CN101874255A - Method and apparatus for estimating total interest of a group of users directing to a content - Google Patents
Method and apparatus for estimating total interest of a group of users directing to a content Download PDFInfo
- Publication number
- CN101874255A CN101874255A CN200680018627A CN200680018627A CN101874255A CN 101874255 A CN101874255 A CN 101874255A CN 200680018627 A CN200680018627 A CN 200680018627A CN 200680018627 A CN200680018627 A CN 200680018627A CN 101874255 A CN101874255 A CN 101874255A
- Authority
- CN
- China
- Prior art keywords
- value
- user
- degree
- relative importance
- group
- 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
- 238000000034 method Methods 0.000 title claims abstract description 58
- 238000004590 computer program Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000006870 function Effects 0.000 description 15
- 230000008569 process Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 7
- 230000036962 time dependent Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000011218 segmentation Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000003203 everyday effect Effects 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 240000004859 Gamochaeta purpurea Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000003556 assay Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 238000013481 data capture Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000007620 mathematical function Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000014860 sensory perception of taste Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Finance (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The present invention provides a method for estimating total interest of a group of users directing to a content, wherein the group of users includes at least two members, and the method comprising the steps of: obtaining information relating to available time slice of the content and features contained in the content; obtaining value of each member's priority in different time slice and preference of the member directing to the contained features; and determining the total interest according to the value of each member's priority in the available time slice and preference of the member directing to the contained features. In the present invention, a way may be used in which priority of each member of the group varies from time to time to adapt to the many states where each group member's influence for content selection is dynamic, so as to recommend content to a group of users more correctly.
Description
Technical field
The present invention relates to the information processing technology, particularly a kind of method and apparatus of estimating user's group to the total interest degree of a content.
Background technology
Along with the development and the development of technology of society, the knowledge that the mankind obtain is just with the rate increase of geometric series.In the face of vast as the open sea information, pressing for provides and can carry out the method and apparatus of rough handling automatically to information, thereby people are freed from daily trival matters, more effectively is engaged in performing creative labour and enjoys life better.
Commending contents is quite popular technology in the information processing, thereby its ultimate principle is for to obtain the interest level (following again be called interest-degree) of specific user/customer group to content according to predefined decision condition to content analysis, and whether decision recommends this content to this user/customer group by this then.This technology has broad application prospects in television program recommendations, and therefore many research work all with this as a setting.
The object of commending contents can be a unique user, also can be the user's group by at least two users or member composition, and for example user's group can be made up of the kinsfolk, perhaps forms by staying in a classmate in the bedroom.Compare with unique user, user's group is owing to the problem that influences each other that relates between the user group membership, so the process more complicated of interest-degree assay.
The U.S. Patent application of submitting in March 28 calendar year 2001 09/819 that is entitled as " method and apparatus (Method And Apparatus For Generating Recommendations For APlurality Of Users) that produces multi-user's recommendation results ", in 440, a kind of method of determining to organize to the user recommendation results is disclosed, the invention people of this patented claim is LalithaAgnihotri and Srinivas Gutta, and the people that assigns is Philips Electronics NorthAmerica Corp..In above-mentioned application documents disclosed embodiment, recommendation results must assign to represent that to recommend score or combined recommendation each user calculated for the fancy grade of the feature that information or content comprised in this recommendation score or combined recommendation score were organized according to the user.In this mode, add the content that U.S. Patent application 09/819,440 discloses to insert.
In a specific embodiments of U.S. Patent application 09/819,440, recommend score to determine that method is: at first to obtain electronic program guides (EPG), also promptly have television schedule information with electronic form.Then, obtain user group member's archives 300, these archives have promptly comprised the information of each user for the fancy grade of each feature.Then, if necessary, each user is converted to same numerical measure for the fancy grade of each feature.Treatment step subsequently is to determine that according to aforementioned files on each of customers a program is to recommendable degree of each user or recommendation score.After the recommendation score of having obtained all users in the user organizes, all users' recommendation score to be combined to form total recommendation score of this user's group, array mode for example comprises gets weighted mean value or arithmetic mean to all users' recommendation score.At last, export the recommendation score of the user's group that calculates.
In order to determine that better user's group is the total interest degree to the interest level of content generally, can also include more factors in limit of consideration.
Summary of the invention
An object of the present invention is to determine exactly the total interest degree of user's group, thereby organize content recommendation more targetedly to the user to content.
One aspect of the present invention provides a kind of method of estimating user's group to the total interest degree of a content, wherein this user group comprises at least two members, and the method comprising the steps of: obtain relate to this content can be with the information of contained feature in period and this content; Obtain each member's the value of relative importance value on the different periods and the fancy grade of this member to this contained feature; The value of relative importance value on this available period and this member according to each member determine this total interest degree to the fancy grade of this contained feature.
In one embodiment, described each member's relative importance value value is set to: one of them described member's relative importance value is the value difference at least two periods, these two free a plurality of periods that were divided in one day of period.Perhaps, described each member's relative importance value value is set to: at least two same periods of same date not, a described member's relative importance value value difference.
In another embodiment, determining step may further comprise the steps: according to each member's the relative importance value value on the different periods with the fancy grade of contained feature is determined interest-degree the value on different periods of this user's group to this contained feature; Determine the interest-degree value of this user's group according to the described available period; And the interest-degree value that makes up this user group is to obtain described total interest degree.
Another aspect of the present invention provides a kind ofly organizes the method for content recommendation to a user, and wherein this user's group comprises at least two members, and the method comprising the steps of: obtain relate to this content can be with the information of contained feature in period and this content; Obtain each member's the value of relative importance value on the different periods and the fancy grade of this member to this contained feature; The value of relative importance value on this available period and this member according to each member determine this total interest degree to the fancy grade of this contained feature; Recommend to this user's group according to this total interest degree.
Another aspect of the present invention provides a kind of equipment of estimating user's group to the total interest degree of a content, wherein this user's group comprises at least two members, this equipment comprises: deriving means, be used for obtaining relate to this content can be with the information of period and the contained feature of this content; Receiving trap, the relative importance value that is used to receive each member on the different periods value and this member to the fancy grade of this contained feature; Determine device, be used for the fancy grade of this contained feature being determined this total interest degree according to the value of relative importance value on this available period and this member of each member.
In one embodiment, described each member's relative importance value value is set to: a described member's relative importance value is the value difference at least two periods, these two free a plurality of periods that were divided in one day of period.Perhaps, described each member's relative importance value value is set to: at least two same periods of same date not, a described member's relative importance value value difference.In another embodiment, described definite device comprises: generating apparatus, be used for according to each member's the relative importance value value on the different periods with the fancy grade of contained feature is determined interest-degree the value on different periods of this user's group to this contained feature; Search device, be used for determining the interest-degree value of this user's group according to the described available period; Calculation element is used to make up the interest-degree value of this user's group to obtain described total interest degree.
Another aspect of the present invention provides a kind ofly organizes the equipment of content recommendation to a user, and wherein this user's group comprises at least two members, comprising: deriving means, be used for obtaining relate to this content can be with the information of period and the contained feature of this content; Receiving trap, the relative importance value that is used to receive each member on the different periods value and this member to the fancy grade of this contained feature; Determine device, be used for the fancy grade of this contained feature being determined this total interest degree according to the value of relative importance value on this available period and this member of each member; Recommendation apparatus is used for recommending to this user's group according to this total interest degree.
Another aspect of the present invention provides a kind of computer program of user's group to the total interest degree of a content that be used to estimate, wherein this user group comprises at least two members, and this computer program comprises: obtain relate to this content can be with the code of the information of contained feature in period and this content; Obtain each member's the value of relative importance value on the different periods and the code of this member to the fancy grade of this contained feature; The fancy grade of this contained feature is determined the code of this total interest degree according to each member's the value of relative importance value on this available period and this member.
The invention still further relates to a kind of memory carrier that comprises this computer program.
In the present invention, can by make the user organize in each member's the time dependent mode of relative importance value adapt to the user organize in each member be many situations of dynamic change for the influence power of content choice, thereby can organize content recommendation to a user more accurately.
In one embodiment, relative importance value that can the member is set to change with the period segmentation, and this implementation has simply, advantage flexibly, and the interest-degree that is suitable for TV programme is estimated and recommended.And can precompute use when each period, upward user's group was spent to the interest-degree of each feature and for the subsequent calculations total interest on this basis, this can reduce computing cost, thereby simplifies hardware configuration.
By following description and claim that reference is carried out in conjunction with the accompanying drawings, other purposes of the present invention and advantage will be conspicuous, and the present invention is also had more comprehensively understanding.
Description of drawings
Fig. 1 is a process flow diagram, and it shows according to the preferred embodiment of user's group of estimation of the present invention to the method for the total interest degree of a content.
The process flow diagram of Fig. 2 shows one and calculates the user group example to the interest-degree of a feature based on the fuzzy logic processes mode.
Fig. 3 a, 3b and 3c show the subordinate function that uses in the fuzzy logic processes mode shown in Figure 2, and they correspond respectively to fancy grade, relative importance value and user group membership to the interested degree of characteristic attribute.
Fig. 4 is a process flow diagram, and it shows according to another preferred embodiment to the method for the total interest degree of a content of user's group of estimation of the present invention.
Fig. 5 is a process flow diagram, and it shows a preferred embodiment organizing the method for content recommendation to a user according to of the present invention.
Fig. 6 is a block scheme, and it shows according to the embodiment of user's group of estimation of the present invention to the equipment of the total interest degree of a content.
Fig. 7 a is a block scheme, its show according to of the present invention to a user organize content recommendation an embodiment of equipment.
Fig. 7 b is a block scheme, its show according to of the present invention to a user organize content recommendation another embodiment of equipment.
In all accompanying drawings, identical reference number is represented similar or identical feature and function.
Embodiment
Below in conjunction with accompanying drawing better embodiment of the present invention is described in detail.
The process flow diagram of Fig. 1 shows according to the preferred embodiment of user's group of estimation of the present invention to the method for the total interest degree of a content.For example, this content is a TV programme.
As shown in Figure 1, in step S110, obtain user data, comprise that the user organizes fancy grade and each member the relative importance value when selecting TV programme of interior each member to each feature.These user data and other data relevant with the member can be stored in the files on each of customers for calling, reasonablely be, files on each of customers is the data file with certain format, its alter mode for example can be by program provider regular update, or is revised voluntarily as required by the user group membership.Following table 1 and 2 shows the data structure of this two classes user data respectively.
Table 1
| ??A | ????B | ????C | ????D | ????E | ??F | ????G | ??H | ????I | ????J | ????K | |
| The child 1 | ??1.0 | ????0 | ????0.6 | ????0 | ????0.4 | ??0.8 | ????0 | ??0 | ????-????0.6 | ????0 | ????0.9 |
| The child 2 | ??0.9 | ????0 | ????0 | ????0.6 | ????0 | ??-??0.4 | ????0.1 | ??0.5 | ????0 | ????0.8 | ????0 |
| Mother | ??0.6 | ????0 | ????0 | ????0 | ????0.6 | ??0.7 | ????-????0.5 | ??0.2 | ????0.9 | ????0 | ????0 |
| Father | ??0.7 | ????0 | ????0 | ????0 | ????0 | ??0.3 | ????0.8 | ??0.9 | ????0 | ????-????0.8 | ????0.9 |
Table 2
| Period | The |
The child 2 | Mother | Father |
| ????18:30-19:30 | ????1 | ????1 | ????0.1 | ????0.6 |
| ????19:30-20:30 | ????1 | ????1 | ????0.4 | ????0.4 |
| ????20:30-21:30 | ????0.4 | ????0.5 | ????0.7 | ????0.3 |
| ????21:30- | ????0.2 | ????0.2 | ????0.4 | ????0.8 |
The obtain manner of user data can be varied, for example: if utilize set-top box to realize the function that the total interest degree calculates, then set-top box can obtain by downloading from the remote server of program provider such as cable TV network or LAN (Local Area Network), and the user data that provides on the remote server can be to be the statistics that sample is obtained with a large amount of families; Perhaps set up user data on their own by input equipment and they are stored in the equipment of the calculated population interest-degree such as set-top box by the kinsfolk; Even can also by be installed near the televisor monitoring equipment (for example camera and telepilot etc.) gather automatically the kinsfolk watch TV situation and with these information deliver to set-top box or through Network Transmission to remote server to form user data by statistical study.
Table 1 comprises that the user organizes the fancy grade data of interior each member to each feature, this user's group is by father, mother, child 1 and child 2 form, the TV programme of watching comprises A～K and amounts to 11 features, wherein A represents that this program is an english programs or based on the program of English, B, C and D are respectively and computing machine, the program that mathematics is relevant with chemistry, it is the film with romantic sentiment that E represents this program, F and G represent that respectively this program is cartoon and serial, H represents news program, the advertisement that the I representative is intercutted, J represents that this program has the comedy color, and K represents sports cast.The a certain member of numeric representation in the table is to the fancy grade of a certain feature, that is, also adopt in the present embodiment numerical value describe or the scale user organize in each member for the fancy grade of the feature that content comprised.For example child 1 enjoys a lot to watch cartoon, so its fancy grade assignment for cartoon is 0.8, and since its serial and news are lost interest in, therefore corresponding numerical value is 0, in addition, the also very disagreeable commercial advertisement of child 1, so its fancy grade value to this feature is-0.6.For other kinsfolks, also can make corresponding assignment according to hobby separately.
Table 2 comprises that the user organizes interior each member's relative importance value data.It will be appreciated that by following description to the interest-degree account form the relative importance value here can reflect that a member organizes the influence degree of total interest degree for the user, higher relative importance value means the influence power that this member is stronger to the total interest degree, and vice versa.
As shown in table 2, in this example, the time of TV reception is divided into four periods, the relative importance value of each member in the different periods is not unalterable, for example in 18:30～19:30 period of every day, because time this moment still early, the head of a family allows them to watch TV, therefore child 1 and child's 2 relative importance value all value be 1, and along with the arrival in the late into the night, the possibility that children are sitting in before the televisor is more and more littler, causes relative importance value to reduce gradually, therefore to a certain extent, this time dependent relative importance value can reflect the variation of the agent object of TV reception.And for example, mother is lower than other members (value is 0.1) and the relative importance value of 20:30～21:30 period the highest (value is 0.7) at the relative importance value of 18:30～19:30 period of every day, this is because mother just is being busy with housework in the last period, the TV of at all having no time to watch, therefore its relative importance value can be set very lowly, then a period is mother's spare time, in order to satisfy the eager desire that it televiews, thereby relative importance value can be set to such an extent that very highly guarantee that it has bigger right of speech on program is selected.
In this example, relative importance value value difference in different period in a day, this mode relatively is suitable for the interest-degree of TV programme and estimates and recommendation, in addition, in order to reflect the festivals or holidays factors to the influence of kinsfolk's watching habit, can also make relative importance value that this segmentation changes on ordinary days with get different values festivals or holidays.But it is worthy of note, relative importance value also can adopt other time dependent modes, for example relative importance value can year, month or day for unit value difference with factors such as reflection festivals or holidays and seasons, perhaps relative importance value can also be expressed as a time dependent mathematical function form.
In sum, the combination that is constituted by the relative importance value value that reasonably is provided with on fancy grade and the different time, can when definite total interest is spent, the time dependent factor of the object of received content also be taken into account, thereby help to obtain recommendation results comprehensively and accurately.
Refer again to Fig. 1, after finishing the step S110 that obtains user data, enter step S120, and calculate user's group the interest-degree (being called the feature interest-degree in the present invention again) of each feature in each period.The back will be discussed the concrete mode of calculating this feature interest-degree in detail.
Then enter step S130, create a user and organize this user's group that archives calculate with storage above-mentioned steps S120 to the interest-degree of each feature in each period, table 3 shows an exemplary configurations form of these archives.In table 3, for feature A～K, it has corresponding calculated value separately in four periods, has represented user's group to the interest-degree of one of them feature on each period.Therefore owing to only, all do not marking concrete numerical value in the spaces, and feature A, E and G have just been provided at the calculated value of 20:30-21:30 in the period for the purpose of expression file structure.
Table 3
| Period | ????A | ??B | ??C | ??D | ????E | ??F | ????G | ??H | ??I | ??J | ??K |
| ??18:30-19:30 |
| Period | ????A | ??B | ??C | ??D | ????E | ??F | ????G | ??H | ??I | ??J | ??K |
| ??19:30-20:30 | |||||||||||
| ??20:30-21:30 | ????0.37 | ????0.145 | ????-0.015 | ||||||||
| ??21:30- |
So far, finished the establishment that the user organizes archives.It is worthy of note, user's group depends on the interest-degree of this user's group to contained each feature of program to the total interest degree of a program, therefore in the total interest degree calculation procedure of back, only need from the user organize transfer the archives with the corresponding interest-degree data of the contained feature of program can and need not repetition above-mentioned steps S110～S130.
This user organizes archives and (for example: a week), then at this therebetween, can directly call this user at every turn when carrying out program commending and organize archives, thereby need not repetition above-mentioned steps S110～S130 can remain unchanged in a period of time.For conveniently transferring, for example the user can be organized archives writes in the data file for the total interest degree calculation procedure of back and uses, when the calculating of total interest degree is when realizing in set-top box, the method for calling that the user organizes archives for example can be from downloading or extract from the storer that set-top box carries from remote server through network.It is pointed out that after user data regularly or aperiodically upgrades,, should utilize step S110～S130 to recomputate the user and organize archives in order to reflect the variation of user data.
Referring to Fig. 1, enter step S140 subsequently, obtain and relate to contained feature of TV programme and the data of broadcast time, these data can leave in the database of program provider according to certain data structure or be included in the EPG that sends to user's group, for realizing that the device that the total interest degree obtains function uses.When the device that these data are provided was independent of the device that obtains the total interest degree physically, reasonable was they only to be imported once obtain in the device of total interest degree.Suppose that the current TV programme that needs the calculated population interest-degree is an English master TV series of describing touching love story, then it comprises is characterized as A, E and G, the broadcast time of further supposing this program in addition is 20:40～21:20, therefore should belong to this period of 20:30～21:30.
Then enter step S150, the user who creates from step S130 organizes the interest-degree of searching contained each feature of above-mentioned TV series the archives, here, because its broadcast time is just in time dropped on 20:30～21:30 in the period, therefore can in form shown in Figure 3, directly find corresponding interest-degree data.The interest-degree data of supposing the contained feature A of this TV programme, E and G are respectively 0.37,0.145 and-0.015.Very likely the situation of Fa Shenging is, the broadcast time of a program is striden the period, and for example broadcast time 20:25～21:15 just drops on the 19:30～20:30 shown in the table 2 and 20:30～21:30 respectively in two periods, can adopt following processing mode to this.
First kind of mode is to determine period under this program according to the time that program begins to broadcast, and under Jia She the situation, this program promptly is regarded as belonging to 19:30～20:30 period in the above.The second way is opposite with first kind of situation, and the time that it finishes to broadcast according to program is determined the affiliated period, and this moment, this program was regarded as belonging to 20:30～21:30 period.Also having a kind of mode is according to the broadcast time length of program in two different periods, the interest-degree of each feature is calculated in segmentation, that is, with broadcast time length be weight, calculate this feature interest-degree weighted mean value and with this as user group to the final interest-degree value of this feature.
Then enter among the step S160, user's group that step S150 is obtained to each feature in the corresponding broadcast period interest-degree be combined as total interest degree to this program.For example can be with the user that calculates according to following formula group to the arithmetic mean of the interest-degree of the contained feature of this TV programme or weighted mean value as total interest degree P:
Here, S
j iFor the user organizes the interest-degree of feature j in i period, WS
jBe the weight of j feature, m is the quantity of this TV show features.With regard to step S140 for example with regard to, when getting arithmetic mean, the total interest degree promptly equals the arithmetic mean of feature A, E and these three interest-degrees of G, promptly (0.37+0.145-0.015)/3 ≈ 0.043.
And for example, when feature quantity is more, for example 10, also can adopt the methods of marking that in the various contests of being everlasting, uses, do not consider the feature that the interest-degree value is minimum and maximum, and then average.In a word, can adopt various array modes, as long as can reflect the contribution component of each feature to the total interest degree.
Subsequently, enter step S170, determine whether to also have other to need the TV programme of calculated population interest-degree.If have, then return step S140, otherwise, then finish whole process.
Below the feature interest-degree account form of step S120 is done detailed argumentation.It is evident that, user's group depends on each member's the value of relative importance value on the different periods and the fancy grade of this member to this contained feature to the interest-degree of a feature, therefore there is the mode of multiple account form or combination fancy grade and relative importance value to obtain such interest-degree data, below only provides two examples.
First example is based on average weighted notion, particularly, here the relative importance value of each member in each period made normalized, thereby relative importance value is scaled the weighted value of this member in each period, table 4 shows the user and organizes interior each member through the weighted value in the different periods after normalized, and by table 4 as seen, this user's group is by father, mother and three member compositions of child, in each period, each member's weight sum equals 1 all the time.Table 5 show this user organize in each member to the fancy grade of feature A, B and C, every in this table has the implication identical with table 2, repeats no more herein.
Table 4
| Period | The |
Mother | Father |
| ????18:30-19:30 | ????0.8 | ????0 | ????0.2 |
| ????19:30-20:30 | ????0.4 | ????0.3 | ????0.3 |
| ????20:30-22:30 | ????0.3 | ????0.6 | ????0.1 |
| ????22:30-24:00 | ????0 | ????0.7 | ????0.3 |
| ????24:00- | ????0 | ????0 | ????0 |
Table 5
| ????A | ????B | ????C | |
| The child | ????1 | ????0.3 | ????0.2 |
| ????A | ????B | ????C | |
| Mother | ????0.5 | ????0.6 | ????1 |
| Father | ????0.3 | ????0.9 | ????0.7 |
Suppose to have comprised feature A in the TV programme, and reproduction time is 18:30～19:30, then can calculates the interest-degree of this user's group according to following formula to feature:
Here, S
j iBe user's group interest-degree to feature j in i period, D
k jBe the fancy grade of k user to feature j, W
k iBe the weight of k user in i period, n is this user group member's a quantity, and D
k jWith W
k iProduct represented in i period the interest-degree of k user to feature j.
With regard to this example, can get by question blank 4 and 5, the user organizes father, mother and the child weight in period 18:30～19:30 and is respectively 0.2,0 and 0.8, fancy grade to feature A is respectively 0.3,0.5 and 1, therefore separately the interest-degree of feature A is respectively 0.06,0 and 0.8, the interest level that obtains user's group thus is 0.86.
Below provide an example that realizes above-mentioned steps S120 function based on fuzzy logic processes again.
Fig. 2 shows based on the process flow diagram of fuzzy logic processes mode computation user group to an example of the interest-degree of a feature, is example with the situation shown in table 1 and 2 still here.
For calculate user group certain period (for example in the 18:30～19:30) to the interest-degree of a feature (for example A), as shown in Figure 3, at step S210, import a member (for example father) to the fancy grade of feature A and he relative importance value in the corresponding period, can get according to table 2 and 3, they respectively value be 0.7 and 0.6.
Then enter step S220, utilize selected subordinate function with these two of the fancy grade imported among the step S210 and relative importance values clearly value (crisp value) be mapped as degree of membership to fuzzy value.The form of subordinate function depends on concrete application scenario, for example for the example here, can adopt the subordinate function shown in Fig. 3 a and the 3b, and wherein, Fig. 3 a is the subordinate function of fancy grade, horizontal ordinate e
1Represent fancy grade, ordinate is represented degree of membership μ, utilize this subordinate function fancy grade can be mapped as degree of membership to " disliking ", " it doesn't matter " and " liking " these three fuzzy values (fuzzyvalue), Fig. 3 b is the subordinate function of relative importance value, horizontal ordinate e
2Represent relative importance value, ordinate is represented degree of membership μ, utilizes this subordinate function relative importance value can be mapped as degree of membership to " subordinate ", " generally " and " important " these three fuzzy values.
Enter step S230 subsequently, utilize predetermined fuzzy logic ordination to carry out reasoning, thereby obtain fuzzy output, this is one group of degree of membership to different fuzzy values, has reflected that this member (being father here) is to the interested degree of characteristic attribute A.Fig. 3 c shows the subordinate function of this fuzzy output, horizontal ordinate α representative is to the interested degree of characteristic attribute, ordinate is represented degree of membership μ, as seen from the figure, this member is represented as degree of membership to " disliking very much ", " disliking ", " it doesn't matter ", " liking " and " enjoying a lot " these several fuzzy values to the interested degree of characteristic attribute A.As for concrete inference rule, depend primarily on the characteristics of application scenario, for example in this example, can adopt following rule:
If I is e
1Be " disliking ", and e
2Be " subordinate ", then α is " it doesn't matter ";
If II is e
1Be " disliking ", and e
2Be " generally ", then α is " disliking ";
If III is e
1Be " disliking ", and e
2Be " important ", α is " disliking very much " so;
If IV is e
1Be " it doesn't matter ", and e
2Be " subordinate ", α is " it doesn't matter " so;
If V is e
1Be " it doesn't matter ", and e
2Be " generally ", α is " it doesn't matter " so;
If VI is e
1Be " it doesn't matter ", and e
2Be " important ", α is " it doesn't matter " so;
If VII is e
1Be " liking ", and e
2Be " subordinate ", α is " it doesn't matter " so;
If VIII is e
1Be " liking ", and e
2Be " generally ", α is " liking " so;
If IX is e
1Be " liking ", and e
2Be " important ", α is " enjoying a lot " so.
Then enter step S240, this member that step S230 is obtained is converted to one to the interested degree of characteristic attribute A and clearly is worth S
i, just so-called de-fuzzy is handled, and i is member's a numbering here.De-fuzzy processing mode commonly used comprises " center of attraction method (Center-of-Gravity) ", " maximal value center method (Center-of-Maximum) " and " the maximal value method of average (Mean-of-Maximum) ", when adopting the center of attraction method to do the de-fuzzy processing, can utilize following formula to calculate:
Here μ [1] is the height of the output area of satisfied the 1st inference rule, y
1Be the horizontal coordinate of the center of attraction of the output area that satisfies the 1st inference rule, m is for satisfying the number of rule.
Enter step S250 subsequently, judge whether to calculate all members' S
iIf condition does not satisfy, and then returns step S210, otherwise enters step S260.
In step S260, the user is organized the S of interior all members (for example being father, mother and child in this example)
iAddition, thus this user's group obtained to the interest-degree of a feature in certain period, all members' S in perhaps also the user can being organized
iMean value as this user group to the interest-degree of a feature in certain period.
The process flow diagram of Fig. 4 shows according to the another one example of user's group of estimation of the present invention to the method for the total interest degree of a content.
As shown in Figure 4, in step S410, also can obtain user data by variety of way, its structure can as table 1 and 2 or table 4 and 5 shown in, but also can be other forms.
Enter step S420 subsequently, obtain the feature that relates to a TV programme and the data of reproduction time, suppose that the programme information that extracts is that this TV programme comprises feature B and C and broadcasts in 20:30～21:00 from the program source database.
Then enter step S430, determine corresponding relative importance value according to the reproduction time that obtains, for example for the situation of relative importance value segmentation in time variation, can be by the acquisition of tabling look-up.
Enter step S440 subsequently, the relative importance value of determining among all members' fancy grade and the step S430 in organizing according to the user who obtains among the step S410 calculates user's group to the interest-degree of each feature in this plays the period.Relevant user's group was done detailed argumentation in front to the account form of the interest-degree of a feature, repeated no more herein.
Then enter step S450, user's group that step S440 is obtained is combined as total interest degree to this program to the interest-degree of playing each feature in the period, and concrete account form can be referring to top associated description.
Enter step S460 subsequently, judge whether need obtain in addition other program of total interest degree,, then return step S420 if having, otherwise, whole process then finished.
During a plurality of period in the reproduction time of a TV programme is crossed over table 2 or table 4, also can adopt aforesaid way to handle fully.For example, suppose the employing above-mentioned first or the second way, then in step S430, search program begin or finish broadcast time under period, correspondingly, in step S440, calculate user's group each feature begun or the interest-degree in the period under the concluding time at program.Suppose to adopt the third mode, then in step S430, search all affiliated periods of playing programs time, then, in step S440, be weight with the broadcast time length in each time period of in step S430, determining, calculate different time sections to the weighted mean value of the interest-degree of this feature and with this as the interest-degree value of user's group to this feature.
It is worthy of note, as another feasible approach, in above-mentioned steps S440, also can calculate the user organize in each member to the interest-degree of program rather than user's group interest-degree to each feature.Correspondingly, in step S450, each member was combined as total interest degree to this program to the interest-degree of program in the user organized.By two examples describing by Fig. 1 and Fig. 4 respectively above the comparison as seen, their main difference is, the former at first creates out a user and organizes archives (step S120 and S130), this user organizes archives and has in fact defined the mode that user's group changes with the period the interest-degree of each feature, then for each concrete program, only need invoke user group archives can determine that this user's group is to the interest-degree (step S150) of the contained feature of program in the broadcast period, the latter does not then have the step that the establishment user organizes archives, but for each concrete program, all determine the user group membership at first respectively at the relative importance value (step S430) of this playing programs in the period, calculate this user's group then the interest-degree value (step S450) of the contained feature of program in the broadcast period.
Utilize the above-mentioned total interest degree method of obtaining, can be more comprehensively, organize content recommendation to a user exactly.Below be example with the TV programme, by Fig. 5 a kind of preferred embodiment from the method for content recommendation to the user that organize is described.
As shown in Figure 5, in step S510, utilize and above-mentionedly obtain the total interest degree of user's group one or more TV programme by Fig. 1 and the described method of obtaining the total interest degree of Fig. 4.
Enter subsequently among the step S520, the total interest degree result who obtains according to step S530 determines to organize to this user the TV programme of recommendation.The mode of determining for example can be, total interest degree and a pre-set threshold of each program compared, if greater than this threshold value, then list in to the user recommendation list that provides is provided, otherwise to the processing of making comparisons of the total interest degree of next program.Can also be, at first program total interest degree be sorted that list n before the total interest degree rank program in recommendation list then, the n here is predefined positive integer.
At last, in step S530, the recommendation list that obtains is offered user's group.When commending contents when program provider side is finished, recommendation list can offer the user with EPG through network by program provider and organize equipment, for example set-top box or PC, and when commending contents when the user organizes side and finishes, can organize equipment by the user and handle acquisition by the programme information (for example contained feature of program and broadcast time etc.) that program provider is provided.
Below discuss device or the equipment that is used to implement above-mentioned total interest degree acquisition methods and content recommendation method.
Fig. 6 is a block scheme, and it shows realizes an above-mentioned embodiment who obtains the equipment of total interest degree method.This equipment 600 comprises an acquiring unit 610, a receiving element 620 and a determining unit 630.
Acquiring unit 610 is used to obtain user data, these user data comprise a user organize in each member's the time dependent mode of relative importance value and he to the fancy grade of each feature, its form for example as table 1 and 2 or table 4 and 5 shown in.
At different user data obtain manners, acquiring unit 610 has different ways of realization.For example, if user data obtains by measuring, then acquiring unit or device comprise the automated monitor at the scene of being installed in; And when user data was provided with voluntarily by the user, the equipment such as keyboard, mouse, telepilot, speech input device can be used as the user data acquiring unit; If the calculating of total interest degrees of data be the user organize finish on the equipment (for example televisor, set-top box or PC) and user data provide through network by far-end computer, then the user organizes the module that realizes communication function in the equipment and can be considered as data capture unit.
Receiving element 620 is used to receive the information relevant with programme content (comprising contained feature of program and reproduction time).Certainly, in an embodiment, receiving element 620 can unite two into one with acquiring unit 610 on hardware.
Determining unit 630 is used to obtain the total interest degree of this user's group to a content, and it comprises that a user organizes archives generation unit 631, one and searches unit 632 and a computing unit 633.
First, second and the 3rd storage unit 640,650 and 660 all can be easily to lose type storer or non-volatile type memorizer, and can be realized by a storer, and this moment, they were actually the different storage zone of this storer.
Searching unit 632 links to each other with receiving element 620, it searches corresponding interest-degree data according to the information that relates to contained feature of program and reproduction time that receiving element 620 provides in second storage unit 650, thereby obtains in the period corresponding with the playing programs time interest-degree of this user's group to contained each feature of program.For the situation that the reproduction time of a TV programme is crossed over a plurality of periods, search unit 632 and will carry out search operation according to different processing modes.
Fig. 7 a is a block scheme, and it shows an embodiment who organizes the equipment 700a of content recommendation to a user according to of the present invention.This equipment 700a comprises an above-described equipment 600 and a recommendation unit 710a who obtains the total interest degree.
The 3rd storage unit 660 in threshold value comparing unit 711 and the equipment 600 links to each other, it compares the total interest degree of being stored one by one with a pre-set threshold, if greater than threshold value, then the indications with respective program exports recommendation list storage unit 712 to.This recommendation list can further be shown to the user.
Fig. 7 b is a block scheme, and it shows another embodiment that organizes the equipment 700b of content recommendation to a user according to of the present invention.The difference of itself and Fig. 7 a apparatus shown is, replaces threshold value comparing units 711 with sequencing unit 713 among the recommendation unit 710b.
The 3rd storage unit 660 in sequencing unit 713 and the equipment 600 links to each other, and it sorts the total interest degree of being stored, and exports the indications of previous or a plurality of programs of total interest degree maximum to recommendation list storage unit 712.
Should be appreciated that above-mentioned all these unit with and some or all of the various piece that comprised can also utilize software to realize.
The present invention also can realize by the computing machine of suitable programming, the code that the program that this computing machine is equipped with comprises can offer a processor, form a kind of machine, make the code of on this processor, carrying out realize following function: obtain this user organize in each member's the value of relative importance value on different time and he to the fancy grade of feature; Obtain the information about this content, what this information comprised this content provides time and contained feature; Organize interior each member's relative importance value according to this user and determine this total interest degree in the temporal value of providing of this content with to the fancy grade of the contained feature of this content.This computer program can be stored in the memory carrier.
In the present invention, so-called feature, those that are meant that content comprises influence the feature of user's interest level, therefore for different user's group and content, may have unique characteristics combination.With the TV programme is example, can wait these to influence the attribute that spectators watch wish with performer who goes out personation in broadcasting channel, title, the program and program category (for example drama, comedy, romantic play, action movie or motion race etc.) and characterize.It is worthy of note; be not the problem that the present invention need solve how for user's group and the specific characteristics combination of content customization; do not have direct cause-effect relationship, so the knowledge of this aspect should not constitute the qualification to protection domain of the present invention with the obtained technique effect of the present invention yet.
Though top description all is example with the TV programme, but content alleged among the present invention should be made broad understanding, in fact it comprise any can be by the information that the human organ experienced, information such as vision, the sense of hearing, sense of touch and the sense of taste for example, its physical form is including, but not limited to various forms of signals such as light, electric harmony.Here provide the tourism promotional advertisement of the example of another one content-drop to user's group, it can influence the attribute of addressee's interest level as content characteristic with the place of playing, price and preferential terms etc., and the user organize in each member's relative importance value may change along with the date, for example six. a Children's Day is interim, and the child in the family has than bigger power to make decision on ordinary days for the place selection of playing.Similarly, when the Mother's Day, mother's relative importance value is high and other member's relative importance value is low.
Should be appreciated that those skilled in the art can also make according to above description manyly substitutes, revises and variation.When such substituting, within the spirit and scope that modifications and variations fall into attached claim the time, should being included among the present invention.
Claims (15)
1. estimate that a user organizes the method to the total interest degree of a content for one kind, this user's group comprises at least two members, and the method comprising the steps of:
A. obtain and relate to the information that this content can be used contained feature in period and this content;
B. obtain each member's the value of relative importance value on the different periods and the fancy grade of this member to this contained feature; And
C. the value of relative importance value on this available period and this member according to each member determines this total interest degree to the fancy grade of this contained feature.
2. the method for claim 1, wherein described content is a TV programme.
3. the method for claim 1, wherein each member's relative importance value value is set to: at least one described member's relative importance value is the value difference at least two periods, these two free a plurality of periods that were divided in one day of period.
4. the method for claim 1, wherein each member's relative importance value value is set to: at least two same periods of same date not, at least one described member's described relative importance value value difference.
5. the method for claim 1, wherein described step c may further comprise the steps:
C1. according to each member's the relative importance value value on the different periods with the fancy grade of contained feature is determined interest-degree the value on different periods of this user's group to this contained feature;
C2. determine the interest-degree value of this user's group according to the described available period; And
C3. the interest-degree value that makes up this user's group is to obtain described total interest degree.
6. method as claimed in claim 5, wherein, described user group equals the sum of products of each member to the fancy grade of this contained feature and each member's the value of relative importance value on this period to the value of interest-degree on a period of contained feature.
7. method as claimed in claim 5, step c1 comprises: according to each member's relative importance value on the different periods value and to the fancy grade of contained feature, determine the interest-degree value of this user's group according to the fuzzy logic mode.
8. organize the method for content recommendation to a user for one kind, wherein this user's group comprises at least two members, and the method comprising the steps of:
A. obtain and relate to the information that this content can be used contained feature in period and this content;
B. obtain each member's the value of relative importance value on the different periods and the fancy grade of this member to this contained feature;
C. the value of relative importance value on this available period and this member according to each member determines this total interest degree to the fancy grade of this contained feature; And
D. recommend to this user's group according to this total interest degree.
9. estimate that a user organizes the equipment to the total interest degree of a content for one kind, wherein this user's group comprises at least two members, and this equipment comprises:
Deriving means, be used for obtaining relate to this content can be with the information of period and the contained feature of this content;
Receiving trap, the relative importance value that is used to receive each member on the different periods value and this member to the fancy grade of this contained feature; And
Determine device, be used for the fancy grade of this contained feature being determined this total interest degree according to the value of relative importance value on this available period and this member of each member.
10. equipment as claimed in claim 9, wherein, each member's relative importance value value is set to: at least one described member's relative importance value is the value difference at least two periods, a plurality of periods of this two periods division in free a day.
11. equipment as claimed in claim 9, wherein, each member's relative importance value value is set to: at least two same periods of same date not, at least one described member's relative importance value value difference.
12. equipment as claimed in claim 9, wherein, described definite device comprises:
Generating apparatus, be used for according to each member's the relative importance value value on the different periods with the fancy grade of contained feature is determined interest-degree the value on different periods of this user's group to this contained feature;
Search device, be used for determining the interest-degree value of this user's group according to the described available period; And
Calculation element is used to make up the interest-degree value of this user's group to obtain described total interest degree.
13. organize the equipment of content recommendation to a user for one kind, wherein this user's group comprises at least two members, comprising:
Deriving means, be used for obtaining relate to this content can be with the information of period and the contained feature of this content;
Receiving trap, the relative importance value that is used to receive each member on the different periods value and this member to the fancy grade of this contained feature;
Determine device, be used for the fancy grade of this contained feature being determined this total interest degree according to the value of relative importance value on this available period and this member of each member; And
Recommendation apparatus is used for recommending to this user's group according to this total interest degree.
14. one kind is used to estimate the computer program of user's group to the total interest degree of a content, wherein this user's group comprises at least two members, and this computer program comprises:
Obtain and relate to the code that this content can be used the information of contained feature in period and this content;
Obtain each member's the value of relative importance value on the different periods and the code of this member to the fancy grade of this contained feature; And
The fancy grade of this contained feature is determined the code of this total interest degree according to each member's the value of relative importance value on this available period and this member.
15. memory carrier that comprises as computer program as described in the claim 14.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200680018627A CN101874255A (en) | 2005-05-27 | 2006-05-18 | Method and apparatus for estimating total interest of a group of users directing to a content |
Applications Claiming Priority (4)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200510073942.3 | 2005-05-27 | ||
| CN200510073942 | 2005-05-27 | ||
| CN200680018627A CN101874255A (en) | 2005-05-27 | 2006-05-18 | Method and apparatus for estimating total interest of a group of users directing to a content |
| PCT/IB2006/051570 WO2006126147A2 (en) | 2005-05-27 | 2006-05-18 | Method and apparatus for estimating total interest of a group of users directing to a content |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101874255A true CN101874255A (en) | 2010-10-27 |
Family
ID=36791701
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200680018627A Pending CN101874255A (en) | 2005-05-27 | 2006-05-18 | Method and apparatus for estimating total interest of a group of users directing to a content |
Country Status (5)
| Country | Link |
|---|---|
| EP (1) | EP1891588A1 (en) |
| JP (1) | JP2008542870A (en) |
| KR (1) | KR20080021069A (en) |
| CN (1) | CN101874255A (en) |
| WO (1) | WO2006126147A2 (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102957969A (en) * | 2012-05-18 | 2013-03-06 | 华东师范大学 | Device and method for recommending program to IPTV (Internet protocol television) terminal user |
| CN103297853A (en) * | 2013-06-07 | 2013-09-11 | 华东师范大学 | IPTV (internet protocol television) program recommendation method based on context recognition for multiple users |
| CN104035934A (en) * | 2013-03-06 | 2014-09-10 | 腾讯科技（深圳）有限公司 | Multimedia information recommending method and device |
| CN104636816A (en) * | 2013-11-07 | 2015-05-20 | 财团法人资讯工业策进会 | Device and method for establishing power utilization model |
| CN104636950A (en) * | 2013-11-07 | 2015-05-20 | 财团法人资讯工业策进会 | Group object commodity recommendation system and method |
| CN104735535A (en) * | 2015-03-24 | 2015-06-24 | 天脉聚源(北京)传媒科技有限公司 | Program scoring method and device |
| CN105163139A (en) * | 2014-05-28 | 2015-12-16 | 青岛海尔电子有限公司 | Information push method, information push server and intelligent television |
| CN107491501A (en) * | 2017-07-28 | 2017-12-19 | 无锡天脉聚源传媒科技有限公司 | A kind of method and device for being grouped push |
| CN111125507A (en) * | 2018-11-01 | 2020-05-08 | 北京邮电大学 | Group activity recommendation method and device, server and computer storage medium |
| US10699181B2 (en) | 2016-12-30 | 2020-06-30 | Google Llc | Virtual assistant generation of group recommendations |
| CN112949322A (en) * | 2021-04-27 | 2021-06-11 | 李蕊男 | E-commerce opinion mining recommendation system driven by online text comments |
| CN113506124A (en) * | 2021-06-21 | 2021-10-15 | 安徽西柚酷媒信息科技有限公司 | Method for evaluating media advertisement putting effect in intelligent business district |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4433326B2 (en) * | 2007-12-04 | 2010-03-17 | ソニー株式会社 | Information processing apparatus and method, and program |
| KR101404010B1 (en) | 2008-03-06 | 2014-06-12 | 주성엔지니어링(주) | Etcher of substrate edge and method of etching substrate edge |
| JP5337748B2 (en) * | 2010-03-09 | 2013-11-06 | 日本電信電話株式会社 | Information processing apparatus and information processing program |
| JP5116811B2 (en) * | 2010-07-02 | 2013-01-09 | 日本電信電話株式会社 | Program recommendation device, method and program |
| JP5508987B2 (en) * | 2010-08-13 | 2014-06-04 | 日本電信電話株式会社 | Provided information selection apparatus, method and program |
| JP2012222569A (en) * | 2011-04-07 | 2012-11-12 | Nippon Telegr & Teleph Corp <Ntt> | Broadcast-program recommending device, method and program |
| KR102232798B1 (en) * | 2014-03-18 | 2021-03-26 | 에스케이플래닛 주식회사 | Service apparatus, user apparatus estimating regions of interest and method therefor, computer readable medium having computer program recorded therefor |
-
2006
- 2006-05-18 JP JP2008512984A patent/JP2008542870A/en not_active Withdrawn
- 2006-05-18 WO PCT/IB2006/051570 patent/WO2006126147A2/en not_active Application Discontinuation
- 2006-05-18 CN CN200680018627A patent/CN101874255A/en active Pending
- 2006-05-18 KR KR1020077030490A patent/KR20080021069A/en not_active Application Discontinuation
- 2006-05-18 EP EP06744970A patent/EP1891588A1/en not_active Withdrawn
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102957969A (en) * | 2012-05-18 | 2013-03-06 | 华东师范大学 | Device and method for recommending program to IPTV (Internet protocol television) terminal user |
| CN104035934A (en) * | 2013-03-06 | 2014-09-10 | 腾讯科技（深圳）有限公司 | Multimedia information recommending method and device |
| WO2014134944A1 (en) * | 2013-03-06 | 2014-09-12 | Tencent Technology (Shenzhen) Company Limited | System and method for recommending multimedia information |
| CN104035934B (en) * | 2013-03-06 | 2019-01-15 | 腾讯科技（深圳）有限公司 | A kind of method and device that multimedia messages are recommended |
| CN103297853B (en) * | 2013-06-07 | 2016-04-06 | 华东师范大学 | A kind of IPTV program commending method based on multi-user's context identification |
| CN103297853A (en) * | 2013-06-07 | 2013-09-11 | 华东师范大学 | IPTV (internet protocol television) program recommendation method based on context recognition for multiple users |
| CN104636816A (en) * | 2013-11-07 | 2015-05-20 | 财团法人资讯工业策进会 | Device and method for establishing power utilization model |
| CN104636950A (en) * | 2013-11-07 | 2015-05-20 | 财团法人资讯工业策进会 | Group object commodity recommendation system and method |
| CN105163139B (en) * | 2014-05-28 | 2018-06-01 | 青岛海尔电子有限公司 | Information-pushing method, Information Push Server and smart television |
| CN105163139A (en) * | 2014-05-28 | 2015-12-16 | 青岛海尔电子有限公司 | Information push method, information push server and intelligent television |
| CN104735535A (en) * | 2015-03-24 | 2015-06-24 | 天脉聚源(北京)传媒科技有限公司 | Program scoring method and device |
| US10699181B2 (en) | 2016-12-30 | 2020-06-30 | Google Llc | Virtual assistant generation of group recommendations |
| TWI720255B (en) * | 2016-12-30 | 2021-03-01 | 美商谷歌有限責任公司 | Method and computing device for generating group recommendations, and non-transitory computer-readable storage medium |
| US11562198B2 (en) | 2016-12-30 | 2023-01-24 | Google Llc | Virtual assistant generation of group recommendations |
| CN107491501A (en) * | 2017-07-28 | 2017-12-19 | 无锡天脉聚源传媒科技有限公司 | A kind of method and device for being grouped push |
| CN111125507A (en) * | 2018-11-01 | 2020-05-08 | 北京邮电大学 | Group activity recommendation method and device, server and computer storage medium |
| CN111125507B (en) * | 2018-11-01 | 2023-07-21 | 北京邮电大学 | Group activity recommendation method and device, server and computer storage medium |
| CN112949322A (en) * | 2021-04-27 | 2021-06-11 | 李蕊男 | E-commerce opinion mining recommendation system driven by online text comments |
| CN113506124A (en) * | 2021-06-21 | 2021-10-15 | 安徽西柚酷媒信息科技有限公司 | Method for evaluating media advertisement putting effect in intelligent business district |
| CN113506124B (en) * | 2021-06-21 | 2022-03-25 | 安徽西柚酷媒信息科技有限公司 | Method for evaluating media advertisement putting effect in intelligent business district |
Also Published As
| Publication number | Publication date |
|---|---|
| KR20080021069A (en) | 2008-03-06 |
| JP2008542870A (en) | 2008-11-27 |
| EP1891588A1 (en) | 2008-02-27 |
| WO2006126147A2 (en) | 2006-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101874255A (en) | Method and apparatus for estimating total interest of a group of users directing to a content | |
| Tewksbury et al. | News on the Internet: Information and Citizenship in the 21st Century | |
| US4996642A (en) | System and method for recommending items | |
| CN1647073B (en) | Information search system, information processing apparatus and method, and information search apparatus and method | |
| US20200193288A1 (en) | System and Method for Content Discovery | |
| US7853600B2 (en) | System and method for providing access to video programs and other data using customer profiles | |
| CN100551031C (en) | In the project recommendation device, a plurality of items are divided into the method and the device of similar group | |
| CN105681835B (en) | A kind of method and server of information push | |
| JP4358219B2 (en) | Program information providing apparatus, program information providing method, and program thereof | |
| US20070240181A1 (en) | Subscriber Characterization System with Filters | |
| CN110222233B (en) | Video recommendation method and device, server and storage medium | |
| JP2006524009A (en) | Generating audience analysis results | |
| CN110474944B (en) | Network information processing method, device and storage medium | |
| CN103649981A (en) | Method and apparatus for delivering targeted content | |
| US10165315B2 (en) | Systems and methods for predicting audience measurements of a television program | |
| JP2005056361A (en) | Information processor and method, program, and storage medium | |
| CN105516810A (en) | Television user family member analysis method based on LDA (Latent Dirichlet Allocation) model | |
| CN110225361A (en) | Direct broadcasting room recommended method, storage medium, electronic equipment and system | |
| CN109729377A (en) | A kind of method for pushing, device, computer equipment and the storage medium of main broadcaster's information | |
| JP2001134706A (en) | User action predicting method and action mode selector | |
| JP2004519902A (en) | Television viewer profile initializer and related methods | |
| CN109640128B (en) | Television user watching behavior feature extraction method and system | |
| CN109005431B (en) | A kind of video evaluations recommender system | |
| CN109327739A (en) | A kind of method for processing video frequency, calculates equipment and storage medium at device | |
| CN106156270A (en) | Multi-medium data method for pushing and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20101027 |