CN101069424A - Apparatus and method for estimating user interest degree of a program - Google Patents
Apparatus and method for estimating user interest degree of a program Download PDFInfo
- Publication number
- CN101069424A CN101069424A CN 200580041044 CN200580041044A CN101069424A CN 101069424 A CN101069424 A CN 101069424A CN 200580041044 CN200580041044 CN 200580041044 CN 200580041044 A CN200580041044 A CN 200580041044A CN 101069424 A CN101069424 A CN 101069424A
- Authority
- CN
- China
- Prior art keywords
- program
- user
- length
- plays clip
- time
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
A multimedia collaboration system that integrates separate real-time and asynchronous networks-the former for real-time audio and video, and the latter for control signals and textual, graphical and other data-in a manner that is interoperable across different computer and network operating system platforms and which closely approximates the experience of face-to-face collaboration, while liberating the participants from the limitations of time and distance. These capabilities are achieved by exploiting a variety of hardware, software and networking technologies in a manner that preserves the quality and integrity of audio/video/data and other multimedia information, even after wide area transmission, and at a significantly reduced networking cost as compared to what would be required by presently known approaches. The system architecture is readily scalable to the largest enterprise network environments. It accommodates differing levels of collaborative capabilities available to individual users and permits high-quality audio and video capabilities to be readily superimposed onto existing personal computers and workstations and their interconnecting LANs and WANs. In a particular preferred embodiment, a plurality of geographically dispersed multimedia LANs are interconnected by a WAN. The demands made on the WAN are significantly reduced by employing multi-hopping techniques, including dynamically avoiding the unnecessary decompression of data at intermediate hops, and exploiting video mosaicing, cut-and-paste and audio mixing technologies so that significantly fewer wide area transmission paths are required while maintaining the high quality of the transmitted audio/video.
Description
Technical background
The present invention relates to a kind of information recommendation system, relate in particular to the method and apparatus of a kind of estimating user the interest-degree of a program.
Along with the development of modern communication technology, people can obtain bulk information at any time.Yet the rapid increase of amount of information makes people often at a loss as to what to do, and people have pressed for a kind of instrument and can help them to find real interested content, i.e. Ge Xinghua information recommendation system fast.
In order to satisfy the user interest that constantly changes, the files on each of customers in the information recommendation system (User Profile) also needs constantly to change.So, how, dynamically revise the files on each of customers in the commending system according to user's interest (hobby), to recommend the real interested content of user to the user, become the present problem that needs solution.
At present usually watch the manner of a specific program to revise in the files on each of customers user to the fancy grade and the weight of the content characteristic in this specific program according to the user.Described user's manner comprises that the user watches the time span of described specific program, the program that comprises described content characteristic watched number of times and deletion number of times etc.
Described content characteristic is meant the actor names that comprises in the program (Fan Bingbing, Pueraria lobota excellent), program category (literary film, romance movie, horror film etc.), director (Zhang Yimou, Feng Xiaogang etc.) etc., these content characteristics can be from information sources such as broadcasting, TV or internet, the most representatively sends to the user by electronic program guide of digital television (EPG) with program exactly.
When watching manner to revise files on each of customers according to described user, normally obtaining the user earlier watches the time span of certain specific program and deducts a preset threshold value, be scheduled to the ratio of the time span of broadcast by resulting difference and this specific program again, draw the interest-degree of user, to be used for revising files on each of customers to this program.
Described interest-degree can be expressed as:
W wherein
DiThe time span that the expression user watches a specific program; θ is a predetermined threshold value, can be presetted by program provider; R
DiThe total time length of representing the predetermined broadcast of this specific program.
The method of this acquisition interest-degree is too simple, can not reflect user's interests change very exactly.Because also there are a lot of other situations in the user when watching a specific program, as try this program with confirm whether oneself likes this program, when changing platform through this program, see in the process of this program and often switch to other program, in seeing other program process, switch to this program, suspend and watch etc., still only seen programme content how long decide the user whether to like this program and content corresponding feature in these cases, be difficult to very accurately obtain user's interest-degree with the user.Aftermentioned is referred to as interruption with these other situations.
For example a specific program is original just very short, after the user tries more than half programme content from the starting end of program, just find to control oneself not to like this specific program, if seen more than half with the user as foundation, think that the user likes this program and content corresponding feature, then be difficult to reflection user's true interest, too short because of program, the user just tries and has just play most.
Moreover, if the user begins to watch this program to program to finish from described more than half programme content, like this, just seen a small half of as foundation with the user, think that the user dislikes this program and content corresponding feature, also be difficult to reflection user's true interest, because probably be that the user is after finishing watching other program, find just that in progress this specific program is also fine and see, so just see this program that is over from this more than half programme content.
In addition, often interrupting, as switching or suspending and watch, also might be because the user is delithted with to watch the reason of this program to cause.In this case, even the user is delithted with to watch this program, under the situation of repeatedly changing platform or time-out, also might be because of the very long one section content that adds up to have watched this program, if only the viewing time length with this very long one section content correspondence serves as according to obtaining interest-degree, also being difficult to reflect very exactly user's interest.
In sum, utilize the user to watch the predetermined total time length of playing of the time span of a specific program and this specific program to come estimating user to the interest-degree of a program and revise the method for files on each of customers at present for the basis, perfect not enough, can not reflect the interests change that the user is real very exactly.
Therefore, the invention provides method and the device of a kind of estimating user,, thereby dynamically upgrade files on each of customers so that estimating user is to the interest-degree of a program more accurately to the interest-degree of a program.
Summary of the invention
One of purpose of the present invention is in order to obtain user's interest-degree more exactly, and the method and apparatus of a kind of estimating user to the interest-degree of a program is provided, and utilizes this interest-degree to upgrade the method and the device of files on each of customers.
Estimating user of the present invention comprises step to the method for the interest-degree of a program in broadcast: monitor this user's the behavior at this program, this program comprises at least two fragments, corresponds respectively to different time weightings; Determine the length and the time weight of the pairing plays clip of this user behavior; Length and time weight with according to described plays clip obtain the interest-degree of this user to this program.
One embodiment of the present of invention are, described user behavior comprises the broadcast interruption times of user to described program, when obtaining this user to the interest-degree of this program, also consider the influence of this interruption times.
In the method for estimating user of the present invention to the interest-degree of program, the different fragments in program can be corresponding to different time weightings.The time weighting corresponding different respectively according at least two fragments, adjust the length of the pairing plays clip of user behavior, no matter the user has watched any plays clip of program in the prior art to avoid, as long as the length of plays clip is identical, show that then the user is identical situation to the hobby of each content characteristic in this program, thereby reduce the inaccurate degree of estimating user interest-degree.
The method of renewal files on each of customers of the present invention comprises step: the monitoring user at a behavior in broadcast, this program comprises at least two fragments, corresponds respectively to different time weightings; Determine the length and the time weight of the pairing plays clip of this user behavior; Length and time weight according to described plays clip obtain the interest-degree of this user to this program; With revise described files on each of customers according to this interest-degree.
A kind of estimating user of the present invention comprises an interactive device, a definite device and an acquisition device to the device of the interest-degree of a program in broadcast.Wherein this interactive device is used to monitor this user's the behavior at this program, and this program comprises at least two fragments, corresponds respectively to different time weightings; Should determine device, be used for determining the length and the time weight of the pairing plays clip of this user behavior; This acquisition device is used for obtaining the interest-degree of this user to this program according to the length of described plays clip and time weight.
The device of renewal files on each of customers of the present invention comprises an interactive device, a definite device, an acquisition device and a correcting device.Wherein this interactive device is used to monitor this user's the behavior at this program, and this program comprises at least two fragments, corresponds respectively to different time weightings.Should determine device, be used for determining the length and the time weight of the pairing plays clip of this user behavior.This obtains device, is used for length and time weight according to described plays clip, obtains the interest-degree of this user to this program.This correcting device is used for revising described files on each of customers according to this interest-degree.
By following description and the claim that reference is carried out in conjunction with the accompanying drawings, other purpose of the present invention and achievement will be conspicuous, and the present invention is also had more comprehensively understanding.
Description of drawings
The present invention carries out detailed explanation by the mode of example with reference to accompanying drawing, wherein:
Fig. 1 is the structural representation of an information recommendation system according to an embodiment of the invention;
Fig. 2 is the curve chart that changes with the playing programs time of time weighting according to an embodiment of the invention;
Fig. 3 is the curve chart that changes with the playing programs time of another time weighting according to an embodiment of the invention;
Fig. 4 be according to an embodiment of the invention estimating user to the interest-degree of a program in broadcast and revise the flow chart of files on each of customers;
Fig. 5 utilizes fuzzy logic to obtain the flow chart of time weighting according to an embodiment of the invention;
Fig. 6 is the fuzzy set curve chart of input variable e1 among Fig. 5;
Fig. 7 is the fuzzy set curve chart of input variable e2 among Fig. 5;
Fig. 8 is output variable α among Fig. 5
iThe fuzzy set curve chart.
In all accompanying drawings, identical reference number is represented similar or identical feature and function.
Embodiment
Fig. 1 is the structural representation of an information recommendation system according to an embodiment of the invention.
This interactive device 103 is used to monitor this user's the behavior at a specific program, and this program comprises at least two fragments, corresponds respectively to different time weightings.
Wherein this user behavior comprise interruption position that the user plays each interruption of above-mentioned program, interruption times, from the corresponding original position of watching to the length of the plays clip this interruption position etc.This watches original position can be 0 minute position of a program, and this interruption position can be the broadcast final position of a program.
Time weighting is meant and is used for adjusting the coefficient that the user watches the length of a plays clip, is used to reduce, raise or do not change the length of this plays clip.This time weighting can watch the interruption times of this program process to determine according to the position and the user of plays clip in the program, also can be by user preset the fixed or program provider time weighting curve that presets and send to the user determine.
At least comprise two fragments in this program, respectively corresponding different time weightings we can say that also to have the time weighting of two fragments in this program at least different.Wherein above-mentioned plays clip is the fragment that the user watched.For example the total time length of a program predetermination broadcast is 120 minutes, and promptly the total length of this program is 120 minutes, and the time weighting of its preceding 60 minutes fragments is 0.5, and back 60 minutes time weighting is 1.5.
Should determine that device 104 was used for determining the length and the time weight of the pairing plays clip of this user behavior.Should determine that device 104 comprised that a position determining means 1042 and a fragment obtain device 1044.
An interruption position when this position determining means 1042 is used for watching aforementioned specific program according to the user is determined the position of corresponding plays clip.
This fragment obtains length and the time weight that device 1044 is used for obtaining according to the position of above-mentioned plays clip this plays clip.
The length of above-mentioned plays clip is often referred to the time span that the user watches aforementioned specific program fragment, can determine by above-mentioned user behavior.
This time weighting is the time weighting of this plays clip---one specific time weighting at this plays clip can be obtained by the position of this plays clip user preset corresponding with it time weighting curve fixed or that program provider presets; Also can obtaining by the mode of fuzzy reasoning, the process of this fuzzy reasoning can utilize interruption times and the interruption position in the user behavior to be input variable, and the pairing time weighting of this plays clip is an output variable, to obtain the time weighting of this plays clip.
This acquisition device comprises a length deriving means 1056, a weighting device 1054 and a comparison means 1052.
Under the situation of not considering interruption times, this acquisition device 105 can be used for obtaining the interest-degree of this user to this specific program according to the length of above-mentioned plays clip and time weight.
This length deriving means 1056 is used to obtain predetermined total time the length---program total length of playing of this specific program; This weighting device 1054 is used for the length of above-mentioned plays clip and time weight are handled, thereby obtain the plays clip length of a weighting, normally the plays clip length that the length and the corresponding time weighting of plays clip multiplied each other and draw described weighting; This comparison means 1052 is used for this program predetermination is play total time length and the plays clip length of weighting compare, thereby obtain the interest-degree of this user to this program.
This interest-degree can be expressed as:
Wherein i represents the content characteristic that comprises in this program, W
DiThe take the opportunity plays clip length of the weighting after the weight of the length that equals plays clip, R
DiRepresent the total time length that this program predetermination is play, θ represents a predetermined threshold value, and is can be by user preset fixed or presetted by program provider and to send to the user again.
Wherein the total time length generally play according to program predetermination of the size of θ value is come surely, and generally is no more than the half the time length that program predetermination is play, and is 120 minutes as the broadcast total time length of a program predetermination, and then can establish θ is 20 minutes.
Under the situation of disconnected number of times under consideration, this obtains device 105 also can obtain the interest-degree of this user to this program according to length and the time weight and the broadcast interruption times of above-mentioned plays clip.That is, above-mentioned weighting device 1054 also is used on the basis of plays clip length of weighting and deducts the influence that interruption times is brought again.
This correcting device 106 is used for revising above-mentioned files on each of customers according to this interest-degree.
This system 100 also comprises a programme information receiving system 101, a recommendation apparatus 102 and a files on each of customers management devices 107.
This programme information receiving system 101 is used for program receiving information and with the corresponding electronic program guide of digital television of program (EPG) etc.
This recommendation apparatus 102 is used for filtering out the program that the user may like according to programme information that is received and files on each of customers, recommends the user with the programme information that the user may be liked with a recommendation information tabular form.
This files on each of customers management devices 107 is used for the leading subscriber archives.This files on each of customers comprises fancy grade and the weight of a user at least one content characteristic usually.
The time weighting that this user preset is fixed or program provider presets can be the curve of time weighting with this playing programs time variation.Be illustrated with Fig. 2 and two different embodiment of Fig. 3 below.
Fig. 2 is the curve chart that the time weighting of one embodiment of the invention changed with the playing programs time.Be the curve chart of the time weighting of a film A with the reproduction time variation shown in the figure, this film A comprises actor A.In the figure, time weighting is a discrete type curve with the curve that reproduction time changes.
Abscissa X (R among this figure
D) time that expression film A plays, the total time length of the predetermined broadcast of this film A is 120 minutes, promptly total length is 120 minutes, ordinate Y express time weight.
The figure shows this film A and be divided into four fragments, the time weighting difference of each fragment.The time weighting of 0a fragment is 0.2, and fragment length is 30-0=30 minute; The time weighting of ab fragment is 0.6, and fragment length is 60-30=30 minute; The time weighting of bc fragment is 1.2, and fragment length is 90-30=30 minute; The time weighting of cd fragment is 2, and fragment length is 120-90=30 minute.
Pairing time weighting of each fragment and corresponding fragment length are multiplied each other, just draw the time span of a respective weight.For example the time span of the weighting of 0a fragment is 0.2*30=6 minute; The time span of ab fragment weighting is 0.6*30=18 minute; The time span of bc fragment weighting is 1.2*30=36 minute; The time span of cd fragment weighting is 2*30=60 minute.
The fragment length sum of the weighting of above-mentioned each fragment is 6+18+36+60=120 minute, equals the predetermined broadcast total time length of film A.The time span sum that is to say the weighting of next all fragment of program of normal conditions equals the total time length that this program presets broadcast.
If the user has only seen ab fragment and cd fragment among the film A, promptly the user has only watched plays clip ab and plays clip cd, then the user to watch the plays clip length of the weighting of film A be 18+60=78 minute.
Wherein a, c be respectively the user watch film A two plays clip watch original position, lay respectively at and located and located in 90 minutes in 30 minutes; B, d are respectively the interruption position of above-mentioned two plays clip, lay respectively to locate and located in 120 minutes in 60 minutes, and wherein b is the 1st interruptions, and d is the 2nd interruptions.
Hence one can see that, and the length of plays clip and corresponding time weighting all can be determined by corresponding original position and the interruption position of watching, and 0 minute position of program also can be used as and watch original position, and 120 minutes positions of the terminal point of program also can be used as interruption position.
Certainly, it only is an example that above-mentioned fragment is divided, user's fragment of watching in fact, i.e. and plays clip can start from any time point of a program, terminates in another time point after a while in this program.For example, the user begins to watch this film A from the intermediate point of ab fragment, interrupts when watching the mid point of cd fragment, and then ab fragment mid point is the plays clip that the user watches this film A to the fragment between the cd fragment mid point.
After drawing the plays clip length of weighting, just can with the plays clip length of this weighting and obtain the interest-degree of user to film A.
Fig. 3 be another embodiment of the present invention the time weight with the curve chart of playing programs time to change.If be the curve chart of the time weighting of a film B with the reproduction time variation shown in the figure, this film B comprises performer third.Among this figure, time weighting is the continuous type curve with the curve that reproduction time changes.
Abscissa X (R among this figure
D) time span that expression film B plays, the predetermined total time length of playing of this film B is 120 minutes, promptly total length is 120 minutes, ordinate Y express time weight.The figure shows the passing backward along with this playing programs time, its corresponding time weighting is constantly to change, and time weighting and the proportional Y=X/60 of reproduction time.
The figure shows this film B and can be divided into five fragments, 0C fragment, CD fragment, DG fragment, GH fragment and HK fragment, and the time weighting difference of each fragment.Wherein,
The 0C fragment: fragment length is 12 minutes;
The fragment length of weighting is 12*0.2*0.5=1.2 minute;
Time weighting is 1.2/12=0.1;
The CD fragment: fragment length is 24-12=12 minute;
The fragment length of weighting is (0.2+0.4) * 12/2=3.6 minute;
Time weighting is 3.6/12=0.3;
The DG fragment: fragment length is 60-24=36 minute;
The fragment length of weighting is (0.4+1) * 36/2=25.2 minute;
Time weighting is 25.2/36=0.7;
The GH fragment: fragment length is 102-60=42 minute;
The fragment length of weighting is (1+1.7) * 42/2=56.7 minute;
Time weighting is 56.7/42=1.35;
The HK fragment: fragment length is 120-102=18 minute;
The fragment length of weighting is (1.7+2) * 18/2=33.3 minute;
Time weighting is 33.3/18=1.85.
Certainly, it only is an example that above-mentioned fragment is divided, user's fragment of watching in fact, i.e. and plays clip can start from any time point of a program, terminates in another time point after a while in this program.For example, the user begins to watch this film B from the intermediate point of CD fragment, interrupts when watching the mid point of GH fragment, then is the plays clip that the user watches this film A from CD fragment mid point to the fragment the GH fragment mid point.
From the above, each fragment and the area that corresponding curve is surrounded are the fragment length of the weighting of this fragment, again with this area divided by fragment length, then can draw corresponding time weighting, weight average time that this corresponding time weighting is this fragment.
The fragment length sum of the weighting of above-mentioned each fragment is 1.2+3.6+25.2+56.7+33.3=120 minute, equals predetermined broadcast total time the length---total length of film B.The time span sum that is to say the weighting of next all fragment of program of normal conditions equals the total time length that this program presets broadcast.
If the user has only seen CD fragment and HK fragment among the film B, promptly the user has only watched plays clip CD and plays clip HK, then the user to watch the plays clip length of the weighting of film B be 3.6+33.3=36.9 minute.
Wherein C, H be respectively the user watch film B two plays clip watch original position, lay respectively at and located and located in 102 minutes in 12 minutes; D, K are respectively the interruption position of above-mentioned two plays clip, lay respectively to locate and located in 120 minutes in 24 minutes, and wherein D is the 1st interruptions, and K is the 2nd interruptions.
The length of above-mentioned plays clip and corresponding time weighting all can be determined by corresponding original position and the interruption position of watching, and 0 minute position of program also can be used as and watch original position, and 120 minutes positions of the terminal point of program also can be used as interruption position.
After drawing the plays clip length of weighting, just can with the plays clip length of this weighting and obtain the interest-degree of user to film A.
Certainly, as long as have the time weighting difference of two fragments at least, time weighting can also be other different shape with the curve that program predetermination reproduction time changes, and the length of the plays clip of its respective weight can surround regional area by a plays clip corresponding time weighting curve of institute and X-axis and obtain.But each fragment time corresponding weighting curve of program and this X-axis surround the area in zone and equal the predetermined broadcast total time length of respective program.
Fig. 4 be according to an embodiment of the invention estimating user to the interest-degree of a program in broadcast and revise the flow chart of files on each of customers.
At first, set up a files on each of customers (step S410).This files on each of customers comprises fancy grade and the weight of a user at least one content characteristic.Certainly, if this files on each of customers exists, then can save this step.
Wherein above-mentioned content characteristic is meant the performer that comprises in the program (Fan Bingbing, Pueraria lobota excellent), program category (literary film, romance movie, horror film etc.), director (Zhang Yimou, Feng Xiaogang etc.) etc., these content characteristics can be from information sources such as broadcasting, TV or internet, the most representatively sends to the user by electronic program guide of digital television (EPG) with program exactly.
Content characteristic in the files on each of customers can be one, as has only certain performer.Certainly, the content characteristic in the files on each of customers also can have a plurality of, and at this moment, corresponding recommendation results will be more accurate.
This fancy grade is meant the sensation of user to each content characteristic, can preset a number range by user or supplier, as [50 ,+50].
This weight is meant the user when selecting program, various dissimilar content characteristics, as, performer, director, program category etc. are to the influence of its selection result.What also we can say the user when the program of selecting oneself to like, be standard normally with, promptly still directs according to performer or program category and wait the program of selecting it to like.Wherein all performers' weight can be the same, and the weight of all program categories can be the same, and all directors' weight also can be the same.This weight also can preset a number range by supplier, as [0,100].
Above-mentioned files on each of customers can be filled in and carried out initialization by user oneself.Certainly, this is not unique mode, can obtain files on each of customers by other approach, as by production firm this commending system being carried out the files on each of customers initialization according to user basic information (as sex, the age) yet.
In above-mentioned files on each of customers, comprise a series of content characteristic.Each content characteristic comprises a ternary array (content characteristic term, favorable rating Like-Degree, weights W eight).Like this, this files on each of customers (User Profile, be called for short UP) can be expressed as a ternary array vector (w), if m different content characteristic arranged in files on each of customers, it can be represented with following this vector array for t, ld:
UP＝((t
1，ld
1，w
1)，(t
2，ld
2，w
2)，....(t
i，ld
i，w
i).....，(t
m，ld
m，w
m)) (1)
Here t
iBe a content characteristic, i is content characteristic t
iSequence number, ld
iBe for this content characteristic t
iFavorable rating, w
iBe content characteristic t
iWeight.
Suppose to have a files on each of customers, files on each of customers is such after its initialization:
Program category: weight=90
Film like-degree=50
Opera like-degree=30
News like-degree=-20, wherein negative number representation is disliked degree,
Then, the user interest ternary array of content characteristic is (film, 50,90) in the above-mentioned program category;
Performer: weight=80
First like-degree=10
Second like-degree=45
Third like-degree=-12.5;
Then, the user is (first, 10,80) to the user interest ternary array of first among the above-mentioned performer;
......
Secondly, monitor the behavior of a user at a program in broadcast, this program comprises at least two fragments, corresponds respectively to different time weighting (step S420).
This user behavior comprises interruption position that the user plays each interruption of above-mentioned program, interruption times, from the corresponding original position of watching to the length of the plays clip between the interruption position etc., 0 minute position that this original position can above-mentioned program wherein, this interruption position can be the terminal point play position of above-mentioned program.This interruption times is because the user switches to other program, time-out and termination when watching a specific program watch etc. caused.
At least comprise two fragments in this program, respectively corresponding different time weightings we can say that also to have the time weighting of two fragments in this program at least different.Wherein above-mentioned plays clip is the fragment that the user watched.
For example aforementioned film A comprises four fragments, the time weighting difference of each fragment, and the time weighting of plays clip ab wherein and plays clip cd is respectively 0.6 and 2.
Afterwards, determine the length and the time weight (step S430) of the pairing plays clip of this user behavior.
Usually according to the interruption position in this user behavior and watch original position to determine the position of this plays clip and the length and the corresponding time weighting of definite this plays clip.
As plays clip ab and the plays clip cd of film A, watch original position be respectively a place 30 minutes, c place 90 minutes, interrupting device is respectively the b place 60 minutes, d place 120 minutes.Wherein, b is the 1st interruptions, and d is the 2nd interruptions.
The time weighting of plays clip ab is 0.6, and the length of plays clip is 60-30=30 minute; The time weighting of plays clip is 2, and the length of plays clip is 120-90=30 minute.
Certainly, time weighting also can be determined by fuzzy logic inference by interruption times and interruption position.
Then, according to the length and the time weight of this plays clip, obtain the interest-degree (step S440) of user to this program.
Usually can be according to formula
Obtain interest-degree.For this reason, the first step must be obtained the total time length that a program predetermination is play, i.e. program total length R
DiSecond step, the plays clip length W of acquisition weighting
Di, can obtain by the length and the corresponding time weighting of above-mentioned plays clip; In the 3rd step, aforementioned two values and the above-mentioned formula of preset threshold value θ substitution are obtained interest-degree.
For example, the total time length of the predetermined broadcast of above-mentioned film A is 120 minutes, and promptly the total length of this film A is R
Di=120 minutes.
Wherein length that the length of the weighting plays clip of plays clip ab and plays clip cd can be by its plays clip and corresponding time weighting multiply each other and obtain: the length of the plays clip of the weighting of plays clip ab is 0.6*30=18 minute; The plays clip length of the weighting of plays clip cd is 2*30=60 minute.And the user watches the plays clip length and the W of the weighting of film A to this
Di=18+60=78 minute.
If predetermined threshold value θ is 20 minutes, respectively with the above-mentioned formula of each data substitution of above-mentioned plays clip ab and plays clip cd
Then can draw the interest-degree of user to this film A.
Because the user is normally often interrupted because of disliking this program, so during disconnected information under consideration, as interruption times or interruption position etc., deduct a suitable value more also can for the plays clip length of weighting, promptly can be according to length, the time weighting of this plays clip, and interruption times or interruption position, obtain the interest-degree of this user to a program.In this case, after according to interruption times the plays clip length of this weighting further being handled, the plays clip length with the weighting after this processing obtains the interest-degree of user to this program again.
Usually can be with the plays clip length W of this weighting
DiDeduct respective interrupt time span and a product that interrupts coefficient N, establishing this interruption coefficient is 0.05, and this interruption coefficient also can be presetted by program provider, and its magnitude range is generally a very little positive decimal, as [0.0001,0.1].
Still two plays clip with above-mentioned film A are example, and corresponding break period, length was respectively length=30 minute and 0 minute (because d place is the terminal point of film A) of fragment bc.The plays clip length W of weighting then
Di=78-(30*0.05)-(0*0.05)=76.5 minute.
Correspondingly, can draw the influence that deducts the interruption situation after, the user to the interest-degree of film A is:
In addition, if the plays clip length of the last weighting that obtains and less than predetermined threshold value θ, then establishing interest-degree is 0; If the time value of last interruptions is less than predetermined threshold value θ, then also establishing interest-degree is 0.
At last, revise this files on each of customers (step S450) according to this interest-degree.Revising this files on each of customers, mainly is to revise in the files on each of customers the fancy grade of content characteristic that should program and the value of weight, can carry out according to following formula usually:
Here, t (Term) represents content characteristic, and i represents a sequence number of content characteristic, i.e. content characteristic i, Weight
TiThe initial weight of expression content characteristic i, Like_degree
iThe expression user is to the initial fancy grade of content characteristic i.Weight '
TiThe weight of content characteristic i after expression changes, Like_degree '
iUser preferences degree after expression changes.WD
iBe the actual time span of watching this program that comprises content characteristic i of this user, RD
iBe the predetermined total time length of playing of this program itself, θ is a predetermined threshold value.
α
tAnd β
iBe respectively weight and adjust coefficient and fancy grade adjustment coefficient, can be for a constant, as α
tValue can be [0.1,1.0], β
iAlso can be [0.1,1.0].Two coefficients are used for adjusting the interest-degree of the affiliated type weight of content characteristic i respectively
And the user is to the interest-degree in the content characteristic i fancy grade
α
tAnd β
iBe generally used for delaying the variation of weight and favorable rating, can also can preset by user preset by supplier.Because the weight of user preferences is relatively stable, so α
t≤ β
i, and same type α
tUsually identical.
For a files on each of customers, suppose that the scope of Weight is [0,100], and the scope of Like-degree is [50,50]
If Weight '
iGreater than 1, make Weight '
i=100
If Weight '
iLess than 0, make Weight '
i=0;
If Like-degree '
iGreater than 0.5, make Like-degree '
i=50
If Like-degree '
iLess than-0.5, make Like-degree '
i=-50.
For example, can obtain actor A weight and fancy grade among the film A with above-mentioned formula.In the files on each of customers, the user is 10 for the fancy grade of actor A, and the user is 80 to performer's weight, establishes α
tBe 0.1, and β
iBe 0.5, and do not consider interrupting information, interruption times or interruption position etc. as the aforementioned, fancy grade and weight after then changing are:
This estimating user of present embodiment is to the interest-degree of program and upgrade the method and the device of files on each of customers, the time weighting corresponding different respectively according at least two fragments, adjust the length of the pairing plays clip of user behavior, no matter the user has watched any plays clip of program in the prior art to avoid, as long as the length of plays clip is identical, show that then the user is identical situation to the hobby of each content characteristic in this program, thereby reduce the inaccurate degree of estimating user interest-degree.
Simultaneously, present embodiment has also been considered the situation that the user is interrupted, may be because the user is not delithted with program, thus this enforcements also repeatedly interrupting the one side of length of situation as the adjustment plays clip, thereby can obtain the interest-degree of user's correspondence program more accurately.
Fig. 5 utilizes fuzzy logic to obtain the flow chart of time weighting according to an embodiment of the invention.In the present embodiment, mainly be the interrupting information when repeatedly interrupting, the position of the plays clip that comprehensive user watches and interruption times obtain time weighting to the influence of user preferences.Among this embodiment, be input variable, obtain the time weighting of each plays clip by fuzzy reasoning with interruption times and interruption position.
The first step obtains interruption times and interruption position (step S510).If interrupted N-1 time, then the user has watched N plays clip, suppose to have obtained i plays clip (t1, initial viewing location t2) is t1, interruption position is t2, then Dui Ying interruption times is i.
Second step was an input variable with number of times i and interruption position t2, and was output variable with the user to the time weighting of i section plays clip, set up input more than one and single variable relation of exporting (step S520).If interruption times e1=i; E2=t2; α
iThe time weighting of=the i plays clip.
In the 3rd step, draw input variable e1 after the obfuscation and the fuzzy curve chart (step S530) of e2.This step can concern by existing fuzzy logic inference and obtains.Both fuzzy reasoning curve charts are respectively as Figure 6 and Figure 7.The fuzzy set curve chart of Fig. 6 input variable e1, the fuzzy set curve chart of Fig. 7 input variable e2.The degree of membership of μ value representation e1 and e2 wherein.
The 4th step, utilize the fuzzy curve chart of two input variables to carry out fuzzy reasoning, with fuzzy curve chart and the fuzzy value (step S540) that draws output variable, that is, and the output variable---time weighting α that draws by fuzzy reasoning
iFuzzy curve chart and fuzzy value.Fig. 8 is output variable α
iFuzzy curve chart, the degree of membership μ of this output variable wherein be according to e1 among Fig. 6 and Fig. 7 and e2 degree of membership obtain by fuzzy reasoning.
According to the big α of e1
iLittle, reach greatly then α of e2
iBig principle is carried out fuzzy logic inference, determines the time weighting α of user to this plays clip i
iThe fuzzy value size.Thus, it is as follows to draw concrete fuzzy inference rule:
If I. e1 is " greatly ", and e2 is " greatly ", α so
iBe " little ";
If II. e1 is " greatly ", and e2 is " medium ", α so
iBe " less ";
If III. e1 is " greatly ", and e2 is " little ", α so
iBe " minimum ";
If IV. e1 is " medium ", and e2 is " greatly ", α so
iBe " medium ";
If V. e1 is " medium ", and e2 also is " medium ", α so
iBe " little ";
If VI. e1 is " medium ", and e2 is " little ", α so
iBe " less ";
If VII. e1 is " little ", and e2 is " greatly ", α so
iBe " greatly ";
If VIII. e1 is " little ", and e2 is " medium ", α so
iBe " medium ";
If IX. e1 is " little ", and e2 is " little ", α so
iBe " little ".
The 5th step, the fuzzy value that is obtained is made clear, to draw the clearly value of output variable, i.e. the time weighting α of this plays clip i
iClearly value (step S550).
Be convenient to understand for final result, the result of fuzzy reasoning must convert clearly value to.Modal deblurring algorithm is area gravity model appoach and maximum average value method.The former synthesizes the rule of all excitation outputs as a result of, is applicable to level and smooth control, is process control method commonly used.
Present embodiment adopts area center of gravity deblurring algorithm, and this algorithm as shown in Equation (4).
Here, μ [l]: the height of output area is inferred in expression from l rule;
Yl: the abscissa of the center of gravity of output area is inferred in expression from l rule;
P: the quantity of the rule of inference that expression is satisfied.
Utilize above-mentioned formula, can obtain α
iClearly value, detailed process can be with reference to No. the 200310123354.7th, Chinese patent application (the application people be a Koninklijke Philips Electronics N.V, and the invention people laughs at fear for executing, and the applying date is on December 15th, 2003).
Then, will pass through the time weighting of the plays clip that obtained, obtain the plays clip length of the weighting of this plays clip, with further acquisition user's interest-degree.
Though through the present invention is described in conjunction with specific embodiments, for the skilled personage in present technique field, will be conspicuous according to manyly substituting of making of narration above, modifications and variations.Therefore, when such substituting, in the spirit and scope that modifications and variations fall into attached claim the time, should being included among the present invention.
Claims (14)
1. an estimating user comprises step to the method for the interest-degree of a program in broadcast:
A. monitor this user's the behavior at this program, this program comprises at least two fragments, corresponds respectively to different time weightings;
B. determine the length and the time weight of the pairing plays clip of this user behavior; With
C. according to the length and the time weight of described plays clip, obtain the interest-degree of this user to this program.
2. the method for claim 1, wherein said time weighting comes from user's setting.
3. the method for claim 1, wherein said time weighting comes from the setting of program provider.
4. the method for claim 1, wherein said user behavior comprise the interruption position that the user plays each interruption of described program, and wherein step (b) comprising:
Determine the position of described plays clip according to described interruption position;
Obtain the length and the time weight of this plays clip according to the position of described plays clip.
5. the method for claim 1, wherein said user behavior comprises the broadcast interruption times of user to described program, wherein step (c) comprises length and time weight and this broadcast interruption times according to described plays clip, obtains the interest-degree of this user to this program.
6. the method for claim 1, wherein step (c) comprising:
(i) obtain the total length of described program;
(ii) the length and the time weight of described plays clip are handled, thereby obtained the plays clip length of a weighting;
(iii) the plays clip length to this total length and weighting compares, thereby obtains the interest-degree of this user to this program.
7. the method for claim 1, wherein said user behavior comprises broadcast interruption times and the interruption position of user to described program, step (b) comprises that with the interruption times of described plays clip correspondence and interruption position be input variable, obtains the pairing time weighting of described plays clip with the method for fuzzy reasoning.
8. the method for claim 1, wherein the time weighting of at least one fragment is a particular value, is used to reduce described interest-degree.
9. method of upgrading files on each of customers comprises step:
A. monitor a user's the behavior at a program in broadcast, this program comprises at least two fragments, corresponds respectively to different time weightings;
B. determine the length and the time weight of the pairing plays clip of this user behavior; With
C. according to the length and the time weight of described plays clip, obtain the interest-degree of this user to this program; With
D. revise described files on each of customers according to this interest-degree.
10. an estimating user comprises the device of the interest-degree of a program in broadcast:
An interactive device is used to monitor the behavior of this user at this program, and this program comprises at least two fragments, corresponds respectively to different time weightings;
A definite device is used for determining the length and the time weight of the pairing plays clip of this user behavior; With
One obtains device, is used for length and time weight according to described plays clip, obtains the interest-degree of this user to this program.
11. device as claimed in claim 10, wherein said user behavior comprise the interruption position that the user plays each interruption of described program, this determines that device comprises:
A position determining means is used for determining according to described interruption position the position of described plays clip; With
A fragment obtains device, is used for obtaining according to the position of described plays clip the length and the time weight of this plays clip.
12. device as claimed in claim 10, wherein said user behavior comprises the broadcast interruption times of user to described program, wherein said acquisition device also is used for obtaining the interest-degree of this user to this program according to the length of described plays clip and time weight and this broadcast interruption times.
13. device as claimed in claim 10, wherein this acquisition device comprises:
A length deriving means is used for obtaining the total length of described program;
A weighted device is used for the length and the time weight of described plays clip are handled, thereby obtains the plays clip length of a weighting; With
A comparison means is used for the plays clip length of this total length and weighting is compared, thereby obtains the interest-degree of this user to this program.
14. a device that upgrades files on each of customers comprises:
An interactive device is used for monitoring this user's the behavior at the program of a broadcast, and this program comprises at least two fragments, corresponds respectively to different time weightings;
A definite device is used for determining the length and the time weight of the pairing plays clip of this user behavior;
One obtains device, is used for length and time weight according to described plays clip, obtains the interest-degree of this user to this program; With
A correcting device is used for revising described files on each of customers according to this interest-degree.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 200580041044 CN101069424A (en) | 2004-11-30 | 2005-11-28 | Apparatus and method for estimating user interest degree of a program |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200410100111.6 | 2004-11-30 | ||
| CN200410100111 | 2004-11-30 | ||
| CN 200580041044 CN101069424A (en) | 2004-11-30 | 2005-11-28 | Apparatus and method for estimating user interest degree of a program |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101069424A true CN101069424A (en) | 2007-11-07 |
Family
ID=38880975
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 200580041044 Pending CN101069424A (en) | 2004-11-30 | 2005-11-28 | Apparatus and method for estimating user interest degree of a program |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101069424A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101867752A (en) * | 2010-06-13 | 2010-10-20 | 宇龙计算机通信科技(深圳)有限公司 | Method, device and terminal for playing television programs |
| CN108133036A (en) * | 2018-01-09 | 2018-06-08 | 北京奇艺世纪科技有限公司 | A kind of weighted value of user interest point determines method, apparatus and electronic equipment |
| CN108307153A (en) * | 2017-01-13 | 2018-07-20 | 佳能株式会社 | Video monitoring devices and its control method, storage medium and video monitoring system |
| CN110446093A (en) * | 2019-08-15 | 2019-11-12 | 天脉聚源(杭州)传媒科技有限公司 | A kind of video progress bar display methods, device and storage medium |
| CN117708373A (en) * | 2024-02-06 | 2024-03-15 | 成都数之联科技股份有限公司 | Animation interaction method and system |
-
2005
- 2005-11-28 CN CN 200580041044 patent/CN101069424A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101867752A (en) * | 2010-06-13 | 2010-10-20 | 宇龙计算机通信科技(深圳)有限公司 | Method, device and terminal for playing television programs |
| CN101867752B (en) * | 2010-06-13 | 2012-09-12 | 宇龙计算机通信科技(深圳)有限公司 | Method, device and terminal for playing television programs |
| CN108307153A (en) * | 2017-01-13 | 2018-07-20 | 佳能株式会社 | Video monitoring devices and its control method, storage medium and video monitoring system |
| US11048947B2 (en) | 2017-01-13 | 2021-06-29 | Canon Kabushiki Kaisha | Video monitoring apparatus, method of controlling the same, computer-readable storage medium, and video monitoring system |
| US11830251B2 (en) | 2017-01-13 | 2023-11-28 | Canon Kabushiki Kaisha | Video monitoring apparatus, method of controlling the same, computer-readable storage medium, and video monitoring system |
| CN108133036A (en) * | 2018-01-09 | 2018-06-08 | 北京奇艺世纪科技有限公司 | A kind of weighted value of user interest point determines method, apparatus and electronic equipment |
| CN110446093A (en) * | 2019-08-15 | 2019-11-12 | 天脉聚源(杭州)传媒科技有限公司 | A kind of video progress bar display methods, device and storage medium |
| CN117708373A (en) * | 2024-02-06 | 2024-03-15 | 成都数之联科技股份有限公司 | Animation interaction method and system |
| CN117708373B (en) * | 2024-02-06 | 2024-04-05 | 成都数之联科技股份有限公司 | Animation interaction method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1122235C (en) | Media information recommending apparatus | |
| CN1579092A (en) | Information processing system, information processing device and method, recording medium, and program | |
| CN1264340C (en) | Broadcasting reception apparatus and data broadcasting method | |
| CN101061713A (en) | Method and device for updating user profile | |
| CN1282934C (en) | Information processing device and method, content distribution device and method and computer program | |
| CN100348021C (en) | Information processing apparatus | |
| CN1550101A (en) | Broadcasting receiver | |
| CN1586071A (en) | Automatic content recording device and method, recording medium, and program | |
| CN101047827A (en) | Electronic apparatus, method and system for collecting broadcast program information, and storage medium | |
| CN1795678A (en) | Content notification and delivery | |
| CN1698353A (en) | Information processing apparatus, information processing method, and computer program | |
| CN1574972A (en) | Fast start-up for digital video streams | |
| CN1410913A (en) | Information processing system, output, processor and method, recording medium and program | |
| CN1992890A (en) | Apparatus and method for providing media program | |
| CN1242609C (en) | Television receiver and for counting the number of accesses from the receiver | |
| CN1902911A (en) | Program recommendation device, program recommendation method of program recommendation device, and computer program | |
| CN1901638A (en) | Apparatus, method and program for information processing | |
| CN1757020A (en) | Information processing device, information processing method, and computer program | |
| CN1595985A (en) | Content supplying apparatus and method, and recording medium | |
| CN1629884A (en) | Information recommendation system and method | |
| CN1713701A (en) | Information processing apparatus, information processing method, and computer program | |
| CN1898950A (en) | Recommended program notification method and recommended program notification device | |
| CN1791205A (en) | Data recording controller, method of controlling data recording controller, program, data recording system, and data recording device | |
| CN1756337A (en) | Method, apparatus and program for recording and playing back content data | |
| CN1286017C (en) | Recording/reproduction device, recording management method, recording medium, and program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| ASS | Succession or assignment of patent right |
Owner name: PACE MICRO TECHNOLOGY CO., LTD. Free format text: FORMER OWNER: KONINKLIJKE PHILIPS ELECTRONICS N.V. Effective date: 20080808 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20080808 Address after: West Yorkshire Applicant after: Koninkl Philips Electronics NV Address before: Holland Ian Deho Finn Applicant before: Koninklijke Philips Electronics N.V. |
|
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |