise: time, place and information category.
Particularly, described in step S2, the analysis of various dimensions is carried out to described user behavior data and statistics is:
To user habit place, in the custom time period, the information of all kinds or user preferences kind analyzes and adds up;
Or
To data analysis and the statistics of user habit place, all kinds or user preferences kind;
Or
To data analysis and the statistics of user habit time, all kinds or user preferences kind.
A kind of intelligent recommendation system, comprising:
Data reception module: for receiving and storing user behavior data;
Data processing module: for analyzing described user behavior data and add up, obtains the statistics of the described user behavior data based on multiple dimension;
Data-pushing module: for according to described statistics, described user behavior data is sorted, then according to described sequence by the data-pushing with described user behavior data identical category to user.
Preferably, described data-pushing module is further: for according to described statistics, set weight and the coefficient of described user behavior data; According to the weight score value of the described user behavior data that the product of described weight and described coefficient calculates, described user behavior data is sorted, then according to described sequence by the data-pushing with described user behavior data identical category to user.
The invention has the beneficial effects as follows: directly analyze not to the behavior of user with liking with prior art, user cannot be found fast on the interface of intelligent recommendation system and meet the information oneself be accustomed to and liked, the intelligent recommendation method of taking as the leading factor with user is compared with intelligent recommendation system, the present invention is by incorporating the powerful analysis of server and statistical power, user behavior data is carried out to analysis and the statistics of various dimensions, user behavior data is carried out more deeply and careful analysis, increase the understanding to user behavior and custom, thus ensure that in follow-up intelligent recommendation, more meet user behavior and custom.
Accompanying drawing explanation
Fig. 1 is the flow chart of steps of a kind of intelligent recommendation method that the embodiment of the present invention provides;
Fig. 2 is the flow chart of steps of the another kind of intelligent recommendation method that the embodiment of the present invention provides;
Fig. 3 is that a kind of dimension that the embodiment of the present invention provides selects interface schematic diagram;
Fig. 4 is that a kind of weight that the embodiment of the present invention provides selects interface schematic diagram;
Fig. 5 is the network topological diagram that the embodiment of the present invention provides;
Fig. 6 is a kind of intelligent recommendation system framework figure that the embodiment of the present invention provides;
Fig. 7 is the another kind of intelligent recommendation system framework figure that the embodiment of the present invention provides.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with accompanying drawing, the present invention is further elaborated.Should be appreciated that embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.
As shown in Figure 1, reality of the present invention executes a kind of intelligent recommendation method that example provides, and comprises the steps:
S1, receives and stores user behavior data;
S2, carries out analysis and the statistics of various dimensions, obtains the statistics of described user behavior data to described user behavior data;
S3, according to described statistics, sorts to described user behavior data, then by the data-pushing with described user behavior data identical category to user.
Adopt technique scheme, by incorporating the powerful analysis of server and statistical power, user behavior data is carried out to analysis and the statistics of various dimensions, user behavior data is carried out more deeply and careful analysis, increase the understanding to user behavior and custom, thus ensure that in follow-up intelligent recommendation, more meet user behavior and custom.
As shown in Figure 2, in a kind of intelligent recommendation method that one embodiment of the present of invention provide, step S3, comprises further:
S31, according to described statistics, sets weight and the coefficient of described user behavior data;
As will be understood by the skilled person in the art, wherein, the weight of described user behavior data and coefficient can be unified according to the acquiescence scene arranged to regulate, and also can regulate voluntarily according to providing enterprise's wish of product;
S32, by described weight and described multiplication, calculates the weight score value of described user behavior data;
S33, according to the weight score value of described user behavior data, sorts to described user behavior data;
S34, according to described sequence by the data-pushing with described user behavior data identical category to user.
Such as, based on the time, place and information category three dimensions, analytic statistics is carried out to user behavior data, find user within certain time period, certain place obtains the data of certain kind of information, when this user behavior data is larger, illustrate that this user is within this time period, this place is large to this kind of information amount to obtain, then intelligent recommendation system just can increase weight and the coefficient of this user behavior data, and then the weight score value obtaining this user behavior data is larger, system can be selected to push this place to this user according to the size of weight score value, this kind of information in this time period, thus ensure that the result of intelligent recommendation meets behavior and the custom of user more.
Wherein, propelling movement can specifically comprise the steps:
A. user inputs user name and code entry App (also can anonymity log in);
B.App submits user's logon information to server;
C. server is according to user profile, obtains homepage and pushes content, is pushed to App end by interface;
D. anonymous obtains default treatment queue.
The intelligent recommendation method that embodiments of the invention provide, according to analysis and the statistics of the various dimensions to user behavior data, the weight of setting user behavior data and coefficient, calculate the weight score value of user behavior data, the weight score value of user behavior data can be regulated according to analysis and statistics, and then intelligent recommendation can be changed along with the change of user behavior, all the time be consistent with the behavior of user and custom, thus provide the recommendation more meeting its behavior and custom for user.
In one embodiment of the invention, various dimensions comprise: time, place and information category.Wherein, location information can be obtained by LBS interface.In the present embodiment, location information dimension, for paying close attention to geographical location information residing when user uses system acquisition information, by information such as the content-keyword in comparison database, addresses, improves the weight score value of the user behavior data in this geographic position;
Time dimension for paying close attention to the ageing of user behavior data, time point from current time more close to, improve the weight score value of the user behavior data of this time point
And weight score value is higher, the rank of family behavioral data is more forward, and more forward at the congener information sorting of propelling movement phase, user just more first can see this congener information, is consistent with its custom and behavior.
The dimension administration interface of user behavior data can as shown in Figure 3, and weight administration interface can be as shown in Figure 4.
Particularly, carrying out multi dimensional analysis to described user behavior data described in rapid S2 with statistics can be:
To user habit place, in the custom time period, the information of all kinds or user preferences kind analyzes and adds up;
Or
To data analysis and the statistics of user habit place, all kinds or user preferences kind;
Or
To data analysis and the statistics of user habit time, all kinds or user preferences kind;
In above-mentioned any one situation, after obtaining the behavior of user's obtaining information, then after system extended meeting according to this statistics, select whether to push in this place and/or this category information in this time period and the order pushing this category information to this user to this user, if this user of statistical result showed is in this place and/or to obtain the behavioral data of this category information in this time period larger, then system preferentially can push this category information to this user in this place and/or in this time period, if this user behavior data of statistical result showed is less, then system preferentially can not push this category information to this user.
In one embodiment of the invention, in step s3, described according to described score value described user behavior data sorted after, described again according to described sequence by the data-pushing with described user behavior data identical category to before user, also comprise step, according to the priority principle of setting, second time sequence is carried out to described user behavior data.Wherein, described priority principle comprises: editor selects priority principle and ageing priority principle.Such as, for different user behavior datas, if statistics is identical, then when cannot sort according to statistics, then carry out second time sequence according to the priority principle of setting.This priority principle can select priority principle for editor, also can be ageing priority principle.
In one embodiment of the invention, before step S1, also comprise the steps:
Described user behavior data is verified, gathers effective described user behavior data; After a series of checking and encryption, just approve the validity of data;
Queue storage is carried out to described effective described user behavior data, obtains the described user behavior data that queue stores; Queue stores, be exactly after data acquisition, do not send data to service end at once, send together after continuation collection is neat to troop, as run into power-off, suspension in process, then dress the ranks when software starts again or applied statistics has served time point and be sent to service end;
The described user behavior data that described queue stores is compressed, obtains the described user behavior data compressed;
Send the described user behavior data of described compression to service end.
In SDK, create multiple task queue, store dissimilar user behavior, after user behavior data arrives the magnitude preset, compress and after encrypting, unify to be submitted to service end.Mass data collection and communication often make the quick consumes power of mobile terminal and flow, adopt queue storage and Compression Transmission Technology, greatly can reduce power consumption and the flow of mobile terminal; Wherein, queue memory technology can reduce energy loss 60%, and Compression Transmission Technology can reduce energy loss 50% further; Also improve the concurrent capability of server simultaneously, and utilize limited resource to process more data content at one time.
The above embodiment of the present invention can adopt network topological diagram as shown in Figure 5, wherein, in order to prevent the input and output bottleneck of server, and Data Collection and process separation principle, virtualized server 01 is divided into " data, services 01_a " and " data, services 01_b " after visit capacity reaches 100,000 grades, and they are respectively as data processing database and data processing and propelling data storehouse.
As shown in Figure 6, the embodiment of the present invention additionally provides a kind of intelligent recommendation system, comprising:
Data reception module: for receiving and storing user behavior data;
Data processing module: for analyzing described user behavior data and add up, obtains the statistics of the described user behavior data based on multiple dimension;
Data-pushing module: for according to described statistics, described user behavior data is sorted, then according to described sequence by the data-pushing with described user behavior data identical category to user.
Adopt above-mentioned commending system, by incorporating the powerful analysis of server and statistical power, user behavior data is carried out to analysis and the statistics of various dimensions, user behavior data is carried out more deeply and careful analysis, increase the understanding to user behavior and custom, thus ensure that in follow-up intelligent recommendation, more meet user behavior and custom.
As shown in Figure 7, embodiments provide a kind of intelligent recommendation system, described data-pushing module is further: for according to described statistics, set weight and the coefficient of described user behavior data; According to the weight score value of the described user behavior data that the product of described weight and described coefficient calculates, described user behavior data is sorted, then according to described sequence by the data-pushing with described user behavior data identical category to user.
The intelligent recommendation system that embodiments of the invention provide, according to analysis and the statistics of the various dimensions to user behavior data, the weight of setting user behavior data and coefficient, calculate the weight score value of user behavior data, the weight score value of user behavior data can be regulated according to analysis and statistics, and then intelligent recommendation can be changed along with the change of user behavior, all the time be consistent with the behavior of user and custom, thus provide the recommendation more meeting its behavior and custom for user.
The intelligent recommendation system that embodiments of the invention provide, can carry multiple mobile terminals product and jointly use, and be respectively each product provide independently UUID be used as identify, use a set of backstage to carry out supporting simultaneously.
The intelligent recommendation method and system that embodiments of the invention provide, for the system that Android and iOS etc. is different, adopts response mechanism to carry out the propelling movement of data, and such as, Android uses that Baidu pushes, iOS uses apple to push.
By adopting technique scheme disclosed by the invention, obtain effect useful as follows:
The intelligent recommendation method and system that embodiments of the invention provide, by incorporating the powerful analysis of server and statistical power, user behavior data is carried out to analysis and the statistics of various dimensions, user behavior data is carried out more deeply and careful analysis, increase the understanding to user behavior and custom, thus ensure that in follow-up intelligent recommendation, more meet user behavior and custom; Further, according to analysis and the statistics of the various dimensions to user behavior data, the weight of setting user behavior data and coefficient, calculate the weight score value of user behavior data, the weight score value of user behavior data can be regulated according to analysis and statistics, and then intelligent recommendation can be changed along with the change of user behavior, be consistent with the behavior of user and custom all the time, thus provide the recommendation more meeting its behavior and custom for user.
The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the premise without departing from the principles of the invention; can also make some improvements and modifications, these improvements and modifications also should look protection scope of the present invention.
Claims (9)
1. an intelligent recommendation method, is characterized in that, comprises the steps:
S1, receives and stores user behavior data;
S2, carries out analysis and the statistics of various dimensions, obtains the statistics of described user behavior data to described user behavior data;
S3, according to described statistics, sorts to described user behavior data, then according to described sequence by the data-pushing with described user behavior data identical category to user.
2. intelligent recommendation method according to claim 1, is characterized in that step S3 comprises further:
S31, according to described statistics, sets weight and the coefficient of described user behavior data;
S32, described weight is multiplied with described coefficient, calculates the weight score value of described user behavior data;
S33, according to the weight score value of described user behavior data, sorts to described user behavior data;
S34, according to described sequence by the data-pushing with described user behavior data identical category to user.
3. intelligent recommendation method according to claim 1 and 2, is characterized in that, before step S1, also comprises the steps:
Described user behavior data is verified, gathers effective described user behavior data;
Queue storage is carried out to described effective described user behavior data, obtains the described user behavior data that queue stores;
The described user behavior data that described queue stores is compressed, obtains the described user behavior data compressed;
Send the described user behavior data of described compression to service end.
4. intelligent recommendation method according to claim 1 and 2, it is characterized in that, in step s3, described described user behavior data is sorted after, described again according to described sequence by the data-pushing with described user behavior data identical category to before user, also comprise step, according to the priority principle of setting, second time sequence is carried out to described user behavior data.
5. intelligent recommendation method according to claim 4, is characterized in that, described priority principle comprises: editor selects priority principle and ageing priority principle.
6. intelligent recommendation method according to claim 1 and 2, is characterized in that, described various dimensions comprise: time, place and information category.
7. intelligent recommendation method according to claim 6, is characterized in that, described in step S2, described user behavior data is carried out to analysis and the statistics of various dimensions, is specially:
To user habit place, in the custom time period, the information of all kinds or user preferences kind analyzes and adds up;
Or
To data analysis and the statistics of user habit place, all kinds or user preferences kind;
Or
To data analysis and the statistics of user habit time, all kinds or user preferences kind.
8. an intelligent recommendation system, is characterized in that, comprising:
Data reception module: for receiving and storing user behavior data;
Data processing module: for analyzing described user behavior data and add up, obtains the statistics of the described user behavior data based on multiple dimension;
Data-pushing module: for according to described statistics, described user behavior data is sorted, then according to described sequence by the data-pushing with described user behavior data identical category to user.
9. intelligent recommendation system according to claim 8, is characterized in that, described data-pushing module is further: for according to described statistics, set weight and the coefficient of described user behavior data; According to the weight score value of the described user behavior data that the product of described weight and described coefficient calculates, described user behavior data is sorted, then according to described sequence by the data-pushing with described user behavior data identical category to user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410360843.2A CN105279206A (en) | 2014-07-25 | 2014-07-25 | Intelligent recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410360843.2A CN105279206A (en) | 2014-07-25 | 2014-07-25 | Intelligent recommendation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105279206A true CN105279206A (en) | 2016-01-27 |
Family
ID=55148226
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410360843.2A Pending CN105279206A (en) | 2014-07-25 | 2014-07-25 | Intelligent recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105279206A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017162070A1 (en) * | 2016-03-25 | 2017-09-28 | 阿里巴巴集团控股有限公司 | Method and system for recommending merchandise based on time |
| CN107292398A (en) * | 2017-06-09 | 2017-10-24 | 携程旅游信息技术（上海）有限公司 | Athletic ground intelligent management system |
| WO2017219527A1 (en) * | 2016-06-24 | 2017-12-28 | 乐视控股（北京）有限公司 | Data statistical analysis method and system for an intelligent terminal |
| CN108235126A (en) * | 2017-12-15 | 2018-06-29 | 北京小度互娱科技有限公司 | It is inserted into the method and apparatus of recommendation information in video |
| CN109656903A (en) * | 2018-10-30 | 2019-04-19 | 成都飞机工业（集团）有限责任公司 | A kind of method of intelligently pushing control center module |
| CN110321487A (en) * | 2019-07-01 | 2019-10-11 | 上海硕恩网络科技股份有限公司 | A kind of accurate label recommendations system and its workflow |
| CN111698129A (en) * | 2020-06-09 | 2020-09-22 | 湖南大众传媒职业技术学院 | User flow and behavior analysis system |
| CN115022098A (en) * | 2022-08-09 | 2022-09-06 | 北京瑞莱智慧科技有限公司 | Artificial intelligence safety target range content recommendation method, device and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008191702A (en) * | 2007-01-31 | 2008-08-21 | Sharp Corp | Preference information collecting system and information recommendation system using the same |
| CN102654860A (en) * | 2011-03-01 | 2012-09-05 | 北京彩云在线技术开发有限公司 | Personalized music recommendation method and system |
| CN103136435A (en) * | 2011-11-30 | 2013-06-05 | 深圳市快播科技有限公司 | System, method and game platform capable of recommending games in personalization mode |
| CN103325052A (en) * | 2013-07-03 | 2013-09-25 | 姚明东 | Commodity recommendation method based on multidimensional user consumption propensity modeling |
| CN103353920A (en) * | 2013-05-31 | 2013-10-16 | 北京百度网讯科技有限公司 | Method and device for recommending games based on SNS |
| CN103870448A (en) * | 2012-12-07 | 2014-06-18 | 盛乐信息技术（上海）有限公司 | Method and method for recommending data |
| CN103886090A (en) * | 2014-03-31 | 2014-06-25 | 北京搜狗科技发展有限公司 | Content recommendation method and device based on user favorites |
-
2014
- 2014-07-25 CN CN201410360843.2A patent/CN105279206A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008191702A (en) * | 2007-01-31 | 2008-08-21 | Sharp Corp | Preference information collecting system and information recommendation system using the same |
| CN102654860A (en) * | 2011-03-01 | 2012-09-05 | 北京彩云在线技术开发有限公司 | Personalized music recommendation method and system |
| CN103136435A (en) * | 2011-11-30 | 2013-06-05 | 深圳市快播科技有限公司 | System, method and game platform capable of recommending games in personalization mode |
| CN103870448A (en) * | 2012-12-07 | 2014-06-18 | 盛乐信息技术（上海）有限公司 | Method and method for recommending data |
| CN103353920A (en) * | 2013-05-31 | 2013-10-16 | 北京百度网讯科技有限公司 | Method and device for recommending games based on SNS |
| CN103325052A (en) * | 2013-07-03 | 2013-09-25 | 姚明东 | Commodity recommendation method based on multidimensional user consumption propensity modeling |
| CN103886090A (en) * | 2014-03-31 | 2014-06-25 | 北京搜狗科技发展有限公司 | Content recommendation method and device based on user favorites |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017162070A1 (en) * | 2016-03-25 | 2017-09-28 | 阿里巴巴集团控股有限公司 | Method and system for recommending merchandise based on time |
| WO2017219527A1 (en) * | 2016-06-24 | 2017-12-28 | 乐视控股（北京）有限公司 | Data statistical analysis method and system for an intelligent terminal |
| CN107292398A (en) * | 2017-06-09 | 2017-10-24 | 携程旅游信息技术（上海）有限公司 | Athletic ground intelligent management system |
| CN108235126A (en) * | 2017-12-15 | 2018-06-29 | 北京小度互娱科技有限公司 | It is inserted into the method and apparatus of recommendation information in video |
| CN108235126B (en) * | 2017-12-15 | 2020-03-27 | 北京小度互娱科技有限公司 | Method and device for inserting recommendation information in video |
| CN109656903A (en) * | 2018-10-30 | 2019-04-19 | 成都飞机工业（集团）有限责任公司 | A kind of method of intelligently pushing control center module |
| CN110321487A (en) * | 2019-07-01 | 2019-10-11 | 上海硕恩网络科技股份有限公司 | A kind of accurate label recommendations system and its workflow |
| CN111698129A (en) * | 2020-06-09 | 2020-09-22 | 湖南大众传媒职业技术学院 | User flow and behavior analysis system |
| CN115022098A (en) * | 2022-08-09 | 2022-09-06 | 北京瑞莱智慧科技有限公司 | Artificial intelligence safety target range content recommendation method, device and storage medium |
| CN115022098B (en) * | 2022-08-09 | 2022-11-18 | 北京瑞莱智慧科技有限公司 | Artificial intelligence safety target range content recommendation method, device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105279206A (en) | Intelligent recommendation method and system | |
| US11669431B2 (en) | Analytics personalization framework | |
| WO2017071251A1 (en) | Information pushing method and device | |
| CN108763502B (en) | Information recommendation method and system | |
| CN102073699B (en) | For improving the method for Search Results, device and equipment based on user behavior | |
| CN102054003B (en) | Methods and systems for recommending network information and creating network resource index | |
| CN110321422A (en) | Method, method for pushing, device and the equipment of on-line training model | |
| CN102663064B (en) | A kind of disposal route of favorites data and device | |
| CN103780677A (en) | Method for performing classified information push and system thereof | |
| CN111680254B (en) | Content recommendation method and device | |
| CN102799610A (en) | Method and system for collecting network information | |
| CN103886090A (en) | Content recommendation method and device based on user favorites | |
| CN105245583A (en) | Promotion information pushing method and device | |
| CN103218355A (en) | Method and device for generating tags for user | |
| WO2014180130A1 (en) | Method and system for recommending contents | |
| CN107885783B (en) | Method and device for obtaining high-correlation classification of search terms | |
| CN105005582A (en) | Recommendation method and device for multimedia information | |
| US11423096B2 (en) | Method and apparatus for outputting information | |
| US20140331142A1 (en) | Method and system for recommending contents | |
| CN104423621A (en) | Pinyin string processing method and device | |
| CN104778208A (en) | Method and system for optimally grasping search engine SEO (search engine optimization) website data | |
| CN103997662A (en) | Program pushing method and system | |
| CN110866040A (en) | User portrait generation method, device and system | |
| CN105681089B (en) | Networks congestion control clustering method, device and terminal | |
| US9756122B2 (en) | Using hierarchical reservoir sampling to compute percentiles at scale |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160127 |