CN104144181B - A kind of Video Terminal polymerization and system - Google Patents
A kind of Video Terminal polymerization and system Download PDFInfo
- Publication number
- CN104144181B CN104144181B CN201310166163.2A CN201310166163A CN104144181B CN 104144181 B CN104144181 B CN 104144181B CN 201310166163 A CN201310166163 A CN 201310166163A CN 104144181 B CN104144181 B CN 104144181B
- Authority
- CN
- China
- Prior art keywords
- interest
- video
- multidimensional
- module
- internet video
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Landscapes
- Information Transfer Between Computers (AREA)
Abstract
The invention discloses a kind of Video Terminal polymerization and system, is particluarly suitable for Intelligent television terminal application, including:Subscribing module, for the source for specifying Internet video to polymerize;Reptile module, for extracting Internet video metadata from subscription website;Local data library module, recorded and local video information for storing local play;Pretreatment module, for being pre-processed to local data base data, to adapt to interest digging needs;Interest digging module, for according to local data base, excavating user's multidimensional interest topic;Filtering module is matched, for carrying out filtering sequence to Internet video according to Internet video and user's matching degree interested;UI display modules, for showing Internet video list obtained by filtered, sequence.This method plays record using the abundant user of terminal, excavates user's multidimensional interest topic, and polymerize the Internet video for meeting user interest from the Internet video resource of magnanimity with restrictive conditions such as interest to terminal using subscribing to.
Description
Technical field
The present invention relates to Data Mining, information fusion field, realizes using information of terminal user to user interest
Excavate, extract multidimensional interest topic, and as evidence, in the video website that user is subscribed to, user's video aggregation interested
To user terminal.
Background technology
Traditional terminal media information management is confined to the management and renewal of local media information database, so that user looks into
Read, and submitted necessary information support to interactive system when user operates.Under integration of three networks background, as the network terminal
TV set intelligent terminal operating system, be no longer satisfied with the access to locally store information, and need the needs according to user
More rich network video information is provided.The video image local video interested of user in internet video website is equally presented
Click on and play to user, have become trend.
The means of the Internet video of user's acquisition at present browse including video website, search for, recommend, and a small amount of c/s gathers
Syzygy is united.Web page browsing and search are widely used in pc terminals, but for television set, mobile phone etc. and are bad at key mouse operation
For intelligent terminal, hence it is evident that add burden for users, reduce Consumer's Experience.Existing paradigmatic system all uses server/end
End pattern, being faced with user must register, feed back, and be forced to receive the redundancy predicament such as service end advertisement.
On the other hand, the magnanimity of network information resource increases the continuous expansion with user's contributing content, and money is brought to user
Source selection variation and while autonomy-oriented, the problem of also bringing selection to get lost, and when the information content of internet is increasing, search
Index holds up how this mode that information is presented based on content in itself is improved again, can not also avoid the information of redundancy.Bulk redundancy
The presence of information can also turn into the burden of user and terminal.
The content of the invention
, being capable of effective extending video it is an object of the present invention to provide a kind of method of terminal active converging network video
Source, the network video source for enjoying magnanimity growth to the full are brought resource selection variation and while autonomy-oriented to user, avoided
Its selection brought is got lost problem.Also to avoid needing to register in current major programme simultaneously, submit user profile, explicit acquisition
The intelligent terminal users such as interest, especially TV user are typically reluctant to the thing done.
To achieve the above object, the invention provides a kind of Video Terminal polymerization, methods described to include:
Step 101)By subscribing to the source for specifying Internet video to polymerize;
Step 102)Using reptile Internet video metadata is extracted from website is subscribed to;
Step 103)Local record and the local video information of playing is stored in local data base;
Step 104)Local data base data are pre-processed, to adapt to interest digging needs, wherein, the pretreatment
It is that the video information stored in database is filtered one by one, rejects invalid information record, chooses qualified data for emerging
Interest is excavated;
Step 105)According to local data base, user's multidimensional interest topic is excavated, the local data base is with certain number
Some video metadata descriptions are stored according to structure, these object videos include the video file being locally stored, and user broadcasts
Put the video in record;
Step 106)Filtering sequence is carried out to Internet video according to Internet video and the matching degree of user interest, described
Each Internet video description information is matched with interest topic successively with filtering, filters and retains matching degree higher than threshold value
As a result, and sort;
Step 107)Show Internet video list obtained by filtered, sequence.
Above-mentioned Internet video metadata includes:Video name, video source address, time, director, performer or type, will be all
Metadata forms the multidimensional description information of Internet video.
The website of above-mentioned designated syndicated feeds is the homepage network address of one or more video websites.
One or more webpages that above-mentioned reptile module is specified using subscribing module extract video metadata as initial page,
A metadata description is generated for each video, and the two level page nestedly included to it travels through one by one, meets to obtain
The video element number of condition;Meanwhile the mode for obtaining metadata also alternatively includes what directly harvesting website was issued by certain specification
Video information.
Multidimensional interest topic is the interest topic being described in two or more dimensions, and its basis is, each
Bar video information is all made up of the description information of multiple dimensions.
The extraction of multidimensional interest topic is divided into following steps:
A, one-dimensional interest extraction:The dimension excavated to each plan takes independent interest digging strategy and standard, obtains
Some interest topics in the dimension, turn into a set;
B, two-dimentional interest extraction:Between different dimensions, if two interest topics are appeared in a multidimensional information simultaneously,
The two interest topics are relevant;Appearance is more simultaneously, and association is bigger;The degree of association combining greater than threshold value, turn into
One two-dimentional interest topic, all two-dimentional interest topics are found out using same method;
C, multidimensional interest is extracted:If the theme in certain dimension is appeared in two multidimensional themes, check whether that the two are more
The correlation degree existed in theme between each one-dimensional theme more than threshold value is tieed up, if so, then merging the two multidimensional themes, is turned into
More high-dimensional interest topic;
D, all multidimensional interest topics that can not further merge are recorded.
To achieve these goals, present invention also offers a kind of Video Terminal paradigmatic system, the system to include:
Subscribing module, for the source for specifying Internet video to polymerize;
Internet video metadata is extracted in reptile module, the website of the Internet video syndicated feeds for being obtained from subscribing module;
Local data library module, recorded and local video information for storing local play;
Pretreatment module, for being pre-processed to local data base data, to adapt to interest digging needs;
Interest digging module, extracted for carrying out the one-dimensional interest to multidimensional according to following principle according to local data base:
One-dimensional interest extraction:The dimension excavated to each plan takes independent interest digging strategy and standard, is somebody's turn to do
Some interest topics in dimension, turn into a set;
Two-dimentional interest extraction:Between different dimensions, if two interest topics are appeared in a multidimensional information simultaneously, this
Two interest topics are relevant;Appearance is more simultaneously, and association is bigger;The degree of association linking together greater than threshold value, turn into one
Individual two-dimentional interest topic, all two-dimentional interest topics are found according to this strategy;
Multidimensional interest is extracted:If the theme in certain dimension is appeared in two multidimensional themes, the two multidimensional are checked whether
The correlation degree more than threshold value in theme between each one-dimensional theme be present, if so, then merging the two multidimensional themes, turn into more
High-dimensional interest topic;All multidimensional interest topics that can not further merge are recorded, complete interest digging;
Filtering module is matched, filtering row is carried out to Internet video for the matching degree according to Internet video and user interest
Sequence;
Display module, for showing Internet video list obtained by filtered, sequence.
Above-mentioned subscribing module allows a user to specify the source that one or more video website network address polymerize as video information, quilt
The video for specifying the page of network address and its two level page of reference to include is included in the range of follow-up polymerization, and can be specified
The homepage of video website.
Above-mentioned reptile module extracts video metadata on the page in subscription range, or directly gathers in website by certain
The video information of specification issue, and the metadata for belonging to a video is arranged as a description network video according to data structure
The information of frequency, and the module nestedly captures metadata to the two level page of each page.
Compared with prior art, the technical advantages of the present invention are that:
The present invention provides a kind of method of terminal active converging network video, can effective extended network source video sequence, have
Effect takes into account user and selects diversified autonomy-oriented to avoid selection from getting lost with effectively removing redundancy.This method is regarded based on terminal-pair
The active pull of frequency metadata, for user offer convenience efficient experience while, make full use of the abundant user profile of terminal
Interest topic is implicitly obtained, the terminal users such as registration, scoring is avoided and is typically reluctant to also be inconvenient to the link participated in.Always
It, the present invention changes the existing limitation for polymerizeing and being all based on c/s patterns, and end is pushed to after completing polymerization by server end
End, i.e. technical scheme are actively captured by terminal, therefore need not submit userspersonal information, need not also receive service end
The advertising lamp content pushed by force;In addition, interest digging is introduced into the filter process of information fusion by the present invention, with reference to syndicated feeds
It is customized, effectively increase the degree of accuracy of polymerization result, reduce redundancy.
Brief description of the drawings
Fig. 1 major function composition description figures of the present invention;
Fig. 2 Video Terminal polymerization schematic flow sheets provided by the invention.
Embodiment
Below in conjunction with the accompanying drawings and specific embodiment the invention will be further described.
As shown in figure 1, method described herein, mainly includes three main functional parts, i.e. interest digging part, information
Polymeric part, filter display portion.Interest digging part exports multidimensional interest topic by the excavation to user profile, and information is gathered
Close part and obtain video metadata from network address is subscribed to, and arrange the multidimensional description of output Internet video.
As shown in Fig. 2 information fusion part mainly includes subscribing module and reptile module and provides the mutual of Internet video
Networking.The two of user's specified satisfactory network address in the range of whole internet by subscribing module, the webpage and its reference
The level page can be traveled through extraction useful information, i.e. video metadata by reptile modules nests, and reptile module obtains metadata approach
Including the Internet video metadata directly issued from specified address acquisition according to certain specification.Reptile module is taken after metadata
Can be again according to form as defined in system, the Internet video multidimensional meta data that metadata is integrated into specification describes, each
Description represents an Internet video.
As shown in Fig. 2 interest digging part mainly includes local data library module, data preprocessing module and interest digging
Module.Local data library module stores some video multidimensional meta data descriptions, and what described video included being locally stored regards
Frequency object, and video playback record etc. embody the object video of user interest.Might not be to every in record in view of user
One video is all interesting, such as the video that viewing duration is too short, it is believed that and it is the videograph that can not embody user interest,
Data preprocessing module is responsible for rejecting the data for not embodying user interest from database, leaves the data for embodying user interest,
And it is supplied to interest digging module to be used to extract multidimensional interest topic.
Interest digging module extracts two-dimentional interest according to following steps and gradually forms more high-dimensional interest topic:
A, one-dimensional interest extraction:The dimension excavated to each plan takes independent interest digging strategy and standard, obtains
Some interest topics in the dimension, turn into a set.For example, for directing this dimension, it is only necessary to count different names
The frequency of appearance, frequency are higher than one interest topic of conduct of threshold value.Selecting for threshold value is very crucial, and threshold value is too low, extraction
As a result user interest can not be represented, threshold value is too high, then may leak through some interest.For this unitary number of the actual address of video
According to then needing to use increasingly complex Path mining method interested, and whether a metadata meets certain interest topic determine
Justice also can correspondingly be changed into, and whether the metadata belongs under the interest path of the theme.Therefore, should be adopted in each dimension
Independent interest digging strategy process is taken, also to define different measurement interest criterias.Further, it is contemplated that cost benefit, not necessarily
Need all to include all dimensions the scope of interest digging.
B, two-dimentional interest extraction:Between different dimensions, if two interest topics are appeared in a multidimensional information simultaneously, recognize
It is relevant for the two interest topics;Appearance is more simultaneously, and association is bigger;The degree of association is linked together greater than threshold value, into
For a two-dimentional interest topic.Same method can find out all two-dimentional interest topics.The measurement of the degree of association can use
The relative scale occurred simultaneously, the absolute number occurred simultaneously can also be used.As shown in table 1, transverse axis represents dimension A, the longitudinal axis
Dimension B is represented, each dimension is interested in, and theme is some, and each formation one is gathered(a1,a2,a3,a4,a5,a6,a7),(b1,b2,
b3,b4,b5).Assuming that this use appears in absolute number in same video presentation as the degree of association simultaneously, then above interest
The degree of association of theme is as shown in numerical value in matrix.Assuming that determine that there is interest topic the threshold value of relevance to be on two different dimensions
10, then can determine that two-dimentional interest topic can be extracted in the matrix(a1,b1),(a2,b2),(a3,b3),(a3,b4),(a4,
b2),(a5,b5),(a7,b1).
Table 1, different dimensions interest topic degree of association matrix
a1 | a2 | a3 | a4 | a5 | a6 | a7 | |
b1 | 38 | 9 | 1 | 0 | 0 | 1 | 12 |
b2 | 0 | 16 | 0 | 20 | 2 | 0 | 3 |
b3 | 0 | 0 | 14 | 2 | 0 | 0 | 0 |
b4 | 0 | 8 | 17 | 0 | 0 | 1 | 0 |
b5 | 7 | 3 | 0 | 0 | 14 | 0 | 0 |
C, multidimensional interest is extracted:If the theme in certain dimension appears in two multidimensional(Include two dimension)In theme, inspection is
The correlation degree more than threshold value in the two no multidimensional themes between each one-dimensional theme be present, if so, it is more then to merge the two
Theme is tieed up, turns into more high-dimensional interest topic.For example, dimension C interest topic set be present(c1,c2,c3,c4), exist
Two-dimentional interest topic(a1,c2),(a1,c3),(b1,c2).Three-dimensional interest topic can so be drawn(a1,b1,c2), and because
The degree of association between tri- themes of a1, b1, c2 has been above threshold value, it can be assumed that they come from the same interest of user
Three dimensions.
D, all multidimensional that can not further merge are recorded(Including two dimension)Interest topic.In this process, record
The multidimensional theme that can not further merge, and do not re-record the multidimensional theme being merged, foregoing A, B, C three-dimensionals
In description, by being associated with three-dimensional theme(a1,b1,c2), need not just re-record(a1,b1),(a1,c2),(b1,c2).
As shown in Fig. 2 filtering display portion mainly includes matching filtering module and UI display modules.It is first to filter matching module
All outputs of reptile module, i.e. the Internet video multidimensional description information from network extraction are first taken, then according to every description
With the matching degree of multidimensional interest, determine if to meet user interest, meet user interest in much degree.Filtering module selects
The video information for meeting user interest is taken, and is sorted, finally gives UI display modules.UI display modules are responsible for final network
Video information is presented to reader, and provides the interface needed for response user's clicking operation.
In a word, method provided by the invention makes full use of terminal advantage, and the information that research meets STB terminal scene is gathered
Service is closed, on the basis of polymerization scope is reduced by formulating syndicated feeds, record and media database is played using local, digs
User interest is dug, to reduce polymerization scope and data scale, terminal polymerization is effectively reduced while improving and polymerizeing precision
Resource consumption.That is, the invention provides a kind of solution method based on terminal polymerization, by making full use of the abundant use of terminal
Family information, user interest is implicitly obtained, and according to interest, the video website polymerization freely customized from user meets regarding for interest
Frequency inventory, and user is shown to as being locally stored video.This method allows client independently to customize source video sequence, whole network
Potential video source can be turned into, effectively extend information source, while by formulating syndicated feeds, and the mistake based on interest
Filter, effectively reduces redundancy, most user's video interested is presented on terminal interface at last.The present invention is without user's registration
Or interest is explicitly provided, support more interest topics, and the redundancy that effectively service end pushes away by force in shielding c/s patterns.
On the basis of the present invention program, the various modifications or deformation made still fall within protection scope of the present invention.
Claims (9)
1. a kind of Video Terminal polymerization, methods described include:
Step 101) is by subscribing to the source for specifying Internet video to polymerize;
Step 102) extracts Internet video metadata using reptile from website is subscribed to;
Local record and the local video information of playing is stored in local data base by step 103);
Step 104) pre-processes to local data base data, to adapt to interest digging needs, wherein, the pretreatment is pair
The video information stored in database filters one by one, rejects invalid information record, chooses qualified data and is dug for interest
Pick;
Step 105) excavates user's multidimensional interest topic, the local data base is with certain data knot according to local data base
Structure stores some video metadata descriptions, and these object videos include the video file being locally stored, and user plays note
Video in record;
Step 106) carries out filtering sequence according to Internet video and the matching degree of user interest to Internet video, described to match
Filter successively matches each Internet video description information with interest topic, filters and retain the knot that matching degree is higher than threshold value
Fruit, and sort;
Step 107) display is filtered, Internet video list obtained by sequence.
2. Video Terminal polymerization according to claim 1, it is characterised in that the Internet video metadata bag
Include:Video name, video source address, time, director, performer or type, the multidimensional that all metadata are formed to Internet video describe
Information.
3. Video Terminal polymerization according to claim 1, it is characterised in that the designated syndicated feeds
Website is the homepage network address of one or more video websites.
4. Video Terminal polymerization according to claim 1, it is characterised in that the reptile module is to subscribe to mould
One or more webpages that block is specified are initial page, extract video metadata, and generating a metadata for each video retouches
State, and the two level page nestedly included to it travels through one by one, to obtain qualified video element number;Meanwhile obtain first number
According to mode also alternatively include directly gather in website by certain specification issue video information.
5. Video Terminal polymerization according to claim 1, it is characterised in that the multidimensional interest topic is
In the interest topic that two or more dimensions are described, its basis is that each video information is all by multiple dimensions
Description information forms.
6. Video Terminal polymerization according to claim 1, it is characterised in that the multidimensional interest topic carries
Take and be divided into following steps:
A, one-dimensional interest extraction:The dimension excavated to each plan takes independent interest digging strategy and standard, obtains the dimension
Some interest topics on degree, turn into a set;
B, two-dimentional interest extraction:Between different dimensions, if two interest topics are appeared in a multidimensional information simultaneously, this two
Individual interest topic is relevant;Appearance is more simultaneously, and association is bigger;The degree of association combining greater than threshold value, turn into one
Two-dimentional interest topic, all two-dimentional interest topics are found out using same method;
C, multidimensional interest is extracted:If the theme in certain dimension is appeared in two multidimensional themes, the two multidimensional master is checked whether
The correlation degree more than threshold value in topic between each one-dimensional theme be present, if so, then merging the two multidimensional themes, turn into higher
The interest topic of dimension;
D, all multidimensional interest topics that can not further merge are recorded.
7. a kind of Video Terminal paradigmatic system, it is characterised in that the system includes:
Subscribing module, for the source for specifying Internet video to polymerize;
Internet video metadata is extracted in reptile module, the website of the Internet video syndicated feeds for being obtained from subscribing module;
Local data library module, recorded and local video information for storing local play;
Pretreatment module, for being pre-processed to local data base data, to adapt to interest digging needs;
Interest digging module, extracted for carrying out the one-dimensional interest to multidimensional according to following principle according to local data base:
One-dimensional interest extraction:The dimension excavated to each plan takes independent interest digging strategy and standard, obtains the dimension
On some interest topics, turn into a set;
Two-dimentional interest extraction:Between different dimensions, if two interest topics are appeared in a multidimensional information simultaneously, the two
Interest topic is relevant;Appearance is more simultaneously, and association is bigger;The degree of association linking together greater than threshold value, turn into one two
Interest topic is tieed up, all two-dimentional interest topics are found according to this strategy;
Multidimensional interest is extracted:If the theme in certain dimension is appeared in two multidimensional themes, the two multidimensional themes are checked whether
In correlation degree more than threshold value between each one-dimensional theme be present, if so, then merging the two multidimensional themes, turn into more higher-dimension
The interest topic of degree;All multidimensional interest topics that can not further merge are recorded, complete interest digging;
Filtering module is matched, filtering sequence is carried out to Internet video for the matching degree according to Internet video and user interest;
Display module, for showing Internet video list obtained by filtered, sequence.
8. Video Terminal paradigmatic system according to claim 7, it is characterised in that the subscribing module allows user
The source for specifying one or more video website network address to polymerize as video information, it is designated the page of network address and its two level of reference
The video that the page includes is included in the range of follow-up polymerization, and is capable of the homepage of designated website.
9. Video Terminal paradigmatic system according to claim 7, it is characterised in that the reptile module is subscribing to model
Video metadata is extracted on the page in enclosing, or directly gathers in website and presses the video information of certain specification issue, and will be belonged to together
It is the information for describing Internet video to be arranged in the metadata of a video according to data structure, and the module is to each page
The two level page nestedly capture metadata.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310166163.2A CN104144181B (en) | 2013-05-08 | 2013-05-08 | A kind of Video Terminal polymerization and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310166163.2A CN104144181B (en) | 2013-05-08 | 2013-05-08 | A kind of Video Terminal polymerization and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104144181A CN104144181A (en) | 2014-11-12 |
CN104144181B true CN104144181B (en) | 2017-12-29 |
Family
ID=51853249
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310166163.2A Expired - Fee Related CN104144181B (en) | 2013-05-08 | 2013-05-08 | A kind of Video Terminal polymerization and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104144181B (en) |
Families Citing this family (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106202095B (en) * | 2015-05-05 | 2020-06-30 | 腾讯科技(深圳)有限公司 | Group message display method and device |
CN104834728B (en) * | 2015-05-14 | 2018-03-09 | 无锡天脉聚源传媒科技有限公司 | A kind of method for pushing and device for subscribing to video |
GB2554993A (en) | 2015-06-11 | 2018-04-18 | Google Llc | Methods, systems, and media for aggregating and presenting content relevant to a particular video game |
CN106470349A (en) * | 2015-08-21 | 2017-03-01 | 中国科学院声学研究所 | The extraction of HTML5 video and manipulation optimization method in a kind of intelligent television |
ES2901113T3 (en) * | 2015-12-28 | 2022-03-21 | Bombora Inc | Surge detector for content consumption |
CN106572394B (en) * | 2016-08-30 | 2020-04-14 | 上海二三四五网络科技有限公司 | Movie and television data navigation method |
CN107577808B (en) * | 2017-09-26 | 2020-03-17 | 武汉斗鱼网络科技有限公司 | Method, device, server and medium for sorting multi-level list pages |
CN110309397A (en) * | 2018-03-19 | 2019-10-08 | 北京数码视讯软件技术发展有限公司 | Video screening technique and system |
CN109121001B (en) * | 2018-09-05 | 2021-07-27 | 深圳市酷开网络科技股份有限公司 | Carousel system and carousel method for multiple content providers and smart television |
CN110297942B (en) * | 2019-06-26 | 2024-01-05 | 广州市百果园信息技术有限公司 | Video exploration method, device, equipment and storage medium |
CN110392308A (en) * | 2019-07-08 | 2019-10-29 | 深圳市轱辘汽车维修技术有限公司 | A kind of video recommendation method, video recommendations device and server |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101174273B (en) * | 2007-12-04 | 2010-06-23 | 清华大学 | News event detecting method based on metadata analysis |
US8145679B1 (en) * | 2007-11-01 | 2012-03-27 | Google Inc. | Video-related recommendations using link structure |
CN102800006A (en) * | 2012-07-23 | 2012-11-28 | 姚明东 | Real-time goods recommendation method based on customer shopping intention exploration |
CN103051930A (en) * | 2012-12-21 | 2013-04-17 | 福建邮科通信技术有限公司 | Method and system for recommending mobile video based on flow analysis and user behavior analysis |
-
2013
- 2013-05-08 CN CN201310166163.2A patent/CN104144181B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8145679B1 (en) * | 2007-11-01 | 2012-03-27 | Google Inc. | Video-related recommendations using link structure |
CN101174273B (en) * | 2007-12-04 | 2010-06-23 | 清华大学 | News event detecting method based on metadata analysis |
CN102800006A (en) * | 2012-07-23 | 2012-11-28 | 姚明东 | Real-time goods recommendation method based on customer shopping intention exploration |
CN103051930A (en) * | 2012-12-21 | 2013-04-17 | 福建邮科通信技术有限公司 | Method and system for recommending mobile video based on flow analysis and user behavior analysis |
Also Published As
Publication number | Publication date |
---|---|
CN104144181A (en) | 2014-11-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104144181B (en) | A kind of Video Terminal polymerization and system | |
CN110574387B (en) | Recommending live streaming content using machine learning | |
US8650198B2 (en) | Systems and methods for facilitating the gathering of open source intelligence | |
CN107295361B (en) | A kind of content delivery method | |
US10198776B2 (en) | System and method for delivering an open profile personalization system through social media based on profile data structures that contain interest nodes or channels | |
US9405746B2 (en) | User behavior models based on source domain | |
CN104111941B (en) | The method and apparatus that information is shown | |
WO2010119288A1 (en) | Metadata browser | |
Ramakrishnan et al. | Toward a peopleweb | |
CN102682082B (en) | Network Flash searching system and network Flash searching method based on content structure characteristics | |
CN110633406B (en) | Event thematic generation method and device, storage medium and terminal equipment | |
CN102523511A (en) | Network program aggregation and recommendation system and network program aggregation and recommendation method | |
CN103686231A (en) | Method and system for integrated management, failure replacement and continuous playing of film | |
CN104504059A (en) | Multimedia resource recommending method | |
CN101764661A (en) | Data fusion based video program recommendation system | |
CN106354860A (en) | Method for automatically labelling and pushing information resource based on label sets | |
CN103384883A (en) | Semantic enrichment by exploiting Top-K processing | |
CN106792172A (en) | A kind of method of internet television personalized recommendation video | |
CN103207917A (en) | Method for marking multimedia content and method and system for generating recommended content | |
Thalhammer et al. | Leveraging usage data for linked data movie entity summarization | |
CN106303596A (en) | Video recommendation system, method and device | |
CN112559513A (en) | Link data access method, device, storage medium, processor and electronic device | |
Papadopoulos et al. | Automatic summarization and annotation of videos with lack of metadata information | |
CN107818156A (en) | A kind of real time individual news recommends method and system | |
CN105608158A (en) | Method and apparatus for displaying picture in waterfall flow manner |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20171229 Termination date: 20200508 |
|
CF01 | Termination of patent right due to non-payment of annual fee |