CN101860696B - Program recommending method and streaming media program system - Google Patents
Program recommending method and streaming media program system Download PDFInfo
- Publication number
- CN101860696B CN101860696B CN 201010160136 CN201010160136A CN101860696B CN 101860696 B CN101860696 B CN 101860696B CN 201010160136 CN201010160136 CN 201010160136 CN 201010160136 A CN201010160136 A CN 201010160136A CN 101860696 B CN101860696 B CN 101860696B
- Authority
- CN
- China
- Prior art keywords
- user
- program
- attribute
- commending
- program commending
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The invention discloses a program recommending method and a streaming media program system, wherein the method comprises the following steps that: the attribute weight of a program is generated according to the program playing behaviors of a user; and the program is recommended to the user according to the attribute weight of the program. The streaming media business system comprises a business management platform and a database, the business management platform comprises a user playing behavior analysis module, the user playing behavior analysis module is connected with the database, generates the attribute weight of the program according to the program playing behavior of the user, and recommends the program to the user according to the attribute weight of the program. Compared with the prior art, the invention calculates the preference of the user to programs by analyzing the program playing behaviors of the user according to the program attribute, recommends programs to the user and improves the recommendation precision.
Description
Technical field
The invention belongs to the streaming media service technical field, be specifically related to a kind of program commending method and flow-medium performance system.
Background technology
All fairly simple to the way of recommendation of content in present most audio frequency and video operation system, the taste according to the user does not carry out the individual character recommendation.The for example simple recommendation in IPTV and mobile stream media system.The existing way of recommendation is broadly divided into two kinds: a kind of is that the number of times played according to program produces temperature, has done a ranking list, and its limitation is: Consideration is single, adopts often automatically to add 1 pattern, does not have intelligent.Another kind of is to adopt based on program provider program subjective judgement method, mainly depends on the program novel degrees and has watched the user to the history evaluation analysis of program, excavates the user to program interest, and according to the pattern of discovery, the user is carried out program commending.Its limitation is: modeling is not carried out in user's broadcast behavior, lacked the effective processing to user preference, the precision of recommendation is not high.
Summary of the invention
The purpose of this invention is to provide a kind of program commending method and flow-medium performance system, improved the accuracy of program commending.
For achieving the above object, the present invention has adopted following technical scheme: a kind of program commending method comprises step:
Generate the attribute weight value of program according to user's program broadcast behavior;
Programs recommended to the user according to the attribute weight value of described program.
Preferably, the attribute weight value that generates program according to user's program broadcast behavior comprises step:
Set up this user's program commending table according to user's broadcast behavior, described program commending table is arranged with the attribute weight value of program;
Then the summation of the attribute of all programs that counting user is play calculates the mean value of each weight in user's program commending table.
Preferably, in described calculating user program commending table, the mean value of each weight comprises step: when the user withdraws from performance system, prompting user feedback program commending degree of being satisfied with according to the satisfaction of user feedback, calculates the mean value of each weight in user's program commending table.
Preferably, according to the satisfaction of user feedback, the mean value that calculates each weight in user's program commending table is specially: if the satisfaction of user feedback is on the low side, increase weighted mean; If the satisfaction of user feedback is for just in time, weighted mean is constant; If the satisfaction of user feedback is higher, reduce weighted mean.
Preferably, by the start by set date task is set, the start by set date system automatically recomputate with update system in user's program commending table.
Preferably, described method also comprises step: search is play the similar user of behavior with this user, selects from user's program commending table of similar users in the programs recommended program commending table that adds this user.
A kind of stream media service system, comprise Business Management Platform and database, described Business Management Platform comprises that the user plays the behavioural analysis module, described user plays the behavioural analysis module and is connected with database, described user plays the behavioural analysis module and generates the attribute weight value of program according to user's program broadcast behavior, and programs recommended to the user according to the attribute weight value of described program.
Preferably, described user behavior analysis module is set up this user's program commending table according to user's broadcast behavior, and described program commending table is arranged with the attribute weight value of program; And the summation of the wherein a certain attribute of all programs of playing of counting user, then calculate the mean value of each weight in user's program commending table.
Preferably, described Business Management Platform also is used for prompting user feedback program commending degree of being satisfied with when the user withdraws from performance system, if the satisfaction of user feedback is on the low side, increase weighted mean; If the satisfaction of user feedback is for just in time, the mean value of weight is constant; If the satisfaction of user feedback is higher, reduce the mean value of weight.
Preferably, described Business Management Platform also comprises time block, is connected with the user behavior analysis module, is used for the start by set date task, the start by set date system automatically recomputate with update system in user's program commending table.
Preferably, described Business Management Platform also comprises search module, and described search module is used for search and this user behavior is play the similar user of behavior, selects from user's program commending table of similar users in the programs recommended program commending table that adds this user.
Compared with prior art, the present invention plays behavior and programme attribute by analysis user to program, calculates the user to the preference of program, and is programs recommended to the user, improved the accuracy of recommending.
Description of drawings
A kind of program commending method flow chart that Fig. 1 provides for the embodiment of the present invention;
A kind of stream media service system block diagram that Fig. 2 provides for the embodiment of the present invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, describe the present invention below in conjunction with the accompanying drawings and the specific embodiments.
Core concept of the present invention is: set up Mathematical Modeling according to user's program broadcast behavior, design program intelligent recommendation system, play history record and program characteristics by analysis user, identification user's preference, recommend satisfied program to the user, improve the accuracy of recommending, have intelligent preferably.
(1) increasing programme attribute (asking the inventor to illustrate which attribute programme attribute specifically comprises) in the uploading step of operator's management platform sets, hypothetical program is m, to enter library information be that (Attribute 1 for Gm=to a program, Attribute2, Attribue3, ...), wherein Attribute is the weighted value of corresponding attribute, system increases a programme attribute weight table.Described attribute refers to the type that film comprises, such as action, love, science fiction, terror etc.
(2) increase a program hot statistics table (generally all existing in existing mobile flow medium and IPTV) in system, when user's certain program that finishes playing, this program temperature value adds 1 automatically, suppose that the temperature information definition is HOT=(Content_m, Hot_value, Date), wherein Content_m is programm name, Hot_value is the temperature value, and Date is Time of Day.
See also shown in Figure 1ly, Fig. 1 is a kind of program commending method flow chart that the embodiment of the present invention provides, and comprises the following steps:
S101: when certain user logs in stream media service system, judge this user whether for logging in first, if login user first enters step S102; If not login user first, enter step S104.
S102: directly recommend several the highest (such as 10) programs of present temperature to the user, change step S104 over to.
S103: get this user's program commending table from user's program commending table of database, select to play for the user.
S104: temperature adds a weight unit automatically when user's broadcast program finishes.
Suppose user's program commending table one recorded information, wherein W represents total weight of program attribute, and AllValue represents the total broadcasting time of user, and Content represents programs recommended, Rm=(W1, W2, W3...AllValue, Content1, Content2 ...).
S105: with user's broadcast behavior warehouse-in, suppose that into library information be Action=(UserID, Content, Date), before warehouse-in, this user history information is all emptied.
S106: after the user behavior warehouse-in, the user was upgraded the timed task warehouse-in of recommending the same day, whether at first inquire about this user recommends to upgrade the timed task data and put in storage the same day, again put in storage if put in storage do not needed, otherwise with the timed task warehouse-in, suppose that into library information be Task=(Userid, Date), wherein Userid is user ID, and Date is Time of Day.When other programs are play in user's broadcast again, jump to step S105.
S107: when the user withdraws from whole stream media service system, the prompting user feedback program commending situation of being satisfied with.Preferably, information: program weight n, select: higher, on the low side, just in time, with field feedback access customer feedback data table FB=(UserID, W1, W2, ...), three kinds of state value-1 of W definable, 0,1 wherein, represent respectively on the low side, just in time, higher.
S108-S110: commending system is by timed task, such as in every day when morning, whole traffic carrying capacity was little, start calculating and the renewal of user's program commending table.Traversal user on the same day recommends to upgrade the timed task table, the user when system service in the sky take out one by one calculate delete after the user behavior value Task show in this record.
In the present embodiment, rudimentary algorithm is:
Wherein i represents the programme content sequence valve that the user play the same day, and the n initial value is 1, is broadcasting program number purpose accumulated value in user one day.For example: play 10 films the same day, n=9 so, the value of i is [1,9], and Attribute represents the attribute of film, and Atrribute_Wj represents the weighted value summation of j attribute in all films that the user play the same day., this user's weighted value (W value and AllValue) in renewal user's program commending table (Rm table), basic calculation: use the value of calculating to upgrade Rm literary name section Rm.Wj=Rm.Wj+Atrribute_Wj, Rm.n=Rm.n+1.
Further, the average weight Average (Wi) of each weights W=Rm.Wi/n in calculating Rm table, data and average weight data Average (Wi) in COMPREHENSIVE CALCULATING feedback form (FB).
Preferably, in the present embodiment, if the satisfaction of user feedback is on the low side, increase corresponding weighted mean; If the satisfaction of user feedback is for just in time, corresponding weighted mean is constant; If the satisfaction of user feedback is higher, reduce corresponding weighted mean.Such as if the W value is-1, Average (Wi) * 120%, and the full marks value that exceeds Average (Wi) is calculated with full marks; If be 0, Average (Wi) is constant; If be 1, Average (Wi) * 80%.
S111-S112: start immediate several (preferred 8) programs (called after C8) of final average weight value that search engine search content table (Gm) weight comprehensive grading and user calculate.
Further, search out and immediate several users of this user weight, preferred 2, get up-to-date broadcast in this 2 user Action table and be different from each 1 of the program of C8 in the GM table, have 10 programs altogether, with the Content1...Content10 in these 10 program updating Rm tables.
Repeating step S108-S112 completes until the data in timing updating task table are all deleted.
See also shown in Figure 2ly, Fig. 2 is a kind of stream media service system block diagram that the embodiment of the present invention provides.Stream media service system comprises Business Management Platform 21 and database 22, and described Business Management Platform 21 comprises that the user plays behavioural analysis module 210, time block 211 and search module 212.
Described user plays behavioural analysis module 210 and is connected with database 22, and described user plays the behavioural analysis module and generates the attribute weight value of program according to user's program broadcast behavior, and programs recommended to the user according to the attribute weight value of described program.
Preferably, described user behavior analysis module 210 is set up this user's program commending table according to user's broadcast behavior, and described program commending table is arranged with the attribute weight value of program; And the summation of the wherein a certain attribute of all programs of playing of counting user, then calculate the mean value of each weight in user's program commending table.
Preferably, described Business Management Platform 21 also is used for prompting user feedback program commending degree of being satisfied with when the user withdraws from performance system, and the user has three kinds of selections: on the low side, just in time, higher.
Preferably, described user behavior analysis module 210 is according to the satisfaction of user feedback, and the average weight value of calculating each weight in user's program commending table is specially: if the satisfaction of user feedback is on the low side, increase weighted mean; If the satisfaction of user feedback is for just in time, weighted mean is constant; If the satisfaction of user feedback is higher, reduce weighted mean.
Search module 213 is used for search and plays similar program and the user of behavior with user behavior, and the behavior table of comprehensive similar program and similar users is selected programs recommended this user's of entering program commending table.
Preferably, described Business Management Platform 21 also is used for prompting user feedback program commending degree of being satisfied with when the user withdraws from performance system.
Describe as an example of the media business such as existing IPTV, mobile flow medium and P2P system example, in the existing business system by EPG/ portal page module to user's presenting programs recommendation tables.When operator logged in business management system upload of media file, the business management system module entered database with data.By EPG/ portal page module, data are entered database when user's broadcast or download behavior or feedback recommendation end.Commending system increases timed task module, user behavior analysis module, feedback adjusting module and search engine module newly.Wherein by the timed task module when whole operation system is not busy (for example morning, 2:00 was to 6:000 in morning) trigger user's program commending table system, first trigger the behavior on user behavior analysis module analysis user same day and the data data writing (feedback information is adjusted through the feedback adjusting module) that feedback will be analyzed this moment, after completing, the startup search engine search program relevant with user behavior enters program commending table database.Wherein the business management system module increases the programme attribute setting in uploading step, hypothetical program is m, to enter library information be Gm=(Attribute1 to a program, Attribute2, Attribue3, ...), wherein Attribute is the weighted value of corresponding attribute, system increases a programme attribute weight table.Increase in addition a program hot statistics table (generally all existing in existing mobile flow medium and IPTV) in database, when user's certain program that finishes playing, by 103 modules, program temperature value is added 1 automatically, suppose that the temperature information definition is HOT=(Content_m, Hot_value, Date).
Program is play behavior to User of the present invention and programme attribute has been set up a data model, and define respectively the specific implementation in each stage, data analysis principle and Mathematical Modeling have been fully utilized, by analysis user, program is play behavior and programme attribute, calculate the user to the preference of program, programs recommended to the user, improve and recommend efficient.Further, system can adjust automatically according to user's feedback information, thereby improves the accuracy of recommending, and has intelligent preferably.
The present invention is applicable to any business operation system relevant to media services, by the further user preferences recommend method of setting forth in a kind of stream media service system of the present invention of following specific embodiment, below embodiment is described in detail, but not as a limitation of the invention.
Above content is in conjunction with concrete execution mode further description made for the present invention, can not assert that concrete enforcement of the present invention is confined to these explanations.For the general technical staff of the technical field of the invention, without departing from the inventive concept of the premise, can also make some simple deduction or replace, all should be considered as belonging to protection scope of the present invention.
Claims (9)
1. a program commending method, is characterized in that, comprises step:
Generate the attribute weight value of program according to user's program broadcast behavior, comprising: set up this user's program commending table according to user's broadcast behavior, described program commending table is arranged with the attribute weight value of program;
Then the summation of the attribute of all programs that counting user is play calculates the mean value of each weight in user's program commending table, and described attribute is the type that film comprises;
Programs recommended to the user according to the attribute weight value of described program.
2. method according to claim 1, it is characterized in that, in described calculating user program commending table, the mean value of each weight comprises step: when the user withdraws from performance system, prompting user feedback program commending degree of being satisfied with, according to the satisfaction of user feedback, calculate the mean value of each weight in user's program commending table.
3. method according to claim 2, is characterized in that, according to the satisfaction of user feedback, the mean value that calculates each weight in user's program commending table is specially: if the satisfaction of user feedback is on the low side, increase weighted mean; If the satisfaction of user feedback is for just in time, weighted mean is constant; If the satisfaction of user feedback is higher, reduce weighted mean.
4. the method described according to claim 1 to 3 any one, is characterized in that, by the start by set date task is set, the start by set date system automatically recomputate with update system in user's program commending table.
5. the method described according to claim 1 to 3 any one, is characterized in that, also comprises step: search is play the similar user of behavior with this user, selects from user's program commending table of similar users in the programs recommended program commending table that adds this user.
6. stream media service system, comprise Business Management Platform and database, it is characterized in that, described Business Management Platform comprises that the user plays the behavioural analysis module, described user plays the behavioural analysis module and is connected with database, described user plays the behavioural analysis module and generates the attribute weight value of program according to user's program broadcast behavior, and programs recommended to the user according to the attribute weight value of described program;
Described user plays the behavioural analysis module and comprises according to the attribute weight value that user's program broadcast behavior generates program:
Set up this user's program commending table according to user's broadcast behavior, described program commending table is arranged with the attribute weight value of program; And the summation of the attribute of all programs of playing of counting user, then calculating the mean value of each weight in user's program commending table, described attribute is the type that film comprises.
7. system according to claim 6, is characterized in that, described Business Management Platform also is used for prompting user feedback program commending degree of being satisfied with when the user withdraws from performance system, if the satisfaction of user feedback is on the low side, increase weighted mean; If the satisfaction of user feedback is for just in time, the mean value of weight is constant; If the satisfaction of user feedback is higher, reduce the mean value of weight.
8. according to claim 6 or 7 any one described systems, it is characterized in that, described Business Management Platform also comprises time block, is connected with the user behavior analysis module, be used for the start by set date task, the start by set date system automatically recomputate with update system in user's program commending table.
9. according to claim 6 or 7 any one described systems, it is characterized in that, described Business Management Platform also comprises search module, described search module is used for search and this user behavior is play the similar user of behavior, selects from user's program commending table of similar users in the programs recommended program commending table that adds this user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010160136 CN101860696B (en) | 2010-04-24 | 2010-04-24 | Program recommending method and streaming media program system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010160136 CN101860696B (en) | 2010-04-24 | 2010-04-24 | Program recommending method and streaming media program system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101860696A CN101860696A (en) | 2010-10-13 |
| CN101860696B true CN101860696B (en) | 2013-11-06 |
Family
ID=42946307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010160136 Active CN101860696B (en) | 2010-04-24 | 2010-04-24 | Program recommending method and streaming media program system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101860696B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11902606B2 (en) | 2017-07-27 | 2024-02-13 | Google Llc | Methods, systems, and media for presenting notifications indicating recommended content |
Families Citing this family (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2012186621A (en) * | 2011-03-04 | 2012-09-27 | Sony Corp | Information processing apparatus, information processing method, and program |
| CN102307315B (en) * | 2011-04-22 | 2014-01-29 | 赛特斯信息科技股份有限公司 | User behavior analysis device in Internet protocol television (IPTV) system, and system for realizing analysis application |
| CN103107936B (en) * | 2011-11-11 | 2015-11-11 | 中国移动通信集团上海有限公司 | A kind of method and device sending information |
| CN102769781B (en) * | 2012-07-17 | 2015-01-21 | 青岛海信传媒网络技术有限公司 | Method and device for recommending television program |
| CN102802050B (en) * | 2012-08-24 | 2015-04-01 | 青岛海信电器股份有限公司 | Television program recommendation method and system |
| TWI538491B (en) * | 2013-09-02 | 2016-06-11 | 大同股份有限公司 | Television service system and method for supplying video service |
| CN104519375A (en) * | 2014-11-20 | 2015-04-15 | 无锡悟莘科技有限公司 | Reminding method of set top box |
| CN104506535A (en) * | 2014-12-25 | 2015-04-08 | 深圳中兴网信科技有限公司 | Data pushing method and data pushing system |
| CN104780448A (en) * | 2015-03-18 | 2015-07-15 | 福建新大陆通信科技股份有限公司 | A time-period-based TV serial tracking method and system |
| CN104899265B (en) * | 2015-05-21 | 2018-07-20 | 广东小天才科技有限公司 | Information recommendation method and system |
| CN105430444A (en) * | 2015-11-17 | 2016-03-23 | 深圳创维数字技术有限公司 | Program recommendation method and digital TV receiving terminal |
| WO2017124393A1 (en) * | 2016-01-21 | 2017-07-27 | 阮元 | Information pushing method during user preference-based resource recommendation, and recommendation system |
| WO2017124387A1 (en) * | 2016-01-21 | 2017-07-27 | 阮元 | Information pushing method during time period-based automatic resource recommendation, and recommendation system |
| WO2017128058A1 (en) * | 2016-01-26 | 2017-08-03 | 白宏伟 | Information pushing method for use during automatic video pushing, and pushing system |
| WO2017128059A1 (en) * | 2016-01-26 | 2017-08-03 | 白宏伟 | User preference-based automatic video pushing method, and pushing system |
| WO2017128057A1 (en) * | 2016-01-26 | 2017-08-03 | 白宏伟 | Method for sending technical usage data of video push technology, and push system |
| CN109121007B (en) * | 2018-09-18 | 2020-11-03 | 深圳市酷开网络科技有限公司 | Movie content recommendation method based on multi-face recognition, smart television and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6064980A (en) * | 1998-03-17 | 2000-05-16 | Amazon.Com, Inc. | System and methods for collaborative recommendations |
| CN101163196A (en) * | 2007-11-20 | 2008-04-16 | 腾讯科技（深圳）有限公司 | Method, system and equipment to recommend program to television users |
| CN101287082A (en) * | 2008-05-16 | 2008-10-15 | 华东师范大学 | Collaborative filtered recommendation method introducing hotness degree weight of program |
| CN101383942A (en) * | 2008-08-01 | 2009-03-11 | 深圳市天威视讯股份有限公司 | Hidden customer characteristic extracting method and television program recommendation method and system |
-
2010
- 2010-04-24 CN CN 201010160136 patent/CN101860696B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6064980A (en) * | 1998-03-17 | 2000-05-16 | Amazon.Com, Inc. | System and methods for collaborative recommendations |
| CN101163196A (en) * | 2007-11-20 | 2008-04-16 | 腾讯科技（深圳）有限公司 | Method, system and equipment to recommend program to television users |
| CN101287082A (en) * | 2008-05-16 | 2008-10-15 | 华东师范大学 | Collaborative filtered recommendation method introducing hotness degree weight of program |
| CN101383942A (en) * | 2008-08-01 | 2009-03-11 | 深圳市天威视讯股份有限公司 | Hidden customer characteristic extracting method and television program recommendation method and system |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11902606B2 (en) | 2017-07-27 | 2024-02-13 | Google Llc | Methods, systems, and media for presenting notifications indicating recommended content |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101860696A (en) | 2010-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101860696B (en) | Program recommending method and streaming media program system | |
| CN107071578B (en) | IPTV program commending method | |
| US8620607B2 (en) | Systems and methods for metering and analyzing energy consumption of events within a portable device | |
| US7706740B2 (en) | Apparatus and methods of selective collection and selective presentation of content | |
| US11917243B1 (en) | Optimizing inventory based on predicted viewership | |
| US8515336B2 (en) | Apparatus and methods of selective collection and selective presentation of content | |
| CN104636336B (en) | A kind of method and apparatus of video search | |
| US9166714B2 (en) | Method of and system for presenting enriched video viewing analytics | |
| US7644423B2 (en) | System and method for generating media consumption statistics | |
| US20060015902A1 (en) | Viewing history recording method and viewing history use method | |
| US10025785B2 (en) | Method and system of automatically downloading media content in a preferred network | |
| CN108595492B (en) | Content pushing method and device, storage medium and electronic device | |
| CN103686236A (en) | Method and system for recommending video resource | |
| CN106326431A (en) | Information recommendation method and device | |
| CN103686237A (en) | Method and system for recommending video resource | |
| US20110106584A1 (en) | System and method for measuring customer interest to forecast entity consumption | |
| CN101764661A (en) | Data fusion based video program recommendation system | |
| US8386601B1 (en) | Detecting and reporting on consumption rate changes | |
| US20140259038A1 (en) | Demographic prediction channel | |
| CN102263999A (en) | Face-recognition-based method and system for automatically classifying television programs | |
| CN102265276A (en) | Context-based recommender system | |
| CN101763351A (en) | Data fusion based video program recommendation method | |
| CN102685566A (en) | Recommendation method for audio and video programs | |
| CN101088288A (en) | Method and device for commending contents | |
| CN104598505A (en) | Multimedia resource recommendation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |