WO2005048601A1 - Program recommendation system - Google Patents
Program recommendation system Download PDFInfo
- Publication number
- WO2005048601A1 WO2005048601A1 PCT/IB2004/052241 IB2004052241W WO2005048601A1 WO 2005048601 A1 WO2005048601 A1 WO 2005048601A1 IB 2004052241 W IB2004052241 W IB 2004052241W WO 2005048601 A1 WO2005048601 A1 WO 2005048601A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- content
- sequence
- pieces
- sequences
- piece
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/438—Interfacing the downstream path of the transmission network originating from a server, e.g. retrieving MPEG packets from an IP network
- H04N21/4383—Accessing a communication channel
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/458—Scheduling content for creating a personalised stream, e.g. by combining a locally stored advertisement with an incoming stream; Updating operations, e.g. for OS modules ; time-related management operations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/16—Analogue secrecy systems; Analogue subscription systems
- H04N7/162—Authorising the user terminal, e.g. by paying; Registering the use of a subscription channel, e.g. billing
- H04N7/163—Authorising the user terminal, e.g. by paying; Registering the use of a subscription channel, e.g. billing by receiver means only
Definitions
- the invention relates to a system and a method for program recommendation.
- EPGs electronic program guides
- program content information is transmitted along with the pieces of content, which form the actual program.
- the program content information includes, for each of the different accessible channels, a broadcast time for content pieces and a content description thereof.
- the content description may be limited to the title only but may also include further information, such as a natural language description, or other categorised information such as the type or genre of the piece of content as well as additional information such as the author, artist or actor performing therein, etc..
- recommendation systems Based on the available program information, recommendation systems have been proposed. Generally these recommendation systems rely on a user profile, ⁇ which includes the users preferences. For program recommendation, the above mentioned program content information is accessed and the content description of content pieces is compared to the user profile.
- a corresponding recommender system is given in US-A-6,163,316.
- the recommender system is associated with a television set.
- a data base with program content information on a plurality of channels is accessed.
- the system includes storage means for storing a viewer profile.
- Selection means match the user profile against content descriptions in the program content information and generate a preferred schedule, which indicates the desirability of a particular program relative to other programs.
- recommendation systems of the type discussed above are only oriented towards a single instant. If the user wishes to plan watching television for a specified time interval, which might involve several pieces of content, the known recommendation systems do not assist his choice appropriately. Another issue in assisting the viewer to choose an audio or video program is the way the available content pieces are shown to the user.
- matrix-style appearance is widely used.
- available channels are shown as parallel horizontal rows where start and end times of content pieces are marked on a horizontal time axis.
- An example of this type of representation is given in US-A-6163316, where content pieces are shown with coloured overlays indicating recommended programs.
- a disadvantage of matrix-style per-channel representation is that only a very limited number of channels can be displayed in a clear and concise manner. Thus, channels where interesting content pieces are only shown occasionally (e.g. a local channel, where only the news are interesting to the user) will most likely not be part of the, say, 5 channels represented and therefore be invisible to the user.
- the object of the invention to provide a recommendation system well suited to assist the user in choosing a program from several available content pieces.
- This object is solved according to the invention by a system for program recommendation according to claim 1 and a method for program recommendation according to claim 13.
- Dependent claims relate to preferred embodiments of the invention.
- the system according to the invention employs a profile and corresponding program content information on a plurality of channels.
- the profile is preferably a user profile for a single user or a group of users.
- Content pieces broadcast over the channels are matched against the user profile to determine a piece score.
- the content description comprises one or more of the following information of a content piece: Category with possible subcategories like e.g.
- a user profile may comprise preference values for one or more of the above information indicating the users preference.
- a user profile may contain a preference value of 0.8 for a content piece of the category "sports", but only a preference value of 0.4 for the category "financial”.
- the preference value is used to calculate a piece score of a particular piece of content.
- the recommended program is not selected exclusively based on this piece score. Instead, sequences of content pieces are examined. These sequences cover a specified time interval and contain content pieces consecutively broadcast in this time interval. For a plurality of these sequences, a sequence score is calculated.
- the calculation of the sequence score is based on one hand on the piece scores of the individual pieces contained in the sequence, and on the other hand on correlation of the content descriptions of at least two content pieces in the sequence.
- the sequence score thus determined reflects the presumed desirability of the chosen program (sequence of content pieces) for the user. By not only considering single pieces of content, but by determining a score for whole sequences thereof the finally selected recommendation is more suited to the user's interests.
- a recommendation system solely based on piece scores will, for any given time, only recommend the content with the highest piece score
- the recommendation system according to the invention may use correlation rules to determine a suggestion which, seen over the specified time period, is more likely to appeal to the user. This may be illustrated by a simple example: With the huge variety of television channels available, a TV recommendation system based on piece scores alone might, at any time during the day, propose to the user his favourite program.
- the prior art system may recommend to him a sequence entirely consisting of sports news on different channels, where the same events will be related time after time.
- the recommendation system according to the invention may have a correlation rule penalising sequences with multiple occurrences of the same type of content.
- the recommendation selected here is likely to include a mixture of different features the user is interested in, with only a few sports news shows therein.
- the sequences considered according to the invention contain content pieces consecutively broadcast in the time interval. While there may be pauses in this sequence, i.e. times for which know recommendation is given, it is generally preferred that the sequence covers the whole time interval. It is also possible that pauses are filled by inserting recorded content pieces, which have been previously stored.
- a sequence may include one or more content pieces only partly.
- a sequence may include a content piece on a first channel for a certain time period, and then — after the end of the first content piece - switching into a second content piece on another channel, which has already been running for some time.
- a sequence may include interrupting a first content piece by switching to another channel where a second content piece is presently starting.
- a sequence may also include a switch over from one running content piece into another. It is, however, preferred to select sequences such that the switch over time from a first to a second content piece corresponds to the end time of the first content piece and/or to the start time of the second content piece, so that intermediate switchovers (from running feature into running feature) are avoided.
- the calculation be based on both the individual piece scores and correlation of pieces contained in the sequence.
- the calculation may involve, for example, calculating an average of piece scores of the pieces contained in the sequence and by then modifying (e.g. adding to or multiplying by) this value with one or more correlation values.
- correlation of content descriptions of pieces in the sequences is judged according to one ore more correlation rules. These rules may be dependent on the user, and therefore be contained in the user profile.
- Each correlation rule contains instructions on how to calculate a correlation value. This correlation value is representative of a correlation of content descriptions of two pieces.
- An example of a correlation rule would be to calculate a negative correlation value for every pair of content pieces in the sequence which are of the same type, e.g. several news shows. The more occurrences of content pieces of common type are found, the more negative the correlation value becomes. The finally obtained correlation value would then, e.g. be added to the sum of piece scores to calculate the sequence score, so that a negative correlation value would lower than the overall score.
- the selection means are configured to calculate the sequence score such that it is lower if two or more content pieces in the sequence are of a common type, than if this is not the case. This serves to balance the sequence, so that a sequence with content pieces of different types is more likely to be selected.
- the corresponding (negative) correlation value may be dependent on the type of the content pieces, so that e.g. multiple occurrences of news shows are penalised more than multiple talk shows.
- the sequence score is lower the more switchovers between content pieces are contained in the sequence. This rule favours completion of content pieces, rather than frequent changes.
- calculation of the sequence score may involve checking if content pieces of one or more preferred categories or types are present in the sequence at all. If this is not the case, the sequence score is lower. As explained above, sequence scores are calculated for a number of sequences and, according to the obtained sequence score, at least one sequence is finally selected for recommendation.
- Another object is to provide a graphical representation for a user which shows a number of recommended sequences in a clear and concise manner. This is archived, according to the development of the invention, by showing a plurality of recommended sequences (e.g. N-best) to the user in a lattice representation, where content pieces are represented as edges running between a start time and an end time on a time axis.
- a plurality of recommended sequences e.g. N-best
- a content piece contained in two or more display sequences is only represented as a single edge. This serves to show sequences to the user in a concise manner. From this type of representation, the user can easily recognise his choices between the different sequences presented. In some cases, the same content piece may be comprised in a large number of high scoring sequences. It would then be possible, that for certain time intervals the user would not be shown any alternative. To prevent this, it is proposed according to a development of the invention to select the displayed sequences such that for each point in time at least two alternatives are presented.
- the selection of displayed sequences may not be exclusively based on sequence scores (N-best).
- the user can change either the represented time interval or the shown sequences, or both.
- the user may input commands to scroll along the time axis. This scrolling may follow one specifically chosen path.
- the user may scroll among represented sequences. If, for example, the N-best sequences are shown, the user may input a command to display further sequences with lower scores. In this case, it is preferred that, while showing lower ranking sequences instead of the higher ranking sequences, the top ranking path stays on the display for reference, e. g. on top.
- examples of program recommendation systems according to the invention will be shown with a regard to figures, where
- Fig. 1 shows a diagram with content pieces of a first example in a lattice structure
- Fig. 2 shows a table with the available content pieces from the first example
- Fig. 3 shows a table of content pieces and piece scores in the first example
- Fig. 4 shows a table of paths in the lattice of fig. 1 with associated content pieces, piece scores and path scores
- Fig. 5 shows a program recommendation for the first example presented in lattice form
- Fig. 6 shows a program recommendation presented in matrix-style form
- Fig. 7 shows a diagram representing content pieces of a second example in a lattice structure
- Fig. 8 shows a lattice diagram with possible sequences of the second example represented as paths
- Fig. 9 shows a program recommendation for the second example presented in a first embodiment of a lattice form
- Fig. 10 shows a program recommendation for the second example presented in a second embodiment of a lattice form.
- a person who wants to pick a TV or audio feature for consumption needs to make a quite complex decision, given the wealth of available content.
- the user has personal preferences and prefers some audio or video content pieces (e.g. movies, tv shows, songs, radio shows etc.) over other content pieces.
- the corresponding decision is already not easy if a large number of content pieces are available at a given time.
- the decision gets even more complex if consumption needs to be planed for a certain time frame, instead of a specific moment.
- a program recommendation system and method will be explained and a detail example will be given on how, for a specific available content and a specific user profile, a recommendation is obtained and presented.
- the specific realisation of a device which implements the system or method may differ.
- the method may be implemented as a computer program running on a stand alone computer or on several computers, e.g. as a client/server application, where a client and a server are connected over a computer network.
- the computer would include storage means, such as RAM memory, and optical or magnetic recording devices for storing data such as user profile and correlation rules.
- the computer will further need to access a source of program information, which may be provided e.g. over a computer network.
- the program recommendation system may also be included in an audio or video device, such as a TV set, satellite receiver, radio etc., where the program finally selected may then be played.
- EPGs electronic program guides
- Figure 2 shows as an example for five different channels a matrix-style EPG with the available content pieces per channel from 14.00 h -17.00 h.
- each of the available channels offers three content pieces in the regarded time interval.
- Planning of consumption of a television program within this 3 hour interval will involve choosing an initial channel to watch, a first switch over at a first switch over time to switch to a second channel, a third switch over and so on.
- intermediate switchovers from running feature into running feature
- a lattice is a connected graph, possibly along some fixed axis (e.g. time), with nodes and edges.
- the nodes represent possible switchovers, while the content pieces (or parts thereof) are represented by the edges that connect two of the nodes.
- the idea of a lattice representation serves two purposes. The first purpose is the internal concept of organizing the content pieces in a connected manner, and of running evaluation algorithms to determine program recommendations. This internal concept will generally not be apparent to the user. The second purpose is the concept of presenting the result of the optimisation to the user.
- the displayed graphical representation which may also take a lattice form, will be described later.
- FIG. 1 shows a lattice diagram as an internal organisation structure corresponding to the example of fig. 2.
- the bullets represent possible switchovers, i.e. a change of content pieces (note, however, that viewing two consecutive content pieces of the same channel is treated as switchover, too).
- Solid black bullets designate switchovers where the corresponding content piece was not completed, circles designate switchovers where a content piece starts or ends.
- the edges have a label which gives a descriptor (program information such as title and channel).
- a plurality of paths may be found. Each path represents a sequence of content pieces and involves edges and nodes in the graph.
- a user profile is stored giving personal preferences of the user with the regard to, e.g. involved artists, genre etc..
- This user profile may be time-dependent (e.g. a user likes to watch movies in the evening, but not during the day). For each content piece, additional to channel number, start time and end time, content information is available.
- the content information may include the genre or type of the content piece, a natural language description of the content, names of artists, actors or author of a content piece etc.
- a match of the content description against the user profile is effected to obtain a piece score for each individual content piece in fig. 1 and 2.
- the piece score reflects the users preference for the individual content piece.
- An example of pieces scores for the content pieces of fig. 2 is given in fig. 3.
- a number of paths in the lattice of fig. 1 are scored in a second step by calculating a path score (or sequence score).
- the path score means that not only single pieces of content, but complete sequences are evaluated.
- the calculation of path scores depends on both the piece scores or the content pieces in the path and on a correlation of these content pieces. This correlation can be scored according to a number of rules. These may be implemented similar to the "language model" used in speech recognition, that models the likelihood of word sequences and phrases.
- the rules may define a probability model, assigning conditional probabilities for certain pieces of content, given the predecessor in the sequence. For instance, if a sequence already contains a news show, later news shows will get a lower probability. As another example, a sequence containing documentary elements on a certain subject will receive a higher probability (higher score) for follow-ups or updates.
- Global rules examine whole sequences from start to end and award scores containing to the whole sequence.
- a global rule may, for example, check for completeness of a sequence, so that e.g. sequences are penalised if they do not contain at least one news show.
- Other global rules may model a desired sequence structure by awarding scores according to the proportion of the content types of the different content pieces. For example, sequences with more than 60 % sports, or such with more than 30 % news may be penalised.
- these rules may be dependent on the user's preferences.
- local rules do not examine whole sequences from start to end, but only pertain to a limited number of consecutive content pieces.
- a local rule may be added to provide "smooth" transitions between content pieces i.e. transitions that appeal to the user.
- Such a local rule will only regard two consecutive content pieces and judge, if the transition is adequate.
- the transition rule will penalise "harsh” transitions (e.g. a hard rock song followed by a ballad) and award higher scores for smooth transitions.
- Such a local rule will be applied to all pairs of consecutive content pieces within a sequence.
- a path score will be determined from a plurality of factors.
- the proportion, according to which each of the factors (piece scores, scores contributed by individual rules) is accounted for in the sequence score may be adjusted by weight factors.
- Some rules may be "must" rules, which are required to be fulfilled in a recommended sequence. Such “must” rules may be modelled by assigning a very high (infinitely high) penalty if the corresponding rule is not fulfilled.
- the rules applied to determine a path score may be fixed as part of a recommendation system. It is, however preferred that at least the individual weight factors of the rules, and possibly also some rule definitions are part of the user profile. This will be especially be advisable for global rules (e.g. with desired content type distributions), which reflect the user's individual preferences.
- the number of switchovers in the recommended sequence i.e.
- path scores are calculated as described above.
- Fig. 4 is a table with an example of different paths, the corresponding piece scores and the finally obtained path scores.
- path #1 contains content pieces Al , A2, B3, and C2, which each have a corresponding piece score.
- the whole sequence (path #1) is assigned a path score.
- the final recommendation is found according to the path score.
- the three sequences with the highest path score values are presented to the user for recommendation.
- Another means to restrict the size of the search lattice is to keep only a fixed number of favorable hypotheses and to dispose of all others ("histogram pruning"). It should be noted, that by any pruning method, it is possible to loose the path through the network that has the best score overall in the end. This kind of error is accordingly called pruning or search error, and should of course be minimized as far as possible.
- pruning or search error the concept of a lattice may not only be used as an internal concept, but can also be advantageously applied to the graphical representation shown to the user. For a better overview, it is preferred to present the recommendation result to the user in a lattice representation. A corresponding example is given in fig. 5.
- the graph shows the top-ranking sequences. Information on the type of content piece is included by displaying the different bars in correspondingly assigned colours or line styles.
- the graph shows the top-ranking sequences. In cases, where an alternative path differs only slightly, only the alternative segment is shown (in the example of fig. 6, the features "talk show 3" and "sports" are alternative for the two top-ranking paths.
- the user profile in the example is given as numerical values attributed to different content categories.
- the program content information includes, for each content piece, one or more content categories out of a predetermined list.
- a lattice is determined from the available content.
- the lattice includes nodes (corresponding to start, end and switchovers) and edges interconnecting the nodes (corresponding to content pieces).
- the corresponding "pruned" (scored) lattice is given in fig. 7. In the lattice of fig. 7 all possible sequences (paths) are evaluated.
- sequence scores are calculated.
- the calculation of a sequence score is based on the pieces scores of the individual content pieces and on a number of rules.
- a first set of rules models correlation of the content descriptions of content pieces in the sequence.
- these rules are given as "negative rules” where penalties are given for unwanted sequences.
- the simple rule set given below models a "bigram", i.e. only a sequence of two subsequent categories is taken into account.
- a full grammar typically will model complete paths: 1. Info Info -0.2 2. News Info -0.1 3. Info News 0.0 4. News News -0.3 5. SciFi SciFi 0.0 6. Sports Sports -0.1 7. Other Other -0.3 Additionally, general penalty rules are applied to each paths: 8. No news at all penalty: - 0.3 9. No SciFi at all penalty: - 0.1
- For calculating path scores first the sum of the individual pieces scores is taken (note, that this is only possible in the given example, because each sequence contains three content pieces.
- Fig. 8 shows the three highest scoring paths in an internal lattice representation.
- Fig. 9 shows a preferred graphical representation to present the recommendation to the user. Note, that here content pieces common to the displayed sequences are represented only once. The highest scoring path is shown in a solid, bold line, whereas the nextcoming two paths are shown in dotted and slashed lines. On a coloured display, the paths could alternatively shown in different colours.
- Fig. 10 shows an alternative display, which could be used to present the above recommendation to a user.
- linetypes alternatively: colours
- the given representations may be scrolled by the user horizontally (to view how the sequences continue in time) or vertically (to look at further sequences, that where not among the initially presented N-best sequences). In case of vertical scrolling, it is preferred that the top scoring path stays on the display.
- a number of modifications can be incorporated in the recommendation system.
- the actual current viewing or listening history of the user can be incorporated into the recommendation for the future. For example, if the user already consumed a news update in the last hour before the recommendation is established, sequences including further news updates will received a lower recommendation probability. This can be archived by using the current path of consumed content elements as starting piece for all future paths.
- the content sequence model can then incorporate the actual history into the probabilities for the ongoing parts of the path.
- the further modification concerns the user profile.
- a general (e.g. broad audio viewing) profile can be combined (i.e. interpolated) with a user's specific profile.
- a recommendation for a group of people is needed, several user profiles may be combined.
- Another modification concerns the content pieces. While in the above example only live content (i.e. content pieces that were viewed at the same time they are broadcasted) was considered, the recommendation may also include content pieces from other sources, such as recorded content pieces that are stored in a local or remote accessible memory and may be played at any time.
- the original viewing time and the number of times the content has been already played will preferably be considered in determining a piece score.
- Other possible modifications concern the user's interaction with the recommendation system. While basically the recommendation system may work without any user interaction (i.e. the system simply displays the N-best sequences) it is preferred that the user may interact with the system. For example, the user may review the recommendation and expressly eliminate certain content pieces contained in the recommendation (i.e. a movie he already knows). In addition or alternatively the user may specifically include one or more content pieces. The system may then perform a re-evaluation of possible sequences without the eliminated content pieces, and including the chosen content pieces and present the result to the user. Another user interaction generally required will be to enter the time period for which he wishes to obtain a recommendation.
Abstract
Description
Claims
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US10/578,648 US7594244B2 (en) | 2003-11-12 | 2004-10-29 | Program recommendation system |
JP2006539004A JP4653757B2 (en) | 2003-11-12 | 2004-10-29 | Program recommendation system |
EP04770340A EP1685710A1 (en) | 2003-11-12 | 2004-10-29 | Program recommendation system |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP03104158.5 | 2003-11-12 | ||
EP03104158 | 2003-11-12 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2005048601A1 true WO2005048601A1 (en) | 2005-05-26 |
Family
ID=34585884
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/IB2004/052241 WO2005048601A1 (en) | 2003-11-12 | 2004-10-29 | Program recommendation system |
Country Status (5)
Country | Link |
---|---|
US (1) | US7594244B2 (en) |
EP (1) | EP1685710A1 (en) |
JP (1) | JP4653757B2 (en) |
CN (1) | CN100551033C (en) |
WO (1) | WO2005048601A1 (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070186244A1 (en) * | 2006-02-07 | 2007-08-09 | Samsung Electronics Co., Ltd. | Method and apparatus for providing filtering interface for recording and/or searching for broadcast content |
WO2011038035A1 (en) * | 2009-09-23 | 2011-03-31 | Qualcomm Incorporated | System for power-efficiently delivering personalized contents |
CN107404657A (en) * | 2017-07-03 | 2017-11-28 | 北京奇艺世纪科技有限公司 | Method and apparatus are recommended in a kind of advertisement |
CN108062390A (en) * | 2017-12-15 | 2018-05-22 | 广州酷狗计算机科技有限公司 | The method, apparatus and readable storage medium storing program for executing of recommended user |
CN111246257A (en) * | 2020-03-17 | 2020-06-05 | 百度在线网络技术(北京)有限公司 | Video recommendation method, device, equipment and storage medium |
CN111615800A (en) * | 2017-12-18 | 2020-09-01 | 万向媒体有限公司 | Device for measuring jumping-out rate of broadcast service |
Families Citing this family (39)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6704553B1 (en) | 1997-10-08 | 2004-03-09 | Thomas M. Eubanks | System and method for providing automatic tuning of a radio receiver and for providing automatic control of a CD/tape player |
US20040073924A1 (en) * | 2002-09-30 | 2004-04-15 | Ramesh Pendakur | Broadcast scheduling and content selection based upon aggregated user profile information |
US7719971B1 (en) * | 2004-09-15 | 2010-05-18 | Qurio Holdings, Inc. | Peer proxy binding |
JP2008529117A (en) * | 2005-01-21 | 2008-07-31 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | Method and apparatus for obtaining degree of common interest of user groups |
US7779011B2 (en) | 2005-08-26 | 2010-08-17 | Veveo, Inc. | Method and system for dynamically processing ambiguous, reduced text search queries and highlighting results thereof |
US7788266B2 (en) | 2005-08-26 | 2010-08-31 | Veveo, Inc. | Method and system for processing ambiguous, multi-term search queries |
JP4256371B2 (en) * | 2005-09-08 | 2009-04-22 | 株式会社東芝 | Viewing recommendation apparatus and method |
KR100562427B1 (en) * | 2005-10-04 | 2006-03-17 | 주식회사 솔루션박스 | Apparatus for receiving contents and recording medium storing program therefor |
JP2007142840A (en) * | 2005-11-18 | 2007-06-07 | Canon Inc | Information processing apparatus and information processing method |
US7774341B2 (en) | 2006-03-06 | 2010-08-10 | Veveo, Inc. | Methods and systems for selecting and presenting content based on dynamically identifying microgenres associated with the content |
EP4209927A1 (en) | 2006-04-20 | 2023-07-12 | Veveo, Inc. | User interface methods and systems for selecting and presenting content based on user navigation and selection actions associated with the content |
US7992171B2 (en) | 2006-09-06 | 2011-08-02 | Qurio Holdings, Inc. | System and method for controlled viral distribution of digital content in a social network |
US7925592B1 (en) | 2006-09-27 | 2011-04-12 | Qurio Holdings, Inc. | System and method of using a proxy server to manage lazy content distribution in a social network |
US8554827B2 (en) * | 2006-09-29 | 2013-10-08 | Qurio Holdings, Inc. | Virtual peer for a content sharing system |
US8078884B2 (en) | 2006-11-13 | 2011-12-13 | Veveo, Inc. | Method of and system for selecting and presenting content based on user identification |
US7886334B1 (en) | 2006-12-11 | 2011-02-08 | Qurio Holdings, Inc. | System and method for social network trust assessment |
US7888582B2 (en) * | 2007-02-08 | 2011-02-15 | Kaleidescape, Inc. | Sound sequences with transitions and playlists |
DE102007009414A1 (en) * | 2007-02-23 | 2008-08-28 | Deutsche Telekom Ag | Method and system for trouble-free switching between program channels in a video environment |
JP4730619B2 (en) * | 2007-03-02 | 2011-07-20 | ソニー株式会社 | Information processing apparatus and method, and program |
US20080320513A1 (en) * | 2007-06-22 | 2008-12-25 | Microsoft Corporation | Dynamic channel surfing guide and customized television home page |
US9497286B2 (en) * | 2007-07-07 | 2016-11-15 | Qualcomm Incorporated | Method and system for providing targeted information based on a user profile in a mobile environment |
US20090158342A1 (en) * | 2007-12-18 | 2009-06-18 | Motorola, Inc. | Apparatus and method for generating content program recommendations |
US8943081B2 (en) | 2008-11-17 | 2015-01-27 | At&T Intellectual Property I, L.P. | User-powered recommendation system |
US8359192B2 (en) * | 2008-11-19 | 2013-01-22 | Lemi Technology, Llc | System and method for internet radio station program discovery |
US9166714B2 (en) | 2009-09-11 | 2015-10-20 | Veveo, Inc. | Method of and system for presenting enriched video viewing analytics |
US20110191332A1 (en) * | 2010-02-04 | 2011-08-04 | Veveo, Inc. | Method of and System for Updating Locally Cached Content Descriptor Information |
US8793729B2 (en) * | 2010-03-31 | 2014-07-29 | Verizon Patent And Licensing, Inc. | Preferential program guide |
TWI488493B (en) * | 2010-04-16 | 2015-06-11 | Mitac Int Corp | Favorite program recommendation apparatus and method |
US20120036531A1 (en) * | 2010-08-05 | 2012-02-09 | Morrow Gregory J | Method and apparatus for generating automatic media programming through viewer passive profile |
JPWO2012073718A1 (en) * | 2010-11-29 | 2014-05-19 | 日本電気株式会社 | Content analysis system, content analysis apparatus, content analysis method, and content analysis program |
JP5729001B2 (en) * | 2011-02-18 | 2015-06-03 | ソニー株式会社 | Information providing system, information providing method and program |
US8826313B2 (en) * | 2011-03-04 | 2014-09-02 | CSC Holdings, LLC | Predictive content placement on a managed services systems |
US11049029B2 (en) * | 2015-02-22 | 2021-06-29 | Google Llc | Identifying content appropriate for children algorithmically without human intervention |
WO2016207308A1 (en) * | 2015-06-26 | 2016-12-29 | British Telecommunications Public Limited Company | Digital content provision |
WO2017015098A1 (en) | 2015-07-17 | 2017-01-26 | Tribune Broadcasting Company, Llc | Video-production system with social-media features |
CN104991962B (en) * | 2015-07-22 | 2019-01-18 | 无锡天脉聚源传媒科技有限公司 | A kind of method and device generating recommendation information |
US10387107B2 (en) * | 2015-07-31 | 2019-08-20 | Nike, Inc. | Utilizing athletic activities to augment audible compositions |
US9936239B2 (en) * | 2016-06-28 | 2018-04-03 | Intel Corporation | Multiple stream tuning |
JP6924108B2 (en) * | 2017-09-19 | 2021-08-25 | 日本放送協会 | Plan making device, plan making method and plan making program |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1170722A1 (en) * | 2000-07-04 | 2002-01-09 | Sony France S.A. | Incremental sequence completion system and method |
US20020108113A1 (en) * | 2000-12-06 | 2002-08-08 | Philips Electronics North America Corporation | Recommender system using "fuzzy-now" for real-time events |
US20030126600A1 (en) * | 2001-12-27 | 2003-07-03 | Koninklijke Philips Electronics N.V. | Smart suggestions for upcoming TV programs |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6163316A (en) | 1997-01-03 | 2000-12-19 | Texas Instruments Incorporated | Electronic programming system and method |
-
2004
- 2004-10-29 CN CNB2004800332374A patent/CN100551033C/en not_active Expired - Fee Related
- 2004-10-29 JP JP2006539004A patent/JP4653757B2/en not_active Expired - Fee Related
- 2004-10-29 WO PCT/IB2004/052241 patent/WO2005048601A1/en active Application Filing
- 2004-10-29 EP EP04770340A patent/EP1685710A1/en not_active Withdrawn
- 2004-10-29 US US10/578,648 patent/US7594244B2/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1170722A1 (en) * | 2000-07-04 | 2002-01-09 | Sony France S.A. | Incremental sequence completion system and method |
US20020108113A1 (en) * | 2000-12-06 | 2002-08-08 | Philips Electronics North America Corporation | Recommender system using "fuzzy-now" for real-time events |
US20030126600A1 (en) * | 2001-12-27 | 2003-07-03 | Koninklijke Philips Electronics N.V. | Smart suggestions for upcoming TV programs |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070186244A1 (en) * | 2006-02-07 | 2007-08-09 | Samsung Electronics Co., Ltd. | Method and apparatus for providing filtering interface for recording and/or searching for broadcast content |
US8793734B2 (en) * | 2006-02-07 | 2014-07-29 | Samsung Electronics Co., Ltd. | Method and apparatus for providing filtering interface for recording and/or searching for broadcast content |
WO2011038035A1 (en) * | 2009-09-23 | 2011-03-31 | Qualcomm Incorporated | System for power-efficiently delivering personalized contents |
KR101422369B1 (en) | 2009-09-23 | 2014-07-29 | 퀄컴 인코포레이티드 | System for power-efficiently delivering personalized contents |
CN107404657A (en) * | 2017-07-03 | 2017-11-28 | 北京奇艺世纪科技有限公司 | Method and apparatus are recommended in a kind of advertisement |
CN108062390A (en) * | 2017-12-15 | 2018-05-22 | 广州酷狗计算机科技有限公司 | The method, apparatus and readable storage medium storing program for executing of recommended user |
CN108062390B (en) * | 2017-12-15 | 2021-07-23 | 广州酷狗计算机科技有限公司 | Method and device for recommending user and readable storage medium |
CN111615800A (en) * | 2017-12-18 | 2020-09-01 | 万向媒体有限公司 | Device for measuring jumping-out rate of broadcast service |
CN111246257A (en) * | 2020-03-17 | 2020-06-05 | 百度在线网络技术(北京)有限公司 | Video recommendation method, device, equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
JP2007511154A (en) | 2007-04-26 |
JP4653757B2 (en) | 2011-03-16 |
US20070044122A1 (en) | 2007-02-22 |
CN1879413A (en) | 2006-12-13 |
CN100551033C (en) | 2009-10-14 |
EP1685710A1 (en) | 2006-08-02 |
US7594244B2 (en) | 2009-09-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7594244B2 (en) | Program recommendation system | |
CN102227695B (en) | Audiovisual user interface based on learned user preferences | |
US9402101B2 (en) | Content presentation method, content presentation device, and program | |
JP4505418B2 (en) | Program recommendation device, program recommendation method of program recommendation device, and program | |
CN1659882B (en) | Method and system for implementing content augmentation of personal profiles | |
CN100550014C (en) | Information indexing device | |
KR101832996B1 (en) | Enhanced program guide | |
JP4211940B2 (en) | Information providing apparatus and information providing method | |
US20020059593A1 (en) | Process for dynamic navigation among multimedia documents and multimedia terminal for the implementation of the process | |
JP2010066837A (en) | Information processing apparatus and method, program, and recording medium | |
JP2000013708A (en) | Program selection aiding device | |
JP2004295102A (en) | Speech recognition dictionary generating device and information retrieval device | |
JP2004295102A5 (en) | ||
JP4305865B2 (en) | Program automatic selection device, program automatic selection method, and program automatic selection program | |
JP4219913B2 (en) | Program recording device | |
US20060059517A1 (en) | System and method for creating a play sequence for a radio or tv program | |
JP2006080621A (en) | Video image outline list display apparatus | |
JP2008099012A (en) | Content reproduction system and content storage system | |
JP4305862B2 (en) | Program ranking apparatus, program ranking method, and program ranking program | |
WO2013050265A1 (en) | Method and apparatus for providing information for a multimedia content item | |
JP4305860B2 (en) | Program automatic selection device, program automatic selection method, and program automatic selection program | |
JP4305863B2 (en) | Program ranking apparatus, program ranking method, and program ranking program | |
JP5115127B2 (en) | Program recording / reproducing apparatus and program recording / reproducing method | |
JP4305864B2 (en) | Program automatic selection device, program automatic selection method, and program automatic selection program | |
JP4305861B2 (en) | Program ranking apparatus, program ranking method, and program ranking program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 200480033237.4 Country of ref document: CN |
|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NA NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): BW GH GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IT LU MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2004770340 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2007044122 Country of ref document: US Ref document number: 1602/CHENP/2006 Country of ref document: IN Ref document number: 10578648 Country of ref document: US |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2006539004 Country of ref document: JP |
|
WWP | Wipo information: published in national office |
Ref document number: 2004770340 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 10578648 Country of ref document: US |