CN100581227C - Collaborative filtered recommendation method introducing hotness degree weight of program - Google Patents
Collaborative filtered recommendation method introducing hotness degree weight of program Download PDFInfo
- Publication number
- CN100581227C CN100581227C CN200810037498A CN200810037498A CN100581227C CN 100581227 C CN100581227 C CN 100581227C CN 200810037498 A CN200810037498 A CN 200810037498A CN 200810037498 A CN200810037498 A CN 200810037498A CN 100581227 C CN100581227 C CN 100581227C
- Authority
- CN
- China
- Prior art keywords
- user
- program
- project
- degree weight
- hotness
- 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.)
- Expired - Fee Related
Links
Abstract
The invention discloses a collaborative filtering recommendation method for introducing program popularity weighting, which is characterized in that on the interface of an IPTV program, a visual menu for users to give marks is provided and a program recommendation list is made for target users according to user watching time, conduct operation and program marking data sent by a terminal set-top box. The invention comprises the detailed steps of: collecting the behavior characteristic information of users, working out a 'user-item' scoring matrix A(m, n), calculating popularity weight value, calculating similarity degree and sorting, making forecast score for the target users and sorting, and working out the recommendation list for the target users. Compared with the prior art, the method disclosed by the invention is more in accordance with objective reality, improves the quality of collaborative filtering and the precision degree of recommendation, initiatively cuts own the programs according to user preferences and behavior characteristics, carries out personalized recommendation to the programs which the users like and realizes the purpose that 'watch the program you like whenever you want'.
Description
Technical field
The present invention relates to the IPTV personalized recommendation system, specifically a kind of collaborative filtered recommendation method of introducing hotness degree weight of program.
Background technology
So-called " information overload " and " information is isotropic " phenomenon have appearred in the sharp increase of going up information along with Internet, commending system arises at the historic moment, it can find the resource that is fit to its interest for the user according to information such as user's operation history and feedbacks, for it produces personalized recommendation.Nowadays, recommended technology has been applied in every field such as ecommerce, digital library, video display amusement.Especially IPTV field, continuous development along with the Digital Television and the communication technology, the TV programme resource is more and more abundanter, the user feels very excited for watching so many program on the one hand, on the other hand again for how finding their real favorite program to feel very worried from hundreds and thousands of programs.Collaborative filter techniques is a most successful current personalized recommendation technology, and commending system that some are more famous such as WebWatcher, GroupLens, Firefly, SELECT, LileMinds and Citeseer have adopted the method for collaborative filtering.The score data that basic thought just is based on the similar nearest-neighbors of scoring produces recommendation to the targeted customer, promptly produces recommendation list to the targeted customer according to other user's viewpoint.It is based on such hypothesis: if the user is more similar to the scoring of some projects, then they are also more similar to the scoring of other project.Its starting point is to find the one group user identical with your interest, and term is called " arest neighbors ", and the core of nearest neighbor search is to calculate two users' similarity.For example user A and user B at first need to obtain user A and all scoring items of user B, select a suitable similarity calculating method then, based on scoring item number certificate, calculate the similarity numerical value of user A and user B.Use many similarity algorithms to comprise at present, Pearson's coefficient correlation (PCC), cosine similitude and adjustment cosine similitude.From the above, the committed step of collaborative filtering is to find targeted customer's nearest-neighbors, and can find nearest-neighbors accurately be to recommend whether accurate emphasis, and the similarity of calculating more accurately between the user is that arest neighbors is chosen prerequisite accurately.During but the PCC that uses calculates at present, it is exactly that the project of the common scoring of user is made no exception, do not distinguish the popular degree of project itself, closely just can reflect higher similarity as long as marked jointly and marked, so existing personalized recommendation technology accuracy is relatively poor, is consistent not to the utmost with objective reality.
Summary of the invention
The objective of the invention is a kind of collaborative filtered recommendation method of introducing hotness degree weight of program of designing at the deficiencies in the prior art, it at first defines the popular degree of TV programme, and then calculate its hotness degree weight, and in user's calculating formula of similarity, introduce this weight, the similitude that calculates in view of the above more meets reality, the similitude that calculates is more accurate, therefore can choose targeted customer's nearest-neighbors more accurately, thereby produces more accurate recommendation.
The object of the present invention is achieved like this: a kind of collaborative filtered recommendation method of introducing hotness degree weight of program, characteristics are on the interface of IPTV program, the visual menu that provides the user to mark, and make program commending and tabulate to the targeted customer according to user's viewing time, behavior operation, program score data that terminal set top box transmits, its concrete steps are as follows:
A. collect the user interest data, make " user-project " rating matrix A (m, n);
B. make the hotness degree weight value of off-line computation of Period project;
C. the project that the current active user has been marked finds corresponding hotness degree weight value;
D. make similarity size and ordering between targeted customer a and other user;
E. K the user who chooses the similarity maximum is as its nearest-neighbors collection;
F. according to nearest-neighbors set pair targeted customer not scoring item predict the scoring and the ordering;
G. will predict that the maximum top n project of scoring makes recommendation list and give the targeted customer.
Described " user-project " rating matrix A (m is to carry out arranged with user's score information and user behavior data n), the row representative of consumer, and row representative project, the element value in the matrix is then represented the favorable rating of this row user to this list of items.
The described off-line cycle calculated once by 30 minutes, the hotness degree weight value of project be with
Calculate w
tBe hotness degree weight; P
AllPopular degree summation for all items; P
tPopular degree for the t project;
Similarity between described targeted customer a and other user is incorporated into Pearson correlation coefficient with the hotness degree weight value and calculates.
The present invention compared with prior art has and more meets objective reality, improved the quality of collaborative filtering, recommend more accurate, it is according to user's preference and behavioural characteristic, initiatively program is reduced, the user is wanted that the program of seeing carries out personalized recommendation, realized " seeing that when you want you want the TV of seeing " this target.
Description of drawings
Fig. 1 is a schematic flow sheet of the present invention
Fig. 2 is a project t hotness degree weight calculation process schematic diagram of the present invention
Embodiment
Embodiment
Consult accompanying drawing 1~2, the present invention is on the interface of IPTV program, the visual menu that provides the user to mark, and make program commending and tabulate to the targeted customer according to user's viewing time, behavior operation, program score data that terminal set top box transmits, its concrete steps are as follows:
1, data collection unit features such as the viewing time by following the tracks of the user, behavior operation in the IPTV system are obtained the information of representative of consumer interest, and will be stored in the corresponding database table.
2, above-mentioned user's behavior characteristic information is carried out the processing of initial data by system, and replaces the user to finish evaluation, then according to user's score information and user behavior data, arrangement obtain " user-project " rating matrix A (m, n), the value of scoring from 1 to r
Max(i.e. marking scope be 1-5), this matrix is stored on the recommended engine device as user interest model, the row representative of consumer, row representative project, the element value in the matrix is represented the favorable rating of this row user to this list of items, favorable rating is set to 5 grades, correspond to respectively: (1) is disliked very much, and (2) are relatively disliked, and (3) are general, (4) prefer, (5) are delithted with.If the user did not estimate certain project, in rating matrix, be set to 0 so.
3, to " user-project " rating matrix A (m n) carries out the off-line computation of Period, obtains the hotness degree weight and the storage of each project, and the off-line cycle calculated once by 30 minutes, (also can decide) according to the frequency that the user upgrades wherein, the popular degree P of project t
tBe defined as: the number of times that project t is marked, the number of nonzero term during promptly t is listed as in user-project rating matrix, P
t=| U (t) |, as seen, the number of times of being marked is many more, and project is popular more, and the number of times of being marked is few more, and project is got over unexpected winner.Its hotness degree weight w
tBe defined as:
P wherein
AllBe the popular degree summation of all items, promptly all items total degree of being marked can obtain by all nonzero term numbers in user-project rating matrix.(the hotness degree weight value of each project of employing off-line computation of Period also is recorded among the W (n) for m, the n) data in, and this value has reflected corresponding project role size when calculating user's similarity according to matrix A.For example user 1 and user 2 have estimated film " No. seven, the Changjiang river " and " I am Liu Yuejin ", but by the scanning rating matrix, we find that the number of times that " No. seven, the Changjiang river " is estimated is 200 times, be far longer than " I am Liu Yuejin " by the number of times estimated 40 times, suppose that all the scoring number of times on this film collection are 3000, so the hotness degree weight w in " No. seven, the Changjiang river "
" No. seven, the Changjiang river "=log (3000/200)=2.7, the hotness degree weight w of " I am Liu Yuejin "
" I am Liu Yuejin "=log (3000/40)=4.3, like this, " I am Liu Yuejin " role when weighing user 1 and user's 2 similarity is greater than " No. seven, the Changjiang river ".
4, when targeted customer a arrives, (m n), obtains a scoring item set T to scanning rating matrix A
a, to each project t ∈ T
a, in W (n), find corresponding w
tRecommended engine is according to " user-project " rating matrix A (m, n) and hotness degree weight, adopt the similarity of targeted customer a and other user u to calculate formation user similarity matrix Sim (m, m), in calculating, similarity introduces the hotness degree weight of common scoring item, computational methods are as follows: based on the Pearson correlation coefficient method, the hotness degree weight that wherein adds in the public scoring item each in the molecule, for similarity being limited between-1~1, the hotness degree weight maximum that adds public scoring item in the denominator is done divisor, and formula is as follows:
By computing formula as can be known, as the popular degree P of project t
tBig more, hotness degree weight w
tMore little, this weight joins in the calculating formula of similarity, and the similarity that obtains is just more little.Otherwise, as the popular degree P of t
tMore little, the similarity that obtains is just big more, therefore meets this general knowledge.Watch the high more film of popular degree jointly, the user's similitude that reflects is low more, otherwise, watch the low more film of popular degree jointly, the user's similitude that reflects is just high more.The result sorts from high to low with aforementioned calculation.
5, according to size of the similarity between above-mentioned targeted customer a and other user and ordering, find preceding k the nearest-neighbors the most similar, form nearest-neighbors collection [knn to targeted customer a
1, knn
2, L, knn
k], make sim (a, knn
1)>sim (a, knn
2)>L>sim (a, knn
k).
6, scan A (m, n), the project set T that finds user a not mark
a', at each project j that does not mark of active user a, predictive user a adopts following formula to each t ∈ T to the scoring of project j
a' calculate and predict score value;
The user a that calculates is to all not prediction scorings of scoring item, and according to sorting from big to small, the top n item design recommendation list RecList (N) that chooses the score value maximum gives current active user a with it.
The present invention compares with existing collaborative filtered recommendation method, in similarity is calculated, considered the popular degree difference of project itself, be reflected in the computing formula, made result of calculation more meet objective reality as weight, to recommending accuracy to make moderate progress, improved the recommendation quality to a certain extent.
Claims (4)
1, a kind of collaborative filtered recommendation method of introducing hotness degree weight of program, it is characterized in that on the interface of IPTV program, the visual menu that provides the user to mark, and make program commending and tabulate to the targeted customer according to user's viewing time, behavior operation, program score data that terminal set top box transmits, its concrete steps are as follows:
A. collect the user interest data, make " user-project " rating matrix A (m, n);
B. make the hotness degree weight value of off-line computation of Period project;
C. the project that the current active user has been marked finds corresponding hotness degree weight value;
D. introducing the project that the current active user has been marked finds corresponding hotness degree weight value to calculate the similarity size between targeted customer and other user and sort;
E. K the user who chooses the similarity maximum is as its nearest-neighbors collection;
F. according to nearest-neighbors set pair targeted customer not scoring item predict the scoring and the ordering;
G. will predict that the maximum top n project of scoring makes recommendation list and give the targeted customer.
2, according to the collaborative filtered recommendation method of the described introducing hotness degree weight of program of claim 1, it is characterized in that described " user-project " rating matrix A (m, n) be to carry out arranged with user's score information and user behavior data, the row representative of consumer, row representative project, the element value in the matrix are then represented the favorable rating of this row user to this list of items.
3, according to the collaborative filtered recommendation method of the described introducing hotness degree weight of program of claim 1, it is characterized in that the described off-line cycle calculated once by 30 minutes, the hotness degree weight value of project be with
Calculate w
tBe hotness degree weight; P
AllPopular degree summation for all items; P
tPopular degree for the t project.
4,, it is characterized in that the similarity between described targeted customer and other user is incorporated into Pearson correlation coefficient calculating with the hotness degree weight value according to the collaborative filtered recommendation method of the described introducing hotness degree weight of program of claim 1.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810037498A CN100581227C (en) | 2008-05-16 | 2008-05-16 | Collaborative filtered recommendation method introducing hotness degree weight of program |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810037498A CN100581227C (en) | 2008-05-16 | 2008-05-16 | Collaborative filtered recommendation method introducing hotness degree weight of program |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101287082A CN101287082A (en) | 2008-10-15 |
CN100581227C true CN100581227C (en) | 2010-01-13 |
Family
ID=40059021
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200810037498A Expired - Fee Related CN100581227C (en) | 2008-05-16 | 2008-05-16 | Collaborative filtered recommendation method introducing hotness degree weight of program |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN100581227C (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103353865A (en) * | 2013-04-25 | 2013-10-16 | 杭州电子科技大学 | Barter electronic trading commodity recommendation method based on position |
Families Citing this family (46)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US11263543B2 (en) | 2007-11-02 | 2022-03-01 | Ebay Inc. | Node bootstrapping in a social graph |
US8666909B2 (en) * | 2007-11-02 | 2014-03-04 | Ebay, Inc. | Interestingness recommendations in a computing advice facility |
CN101685458B (en) * | 2008-09-27 | 2012-09-19 | 华为技术有限公司 | Recommendation method and system based on collaborative filtering |
CN102088626B (en) * | 2009-12-02 | 2014-08-13 | Tcl集团股份有限公司 | On-line video recommendation method and video portal service system |
US9087123B2 (en) * | 2009-12-18 | 2015-07-21 | Toyota Jidosha Kabushiki Kaisha | Collaborative filtering using evaluation values of contents from users |
CN101860696B (en) * | 2010-04-24 | 2013-11-06 | 中兴通讯股份有限公司 | Program recommending method and streaming media program system |
CN102654859B (en) * | 2011-03-01 | 2014-04-23 | 北京彩云在线技术开发有限公司 | Method and system for recommending songs |
CN102654860B (en) * | 2011-03-01 | 2015-05-06 | 北京彩云在线技术开发有限公司 | Personalized music recommendation method and system |
CN102164322B (en) * | 2011-03-18 | 2017-06-13 | 海尔集团公司 | The processing method and system of digital television program |
CN102780920A (en) * | 2011-07-05 | 2012-11-14 | 上海奂讯通信安装工程有限公司 | Television program recommending method and system |
EP2546759A1 (en) * | 2011-07-15 | 2013-01-16 | Irdeto Corporate B.V. | Generation of recommendation values |
CN102938121A (en) * | 2011-08-15 | 2013-02-20 | 阿里巴巴集团控股有限公司 | Method and device for releasing commodity information |
CN102722832B (en) * | 2012-01-05 | 2015-08-19 | 合一网络技术(北京)有限公司 | Online Video advertisement becomes more meticulous directed put-on method |
CN103200427A (en) * | 2012-01-09 | 2013-07-10 | 腾讯科技(深圳)有限公司 | Network programme playing method and system, multi-media server and terminal |
CN102722550B (en) * | 2012-05-25 | 2014-10-22 | 清华大学 | Gray forecasting model-based recommending method and system |
CN102750336B (en) * | 2012-06-04 | 2014-09-24 | 电子科技大学 | Resource individuation recommendation method based on user relevance |
CN102929959B (en) * | 2012-10-10 | 2016-02-17 | 杭州东信北邮信息技术有限公司 | A kind of book recommendation method based on user behavior |
CN104918118B (en) * | 2012-10-24 | 2019-08-02 | 北京奇虎科技有限公司 | Video recommendation method and device based on historical information |
CN103905907B (en) * | 2012-12-25 | 2016-11-23 | 腾讯科技(深圳)有限公司 | A kind of media content recommendations method and system |
CN103209342B (en) * | 2013-04-01 | 2016-06-01 | 电子科技大学 | A kind of introduce video popularity and the collaborative filtered recommendation method of user's interests change |
CN104424247A (en) * | 2013-08-28 | 2015-03-18 | 北京闹米科技有限公司 | Product information filtering recommendation method and device |
CN104462093B (en) * | 2013-09-13 | 2019-12-10 | Sap欧洲公司 | personal recommendation scheme |
CN103559206B (en) * | 2013-10-10 | 2017-08-08 | Tcl集团股份有限公司 | A kind of information recommendation method and system |
CN104090892B (en) * | 2013-12-16 | 2016-06-08 | 深圳市腾讯计算机系统有限公司 | A kind of method that off-line algorithm is carried out calculating online and device |
CN103793476B (en) * | 2014-01-08 | 2017-02-15 | 西安电子科技大学 | Network community based collaborative filtering recommendation method |
US20170148083A1 (en) * | 2014-06-12 | 2017-05-25 | Hewlett Packard Enterprise Development Lp | Recommending of an item to a user |
CN105045867B (en) * | 2015-07-14 | 2018-02-13 | 杭州东信北邮信息技术有限公司 | A kind of books writer based on books scoring recommends method |
CN106447364A (en) * | 2015-08-06 | 2017-02-22 | 阿里巴巴集团控股有限公司 | Information recommending method and information recommending device |
CN105608118B (en) * | 2015-12-15 | 2018-12-18 | 成都陌云科技有限公司 | Result method for pushing based on customer interaction information |
CN107305677A (en) * | 2016-04-25 | 2017-10-31 | 北京京东尚科信息技术有限公司 | Product information method for pushing and device |
CN106022838A (en) * | 2016-05-27 | 2016-10-12 | 天津大学 | Optimized selection method of user closest neighbor set of electronic commerce recommendation system |
CN105847901A (en) * | 2016-06-16 | 2016-08-10 | 北京北广传媒地铁电视有限公司 | Subway television media system |
CN108306812B (en) * | 2017-02-08 | 2021-03-26 | 腾讯科技(深圳)有限公司 | Data processing method and server |
CN106980646A (en) * | 2017-02-27 | 2017-07-25 | 上海大学 | The method applied based on popularity to the influencing mechanism analysis of user interest and its in proposed algorithm |
CN110020118B (en) * | 2017-09-29 | 2021-11-16 | 北京国双科技有限公司 | Method and device for calculating similarity between users |
CN108256093B (en) * | 2018-01-29 | 2020-06-19 | 华南理工大学 | Collaborative filtering recommendation algorithm based on multiple interests and interest changes of users |
CN108337544A (en) * | 2018-03-30 | 2018-07-27 | 华东师范大学 | A kind of RTV remote television box based on the study of internet artificial intelligence |
CN108920577A (en) * | 2018-06-25 | 2018-11-30 | 西北工业大学 | Television set intelligently recommended method |
CN108965935B (en) * | 2018-07-31 | 2021-12-10 | 山东广电网络有限公司 | Method and equipment for data acquisition and analysis and information push of multi-terminal broadcast and television service |
CN109327736B (en) * | 2018-10-25 | 2021-01-26 | 合肥澳视智能科技有限公司 | Program recommendation method based on program viewing expression and user information |
CN109726747B (en) * | 2018-12-20 | 2021-09-28 | 西安电子科技大学 | Data fusion ordering method based on social network recommendation platform |
CN109508407A (en) * | 2019-01-14 | 2019-03-22 | 上海电机学院 | The tv product recommended method of time of fusion and Interest Similarity |
CN110351580B (en) * | 2019-07-12 | 2021-07-13 | 四川长虹电器股份有限公司 | Television program topic recommendation method and system based on non-negative matrix factorization |
CN111770383B (en) * | 2020-09-03 | 2021-01-08 | 北京悠易网际科技发展有限公司 | Generation method of equipment association relationship, content recommendation method and device |
CN112423037B (en) * | 2020-10-26 | 2024-02-06 | 深圳Tcl新技术有限公司 | Television program playing method, device, terminal equipment and computer storage medium |
CN113239284B (en) * | 2021-07-13 | 2021-10-15 | 武汉斗鱼鱼乐网络科技有限公司 | Method and device for determining sequence of recommendation information |
-
2008
- 2008-05-16 CN CN200810037498A patent/CN100581227C/en not_active Expired - Fee Related
Non-Patent Citations (5)
Title |
---|
Evaluation of Item-Based Top-N Recommendation Algorithms. George Karypis.ACM Transactions on Information Systems,Vol.22 No.1. 2004 |
Evaluation of Item-Based Top-N Recommendation Algorithms. George Karypis.ACM Transactions on Information Systems,Vol.22 No.1. 2004 * |
Item-Based Collaborative Filtering RecommendationAlgorithms. Badrul Sarwar, George Karypis, Joseph Konstan, andJohnRiedl.Proceedings of the 10th international conference on World Wide Web. 2001 |
Item-Based Collaborative Filtering RecommendationAlgorithms. Badrul Sarwar, George Karypis, Joseph Konstan, andJohnRiedl.Proceedings of the 10th international conference on World Wide Web. 2001 * |
基于兴趣度的协作过滤技术研究. 陈永平等.计算机技术与发展,第18卷第1期. 2008 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103353865A (en) * | 2013-04-25 | 2013-10-16 | 杭州电子科技大学 | Barter electronic trading commodity recommendation method based on position |
CN103353865B (en) * | 2013-04-25 | 2017-04-12 | 杭州电子科技大学 | Barter electronic trading commodity recommendation method based on position |
Also Published As
Publication number | Publication date |
---|---|
CN101287082A (en) | 2008-10-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN100581227C (en) | Collaborative filtered recommendation method introducing hotness degree weight of program | |
CN101489107B (en) | Collaborative filtering recommendation method based on population attribute keyword vector | |
CN103559206B (en) | A kind of information recommendation method and system | |
CN101482884A (en) | Cooperation recommending system based on user predilection grade distribution | |
CN101271559A (en) | Cooperation recommending system based on user partial interest digging | |
CN102063433A (en) | Method and device for recommending related items | |
CN105430505B (en) | A kind of IPTV program commending methods based on combined strategy | |
CN104331459B (en) | A kind of network resource recommended method and device based on on-line study | |
WO2010010653A1 (en) | User model processing device | |
CN103260061B (en) | A kind of IPTV program commending method of context-aware | |
CN102780920A (en) | Television program recommending method and system | |
CN107071578A (en) | IPTV program commending methods | |
CN104935963A (en) | Video recommendation method based on timing sequence data mining | |
CN102929966B (en) | A kind of for providing the method and system of personalized search list | |
CN110737859A (en) | UP main matching method and device | |
CN106131703A (en) | A kind of method of video recommendations and terminal | |
CN112800097A (en) | Special topic recommendation method and device based on deep interest network | |
CN108650532B (en) | Cable television on-demand program recommendation method and system | |
CN106168980A (en) | Multimedia resource recommends sort method and device | |
CN111061945A (en) | Recommendation method and device, electronic equipment and storage medium | |
CN110019943A (en) | Video recommendation method, device, electronic equipment and storage medium | |
CN105338408B (en) | Video recommendation method based on time factor | |
CN106980646A (en) | The method applied based on popularity to the influencing mechanism analysis of user interest and its in proposed algorithm | |
CN104391925A (en) | Video recommendation method and system based on TV (television) user collaborative forecasting | |
WO2016149640A1 (en) | Providing service based on user operation behavior |
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 | ||
C17 | Cessation of patent right | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20100113 Termination date: 20130516 |