CN104216886A - Video recommendation device, system and method - Google Patents
Video recommendation device, system and method Download PDFInfo
- Publication number
- CN104216886A CN104216886A CN201310206954.3A CN201310206954A CN104216886A CN 104216886 A CN104216886 A CN 104216886A CN 201310206954 A CN201310206954 A CN 201310206954A CN 104216886 A CN104216886 A CN 104216886A
- Authority
- CN
- China
- Prior art keywords
- video
- recommendation
- unit
- click
- videos
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Library & Information Science (AREA)
- Multimedia (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a video recommendation device, system and method. The device comprises a related video database unit, a related recommendation engine unit and a click rate model unit. The related recommendation engine unit is used for receiving a request from an external system front end, extracting a recommendation video list of videos to be recommended from the related video database unit, and returning the recommended videos to the external system front end according to the display weight of the obtained video list. The related video database unit is a database used for storing the display results of the related videos. The click rate model unit is used for receiving display information and click information which are sent by the external system front end and renewing the display weight in the related video database unit. The display information and the click information which are from an external user are received in real time, feedback is conducted, the related videos are recommended to the user, the display weight in the related video database unit is renewed, and therefore the recommended videos have higher relevancy, and the videos irrelevant to the video which is played at present are removed.
Description
Technical field
The present invention relates to network Online Video technology, particularly relate to a kind of video recommendations device, system and method.
Background technology
Along with the development of infotech and internet, people have entered into the epoch of information overload from the epoch of absence of information gradually.For the consumer of information, from magnanimity information, oneself interested information is found to be a very difficult thing; Meanwhile, for information producer, the information allowing oneself produce is shown one's talent, and is subject to users and pays close attention to, and is also the very difficult thing of part.Current commending system produces to address this problem just.
The associated recommendation of video website is generally find out relevant video according to the video of current broadcasting by methods such as text matches.But these class methods also may recommend out some uncorrelated or be not the associated video meeting most user's request, is such as " video of apple cultivation " when what play, and what recommended by relevant recommend method is the video that i Phone is relevant.Such as Fig. 1 is the recommendation results that the commending system one of prior art is relevant to " lecture of preparing for the postgraduate qualifying examination ", and its display associated recommendation video only has two to be correlated with, other two then with prepare for the postgraduate qualifying examination or lecture completely uncorrelated.
Therefore in commending system, how to improve recommendation video out and have more correlativity, remove and the incoherent video of current broadcasting, become video recommendation system problem demanding prompt solution.
Summary of the invention
The object of the present invention is to provide a kind of video recommendations device, system and method, according to click or the navigation patterns of active user, carry out Real-time Feedback, recommending relevant videos.
For solving the problem, the invention provides a kind of video recommendations device, comprising associated video data library unit, associated recommendation engine unit and clicking rate model unit, wherein,
Associated recommendation engine unit is in order to receive the request from external system front end, from associated video data library unit, take out the recommendation list of videos of video to be recommended, represent weight according to the list of videos obtained, the video of recommendation is back to external system front end;
Associated video data library unit is the database that storage associated video represents result;
Clicking rate model unit in order to receive the presenting information and click information that send over from outer front end, and upgrades and represents weight in associated video data library unit.
Further, described presenting information comprises video ID to be recommended, and the ID of the list of videos of recommending;
Described click information, comprises the video of current click, and the recommendation video clicked.
Further, what upgrade video in associated video data library unit represents weight based on following formulae discovery:
Represent weight=pv/click+X/100
Wherein, pv, click are respectively this recommendation video as the amount of representing and the clicking rate of recommending video, and X is constant.
Further, the video each to be recommended in associated video data library unit, what comprise this video represents weight.
The present invention also provides a kind of video recommendation system, comprise fore device and above-mentioned arbitrary described video recommendations device, wherein, fore device receives associated recommendation engine unit and sends request, and the presenting information of associated recommendation engine unit transmission video and click information are sent to described clicking rate model unit.
In addition, the present invention also provides a kind of video recommendation method, comprises step as follows:
Fore device sends request to associated recommendation engine unit;
The request of associated recommendation engine unit receiving front-end device, from associated video data library unit, takes out the recommendation list of videos of video to be recommended, represents weight according to the list of videos obtained, and the video of recommendation is back to fore device;
Fore device sends presenting information and the click information extremely described clicking rate model unit of video.
Hit presenting information and click information that rate model unit receiving front-end device sends over, and upgrade video in associated video data library unit represent weight.
Further, described presenting information comprises video ID to be recommended, and the ID of the list of videos of recommending;
Described click information, comprises the video of current click, and the recommendation video clicked.
Further, what upgrade video in associated video data library unit represents weight based on following formulae discovery:
Represent weight=pv/click+X/100
Wherein, pv, click are respectively this recommendation video as the amount of representing and the clicking rate of recommending video, and X is constant.
Compared with prior art, video recommendations device of the present invention, system and method, the information being represented from external user by real-time reception and click, feed back, and then sort result is represented to associated recommendation be optimized, take out the recommendation list of videos of video to be recommended, weight is represented according to the list of videos obtained, by relevant video recommendations to user, and upgrade and represent weight in associated video data library unit, make the video recommending out have more correlativity, remove and the incoherent video of current broadcasting.
Accompanying drawing explanation
Fig. 1 is the schematic diagram that existing video recommendations technology recommends to obtain;
Fig. 2 is the video recommendations device schematic diagram of the embodiment of the present invention one;
Fig. 3 is the video recommendation system schematic diagram of the embodiment of the present invention two.
Embodiment
For enabling above-mentioned purpose of the present invention, feature and advantage become apparent more, and below in conjunction with the drawings and specific embodiments, the present invention is further detailed explanation.
Embodiment one
Refer to Fig. 2, Fig. 2 is the video recommendations device schematic diagram of the present embodiment, and video recommendations device 1 comprises associated video data library unit 10, associated recommendation engine unit 11 and clicking rate model unit 12.
Associated recommendation engine unit 11 is in order to receive the request from external system front end, from associated video data library unit 11, take out the recommendation list of videos of video to be recommended, represent weight according to the list of videos obtained, the video of recommendation is back to external system front end.Such as, suppose that the recommendation list of certain video A has B video and C video, the weight of B video is the weight of 1.2, C video is 1, the probability that represents then calculating B is 1.2/ (1+1.2)=55%, and the probability that represents calculating C is 1/ (1+1.2)=45%.
Associated video data library unit 10 is the database that storage associated video represents result, and the video each to be recommended in associated video data library unit, what comprise this video represents weight.
Clicking rate model unit 12 in order to receive the presenting information and click information that send over from outer front end, and upgrades and represents weight in associated video data library unit.Represent weight based on following formulae discovery:
Represent weight=pv/click+X/100
Wherein, pv, click are respectively this recommendation video as the amount of representing and the clicking rate of recommending video, and X is constant.
Described presenting information comprises video ID to be recommended, and the ID of the list of videos of recommending; Described click information, comprises the video of current click, and the recommendation video clicked.
Lower mask body is described in detail for the associated recommendation of a video.
Associated recommendation is carried out to the video that certain video ID is 1, needs to return 1 recommendation results.In the present embodiment, if X=0.5, initialization, obtains the candidate pool (the recommendation list of videos namely in associated recommendation engine unit) of recommendation results by text or other algorithms, and suppose that being is 10 videos, these 10 video ID are 2 to 11.
At the beginning, the weight of each video is 0.5 and is stored in associated video data library unit 10.
When certain user have accessed video ID1, because the weight of each video to be recommended is equal, so the probability that represents of each video is 10%, suppose to return video ID3 specifically, after external system front end obtains information, video ID3 is presented to user, and notifies clicking rate model unit 12.The weight of video ID3 is changed to 0/1+0.5/100=0.005 by clicking rate model unit 12.
If user clicks this video ID3, then external system front end notice clicking rate model unit 12, the weight of video ID3 is changed to 1/1+0.5/100=1.005 by clicking rate model unit 13.
If there is new user to ask the associated recommendation of this video ID1, now the weight of this video ID1 is higher than other videos, it represents probability is 1.005/ (1.005+0.5*9)=18%, and the probability that represents of other videos is 0.5/ (1.005+0.5*9)=9%.In this way, the video that clicking rate is higher can obtain and more represent chance, and lower video then represents chance and can tail off.
Embodiment two
Refer to Fig. 3, based on the video recommendations device of above-described embodiment one, embodiment adds fore device 2, the video recommendation system that fore device 2 is formed with the video recommendations device 1 of embodiment one, in fore device 2 receiver, video recommendation apparatus 1, associated recommendation engine unit 11 sends request, and associated recommendation engine unit 11 is sent the presenting information of video and click information is sent to described clicking rate model unit 12.In the video recommendation system of the present embodiment each functional unit and the course of work and embodiment one describe video recommendations device similar, specifically refer to embodiment one, be not described in detail in this.
Embodiment three
The present invention also provides a kind of video recommendation method, comprises step as follows:
Fore device 2 sends request to associated recommendation engine unit 11;
The request of associated recommendation engine unit 11 receiving front-end device 2, from associated video data library unit 10, takes out the recommendation list of videos of video to be recommended, represents weight according to the list of videos obtained, and the video of recommendation is back to fore device 2;
Fore device 2 sends presenting information and the click information extremely described clicking rate model unit 12 of video.
The presenting information that clicking rate model unit 12 receiving front-end device 2 sends over and click information, and upgrade video in associated video data library unit 10 represent weight.
The presenting information that fore device 2 sends video comprises video ID to be recommended, and the ID of the list of videos of recommending;
Fore device 2 sends the click information of video, comprises the video of current click, and the recommendation video clicked.
What upgrade video in associated video data library unit represents weight based on following formulae discovery:
Represent weight=pv/click+X/100
Wherein, pv, click are respectively this recommendation video as the amount of representing and the clicking rate of recommending video, and X is constant.
The video recommendations device of above embodiment, system and method, the information being represented from external user by real-time reception and click, feed back, and then sort result is represented to associated recommendation be optimized, take out the recommendation list of videos of video to be recommended, weight is represented according to the list of videos obtained, by relevant video recommendations to user, and upgrade and represent weight in associated video data library unit, the video recommending out is made to have more correlativity, remove and the incoherent video of current broadcasting, and and then the clicking rate of raising user.
In this instructions, each embodiment adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar portion mutually see.For system disclosed in embodiment, owing to corresponding to the method disclosed in Example, so description is fairly simple, relevant part illustrates see method part.
Professional can also recognize further, in conjunction with unit and the algorithm steps of each example of embodiment disclosed herein description, can realize with electronic hardware, computer software or the combination of the two, in order to the interchangeability of hardware and software is clearly described, generally describe composition and the step of each example in the above description according to function.These functions perform with hardware or software mode actually, depend on application-specific and the design constraint of technical scheme.Professional and technical personnel can use distinct methods to realize described function to each specifically should being used for, but this realization should not thought and exceeds scope of the present invention.
Obviously, those skilled in the art can carry out various change and modification to invention and not depart from the spirit and scope of the present invention.Like this, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.
Claims (8)
1. a video recommendations device, is characterized in that, comprises associated video data library unit, associated recommendation engine unit and clicking rate model unit, wherein,
Associated recommendation engine unit is in order to receive the request from external system front end, from associated video data library unit, take out the recommendation list of videos of video to be recommended, represent weight according to the list of videos obtained, the video of recommendation is back to external system front end;
Associated video data library unit is the database that storage associated video represents result;
Clicking rate model unit in order to receive the presenting information and click information that send over from outer front end, and upgrades and represents weight in associated video data library unit.
2. video recommendations device as claimed in claim 1, it is characterized in that, described presenting information comprises video ID to be recommended, and the ID of the list of videos of recommending;
Described click information, comprises the video of current click, and the recommendation video clicked.
3. video recommendations device as claimed in claim 1, is characterized in that, what upgrade video in associated video data library unit represents weight based on following formulae discovery:
Represent weight=pv/click+X/100
Wherein, pv, click are respectively this recommendation video as the amount of representing and the clicking rate of recommending video, and X is constant.
4. video recommendations device as claimed in claim 1, it is characterized in that, the video each to be recommended in associated video data library unit, what comprise this video represents weight.
5. a video recommendation system, is characterized in that, comprise fore device and as arbitrary in Claims 1-4 as described in video recommendations device, wherein,
Fore device receives associated recommendation engine unit and sends request, and the presenting information of associated recommendation engine unit transmission video and click information are sent to described clicking rate model unit.
6. a video recommendation method, is characterized in that, comprises step as follows:
Fore device sends request to associated recommendation engine unit;
The request of associated recommendation engine unit receiving front-end device, from associated video data library unit, takes out the recommendation list of videos of video to be recommended, represents weight according to the list of videos obtained, and the video of recommendation is back to fore device;
Fore device sends presenting information and the click information extremely described clicking rate model unit of video.
Hit presenting information and click information that rate model unit receiving front-end device sends over, and upgrade video in associated video data library unit represent weight.
7. video recommendation method as claimed in claim 6, it is characterized in that, described presenting information comprises video ID to be recommended, and the ID of the list of videos of recommending;
Described click information, comprises the video of current click, and the recommendation video clicked.
8. video recommendation method as claimed in claim 6, is characterized in that, what upgrade video in associated video data library unit represents weight based on following formulae discovery:
Represent weight=pv/click+X/100
Wherein, pv, click are respectively this recommendation video as the amount of representing and the clicking rate of recommending video, and X is constant.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310206954.3A CN104216886A (en) | 2013-05-29 | 2013-05-29 | Video recommendation device, system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310206954.3A CN104216886A (en) | 2013-05-29 | 2013-05-29 | Video recommendation device, system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104216886A true CN104216886A (en) | 2014-12-17 |
Family
ID=52098392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310206954.3A Pending CN104216886A (en) | 2013-05-29 | 2013-05-29 | Video recommendation device, system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104216886A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104506895A (en) * | 2014-12-22 | 2015-04-08 | 合一网络技术(北京)有限公司 | Video classification method |
| CN105163141A (en) * | 2015-08-31 | 2015-12-16 | 小米科技有限责任公司 | Video recommending method and device |
| CN105516738A (en) * | 2015-12-04 | 2016-04-20 | 青岛海信传媒网络技术有限公司 | Video-on-demand processing method, device and equipment |
| CN106250499A (en) * | 2016-08-02 | 2016-12-21 | 合网络技术（北京）有限公司 | A kind of video is to method for digging and device |
| CN108614856A (en) * | 2018-03-21 | 2018-10-02 | 北京奇艺世纪科技有限公司 | A kind of video sequence calibration method and device |
| CN108710635A (en) * | 2018-04-08 | 2018-10-26 | 达而观信息科技（上海）有限公司 | A kind of content recommendation method and device |
| CN108810056A (en) * | 2017-05-04 | 2018-11-13 | 腾讯科技（北京）有限公司 | Information-pushing method and device |
| CN108989397A (en) * | 2018-06-26 | 2018-12-11 | 腾讯音乐娱乐科技（深圳）有限公司 | Data recommendation method, device and storage medium |
| CN109246451A (en) * | 2018-08-23 | 2019-01-18 | 武汉斗鱼网络科技有限公司 | A kind of direct broadcasting room recommended method, device, server and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102685565A (en) * | 2012-05-18 | 2012-09-19 | 合一网络技术(北京)有限公司 | Click feedback type individual recommendation system |
-
2013
- 2013-05-29 CN CN201310206954.3A patent/CN104216886A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102685565A (en) * | 2012-05-18 | 2012-09-19 | 合一网络技术(北京)有限公司 | Click feedback type individual recommendation system |
Non-Patent Citations (2)
| Title |
|---|
| 欧阳友权: "《网络文学词典》", 30 April 2013, 世界图书出版广东有限公司 * |
| 苏青章: "社会网络多策略视频推荐系统研究与设计", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104506895A (en) * | 2014-12-22 | 2015-04-08 | 合一网络技术(北京)有限公司 | Video classification method |
| CN104506895B (en) * | 2014-12-22 | 2017-12-01 | 合一网络技术(北京)有限公司 | Video classifying method |
| CN105163141B (en) * | 2015-08-31 | 2018-12-25 | 小米科技有限责任公司 | The mode and device of video recommendations |
| CN105163141A (en) * | 2015-08-31 | 2015-12-16 | 小米科技有限责任公司 | Video recommending method and device |
| CN105516738A (en) * | 2015-12-04 | 2016-04-20 | 青岛海信传媒网络技术有限公司 | Video-on-demand processing method, device and equipment |
| CN106250499A (en) * | 2016-08-02 | 2016-12-21 | 合网络技术（北京）有限公司 | A kind of video is to method for digging and device |
| CN106250499B (en) * | 2016-08-02 | 2020-07-14 | 阿里巴巴（中国）有限公司 | Video pair mining method and device |
| CN108810056A (en) * | 2017-05-04 | 2018-11-13 | 腾讯科技（北京）有限公司 | Information-pushing method and device |
| CN108614856A (en) * | 2018-03-21 | 2018-10-02 | 北京奇艺世纪科技有限公司 | A kind of video sequence calibration method and device |
| CN108710635A (en) * | 2018-04-08 | 2018-10-26 | 达而观信息科技（上海）有限公司 | A kind of content recommendation method and device |
| CN108710635B (en) * | 2018-04-08 | 2022-03-18 | 达而观信息科技（上海）有限公司 | Content recommendation method and device |
| CN108989397A (en) * | 2018-06-26 | 2018-12-11 | 腾讯音乐娱乐科技（深圳）有限公司 | Data recommendation method, device and storage medium |
| CN108989397B (en) * | 2018-06-26 | 2021-04-20 | 腾讯音乐娱乐科技（深圳）有限公司 | Data recommendation method and device and storage medium |
| CN109246451A (en) * | 2018-08-23 | 2019-01-18 | 武汉斗鱼网络科技有限公司 | A kind of direct broadcasting room recommended method, device, server and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104216886A (en) | Video recommendation device, system and method | |
| CN108108821B (en) | Model training method and device | |
| CN105335519B (en) | Model generation method and device and recommendation method and device | |
| US10180967B2 (en) | Performing application searches | |
| CN107515909B (en) | Video recommendation method and system | |
| US10332184B2 (en) | Personalized application recommendations | |
| US10825047B2 (en) | Apparatus and method of selection and placement of targeted messages into a search engine result page | |
| US8996520B2 (en) | Ranking of native application content | |
| US20180060915A1 (en) | Audience expansion for online social network content | |
| CN104537000B (en) | A kind of method and apparatus for pushed information | |
| US20120066196A1 (en) | Device for determining internet activity | |
| TW201717071A (en) | Recommendation method and device | |
| KR20160058895A (en) | System and method for analyzing and synthesizing social communication data | |
| US20170140056A1 (en) | System and method for generating influencer scores | |
| CN102332006A (en) | Information push control method and device | |
| CN101360098A (en) | Customer behavior attribute determining method, apparatus, system and advertisement releasing method and system | |
| US9794284B2 (en) | Application spam detector | |
| US11321334B2 (en) | Multi-dimensional search | |
| US11290551B1 (en) | Method and system for implementing circle of trust in a social network | |
| WO2016099447A1 (en) | Personalized application recommendations | |
| WO2015005950A1 (en) | Determining whether to send a call-out to a bidder in an online content auction | |
| CN104915359A (en) | Theme label recommending method and device | |
| US11108717B1 (en) | Trends in a messaging platform | |
| CN104731788B (en) | The processing method and equipment of promotion message | |
| US20220005070A1 (en) | Generating dynamic content item recommendations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180608 Address after: 201203 7, 1 Lane 666 lane, Zhang Heng Road, Pudong New Area, Shanghai. Applicant after: SHANGHAI ZHANGMEN TECHNOLOGY CO., LTD. Address before: 300467 Tianjin Binhai New Area Tianjin eco city animation road 126 anime building B1 area two layer 201-243 Applicant before: Cool Sheng (Tianjin) Technology Co., Ltd. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180806 Address after: 300450 Tianjin eco city anime road 126 anime building B1 area two layer 201-243 Applicant after: Cool Sheng (Tianjin) Technology Co., Ltd. Address before: 201203 7, 1 Lane 666 lane, Zhang Heng Road, Pudong New Area, Shanghai. Applicant before: SHANGHAI ZHANGMEN TECHNOLOGY CO., LTD. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20181219 Address after: 201306 N2025 room 24, 2 New Town Road, mud town, Pudong New Area, Shanghai Applicant after: Shanghai Lian Shang network technology Co., Ltd Address before: 300450 Tianjin eco city anime road 126 anime building B1 area two layer 201-243 Applicant before: Cool Sheng (Tianjin) Technology Co., Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141217 |