CN105791910B - A kind of multimedia resource supplying system and method - Google Patents

A kind of multimedia resource supplying system and method Download PDF

Info

Publication number
CN105791910B
CN105791910B CN201610129990.8A CN201610129990A CN105791910B CN 105791910 B CN105791910 B CN 105791910B CN 201610129990 A CN201610129990 A CN 201610129990A CN 105791910 B CN105791910 B CN 105791910B
Authority
CN
China
Prior art keywords
multimedia resource
pushed
user
push
multimedia
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.)
Active
Application number
CN201610129990.8A
Other languages
Chinese (zh)
Other versions
CN105791910A (en
Inventor
陆垚
胡倩
庞一
杨海强
张超
赵亚文
刘杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing State Owned Financial Leasing Co.,Ltd.
Original Assignee
STAR SOFTWARE TECHNOLOGY CO LTD
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by STAR SOFTWARE TECHNOLOGY CO LTD filed Critical STAR SOFTWARE TECHNOLOGY CO LTD
Priority to CN201610129990.8A priority Critical patent/CN105791910B/en
Publication of CN105791910A publication Critical patent/CN105791910A/en
Application granted granted Critical
Publication of CN105791910B publication Critical patent/CN105791910B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/258Client or end-user data management, e.g. managing client capabilities, user preferences or demographics, processing of multiple end-users preferences to derive collaborative data
    • H04N21/25866Management of end-user data
    • H04N21/25891Management of end-user data being end-user preferences
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/23Processing of content or additional data; Elementary server operations; Server middleware
    • H04N21/238Interfacing the downstream path of the transmission network, e.g. adapting the transmission rate of a video stream to network bandwidth; Processing of multiplex streams
    • H04N21/2385Channel allocation; Bandwidth allocation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/20Servers specifically adapted for the distribution of content, e.g. VOD servers; Operations thereof
    • H04N21/25Management operations performed by the server for facilitating the content distribution or administrating data related to end-users or client devices, e.g. end-user or client device authentication, learning user preferences for recommending movies
    • H04N21/251Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/252Processing of multiple end-users' preferences to derive collaborative data
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4667Processing of monitored end-user data, e.g. trend analysis based on the log file of viewer selections
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies

Abstract

The present invention provides a kind of multimedia resource supplying system and method, comprising: user information collection device is used to collect the relevant information that user watches multimedia resource;Multimedia resource information collection apparatus is used to collect the description information of multimedia resource;Classifying content recommended engine is used to watch the relevant information of multimedia resource and the description information of multimedia resource according to user, determines the classification of multimedia resource;DVB radio network push front end is used to the multimedia resource for belonging to radio network transmission classification pushing to ground base station push front end or user terminal;Telecommunications network IP plug-flow platform is used to the multimedia resource for belonging to telecommunications network transmission classification pushing to ground base station push front end;Ground base station push front end is used to multimedia resource pushing to user terminal.Since the program is classified multimedia resource, can there are the Selection utilization radio network of foundation or telecommunications network to be pushed, to reach resource optimal allocation, so that network bandwidth utilization factor is higher.

Description

A kind of multimedia resource supplying system and method
Technical field
The present invention relates to video push technical field, in particular to a kind of multimedia resource supplying system and method.
Background technique
In the area that the telecommunication network of the forms such as cable broadband/optical fiber is not popularized, mainly pass through radio telecommunication network (3G, 4G) transmission order video is not able to satisfy user demand, and (video code rate is too high, and there are many user, radio telecommunication network bandwidth It is too small, and rate are expensive) and then face resource homogeneity using DVB broadcasting network push order video (all users can only Receive same resource), the problems such as low efficiency (push resource and live streaming resource-sharing bandwidth, resource is limited).Meanwhile some The underdeveloped area of electric power, power breakdown will lead to Transmission when DVB broadcasting network pushes, and since DVB broadcasting network is Unilateral network, it is not known that the position that last time push is interrupted, it is therefore desirable to the same resource repeatedly be pushed to ensure user just It often receives, so inevitably generates the very big wasting of resources.
Traditional program request (VOD) technology and traditional push program request based on broadcasting network based on telecommunication network (push vod) technology is each relatively mature, and enters commercial stage, but usually all establishes one's own system, base In two kinds of networks integration technology also in research and the exploratory stage, be some fusion broadcasting networks having disclosed at present below With the video distribution and push technology of telecommunication network.
Patent application " a kind of internet merge Video Distribution Network system in conjunction with Broadcasting Cable Network " (application publication number: CN 104869473 A) the video distribution technological frame of a kind of internet and Broadcasting Cable Network combination is proposed, primarily focus on live video Multiple networks fusion, terminal device support receive from internet is transmitted through come video or from Broadcasting Cable Network be transmitted through come video, and And multiple terminals can be distributed to.But this application there is no proposition and solves how to classify to different videos, selects It is some to be pushed using broadcasting network, select other to be pushed using internet (accessing using telecommunications network etc.), to reach The problem of to resource optimal allocation.
Patent of invention " system and method for downloading resource " (patent authorization number: 1941899 B of CN) proposes one kind Video push system based on user's reservation protocol supports the push of telecommunication network and the push of broadcasting network simultaneously, and And it can inform that reminding user to push completes after push finishes.However, this patent is not addressed how to different videos yet Classify, selection is some to be pushed using radio network, selects other to be pushed using telecommunications network, to reach resource most The problem of optimal sorting is matched.
Summary of the invention
The embodiment of the invention provides a kind of multimedia resource supplying system, different multimedia resources has been carried out point Class, can have the selection of foundation is to be pushed using radio network, or pushed using telecommunications network, to reach resource most Optimal sorting is matched.The system includes:
User information collection device, multimedia resource information collection apparatus, classifying content recommended engine, DVB radio network push away Send front end, telecommunications network IP plug-flow platform;
The user information collection device watches the relevant information of multimedia resource for collecting user;
The multimedia resource information collection apparatus, for collecting the description information of multimedia resource to be pushed;
The classifying content recommended engine, for watching the relevant information of multimedia resource and to be pushed more according to user The description information of media resource determines the classification of multimedia resource to be pushed;
The DVB radio network pushes front end, for inciting somebody to action when the multimedia resource wait push belongs to radio network and sends classification Multimedia resource to be pushed pushes to ground base station push front end or user terminal;
The telecommunications network IP plug-flow platform, for inciting somebody to action when the multimedia resource wait push belongs to telecommunications network and sends classification Multimedia resource to be pushed pushes to ground base station push front end;Wherein, the ground base station pushes front end, for will be wait push away The multimedia resource sent pushes to user terminal.
In one embodiment, the classifying content recommended engine is specifically used for:
When there is multiple ground base stations to cover multiple users, all users that each ground base station is covered are determined as one User group;
The relevant information that multimedia resource is watched according to user, determines the population characteristic of each user group;
According to the description information of each multimedia resource to be pushed, the statistics of multimedia resource each to be pushed is determined Feature;
The statistical nature of the population characteristic of each user group and multimedia resource each to be pushed is matched, is obtained Matching degree numerical value of each user group to each multimedia resource to be pushed;
It is ranked up matching degree numerical value that each user group treats the multimedia resource of push is descending, selected and sorted Multimedia resource to be pushed before the first default value generates the multimedia resource matching degree sequencing table of each user group;
Determine the institute of multimedia resource each to be pushed in the multimedia resource matching degree sequencing table of multiple user groups Accounting example;
Proportion is reached into the multimedia resource to be pushed of the first preset ratio, proportion is preset between first and compared Multimedia resource to be pushed between example and the second preset ratio is determined as radio network and sends classification;By proportion lower than the The multimedia resource to be pushed of two preset ratios is determined as telecommunications network and sends classification;Wherein, first preset ratio is greater than Second preset ratio;
DVB radio network push front end is specifically used for:
The multimedia resource to be pushed that proportion reaches the first preset ratio is pushed into user terminal;
By to be pushed multimedia resource push of the proportion between the first preset ratio and the second preset ratio Front end is pushed to ground base station;
The telecommunications network IP plug-flow platform is specifically used for:
Proportion is pushed into ground base station push front end lower than the multimedia resource to be pushed of the second preset ratio.
In one embodiment, the classifying content recommended engine also particularly useful for:
A multimedia resource table to be pushed is established for each user, includes in the multimedia resource table to be pushed Need the multimedia resource pushed;
The multimedia resource table for treating push is modified, and the received multimedia resource to be pushed of user is deleted;
The relevant information that multimedia resource is watched according to user, determines the user characteristics of each user;
Based on machine learning algorithm model, provided according to the user characteristics of each user, modified multimedia to be pushed The statistical nature of multimedia resource each to be pushed and machine learning algorithm model parameter in the table of source obtain modified wait push away Multimedia resource each to be pushed by favorable rating value in the multimedia resource table sent;
By multimedia resource each to be pushed in modified multimedia resource table to be pushed by favorable rating value It is descending to be ranked up;
To be pushed multimedia resource of the selected and sorted before the second default value is determined as telecommunications network and sends classification;
The telecommunications network IP plug-flow platform also particularly useful for:
To be pushed multimedia resource of the sequence before the second default value is pushed into ground base station push front end;
Wherein, the machine learning algorithm model parameter is determined by machine learning algorithm model.
In one embodiment, the classifying content recommended engine also particularly useful for:
Machine learning algorithm model is determined as follows:
By the statistical nature of each of user characteristics, modified multimedia resource table multimedia resource to be pushed and User matches the program request situation of each multimedia resource to be pushed, and obtains matching result;
According to matching result, the multimedia resource to be pushed of user's program request is formed into program request set, by the non-program request of user Multimedia resource to be pushed form non-program request set;
Program request set and non-program request set are merged into comprehensive collection, comprehensive collection is divided into training set and test set;
Verifying is trained to training set and test set based on machine learning algorithm, determines machine learning algorithm model.
In one embodiment, DVB radio network push front end includes segmentation module, check code setup module and push Module;
The segmentation module is used for when the multimedia resource wait push belongs to radio network and sends classification, will be each wait push Multimedia resource be divided into multistage, obtain multiple multimedia resource sections to be pushed;
The check code setup module is used to that a hash check code to be arranged for multimedia resource section each to be pushed;
The pushing module is used to push multiple multimedia resource sections to be pushed comprising hash check code.
In one embodiment, the telecommunications network IP plug-flow platform is also used to:
The information and request repeat for receiving the not complete received multimedia resource section to be pushed that user terminal is sent refer to It enables;The information of the not complete received multimedia resource section to be pushed, by user terminal to including the more of hash check code A multimedia resource section wait push determines after being verified;
According to the information and request repeat instruction of the not complete received multimedia resource section to be pushed, judge whether Have and does not receive the not complete received multimedia resource section to be pushed completely greater than the user of third default value, if so, Then indicate that the not complete received multimedia resource section to be pushed is pushed to user terminal by ground base station push front end;
If whether having at the other user terminals that no, judgement is connected on telecommunications network described not complete received wait push away The multimedia resource section sent, if so, then the other user terminals that are connected on telecommunications network of instruction will it is described it is complete it is received to The multimedia resource section of push pushes to user terminal;
If the not complete received multimedia resource section to be pushed directly no, is pushed to user terminal.
The embodiment of the invention provides a kind of multimedia resource method for pushing, different multimedia resources has been carried out point Class, can have the selection of foundation is to be pushed using radio network, or pushed using telecommunications network, to reach resource most Optimal sorting is matched.This method comprises:
Collect the relevant information that user watches multimedia resource;
Collect the description information of multimedia resource to be pushed;
According to user watch multimedia resource relevant information and multimedia resource to be pushed description information, determine to The classification of the multimedia resource of push;
When the multimedia resource wait push belongs to radio network and sends classification, multimedia resource to be pushed is passed through into DVB Radio network push front end pushes to user terminal, or multimedia resource to be pushed is pushed front end push by DVB radio network Front end is pushed to ground base station, user terminal is pushed to by ground base station push front end;
When the multimedia resource wait push belongs to telecommunications network and sends classification, multimedia resource to be pushed is passed through into telecommunications Net IP plug-flow platform pushes to ground base station push front end, pushes to user terminal by ground base station push front end.
In one embodiment, the relevant information and multimedia money to be pushed that multimedia resource is watched according to user The description information in source determines the classification of multimedia resource to be pushed, comprising:
When there is multiple ground base stations to cover multiple users, all users that each ground base station is covered are determined as one User group;
The relevant information that multimedia resource is watched according to user, determines the population characteristic of each user group;
According to the description information of each multimedia resource to be pushed, the statistics of multimedia resource each to be pushed is determined Feature;
The statistical nature of the population characteristic of each user group and multimedia resource each to be pushed is matched, is obtained Matching degree numerical value of each user group to each multimedia resource to be pushed;
It is ranked up matching degree numerical value that each user group treats the multimedia resource of push is descending, selected and sorted Multimedia resource to be pushed before the first default value generates the multimedia resource matching degree sequencing table of each user group;
Determine the institute of multimedia resource each to be pushed in the multimedia resource matching degree sequencing table of multiple user groups Accounting example;
Proportion is reached into the multimedia resource to be pushed of the first preset ratio, proportion is preset between first and compared Multimedia resource to be pushed between example and the second preset ratio is determined as radio network and sends classification;
Proportion is determined as telecommunications network lower than the multimedia resource to be pushed of the second preset ratio and sends classification;
Wherein, first preset ratio is greater than the second preset ratio;
When the multimedia resource wait push belongs to radio network transmission classification, multimedia resource to be pushed is passed through DVB radio network push front end pushes to ground base station push front end or user terminal, comprising:
Front end is pushed by DVB radio network to push away the multimedia resource to be pushed that proportion reaches the first preset ratio It send to user terminal;
By DVB radio network push front end by proportion between the first preset ratio and the second preset ratio to The multimedia resource of push pushes to ground base station push front end;
When the multimedia resource wait push belongs to telecommunications network transmission classification, multimedia resource to be pushed is passed through Telecommunications network IP plug-flow platform pushes to ground base station push front end, comprising:
Multimedia resource to be pushed by telecommunications network IP plug-flow platform by proportion lower than the second preset ratio pushes away It send to ground base station and pushes front end.
In one embodiment, the relevant information and multimedia money to be pushed that multimedia resource is watched according to user The description information in source determines the classification of multimedia resource to be pushed, further includes:
A multimedia resource table to be pushed is established for each user, includes in the multimedia resource table to be pushed Need the multimedia resource pushed;
The multimedia resource table for treating push is modified, and the received multimedia resource to be pushed of user is deleted;
The relevant information that multimedia resource is watched according to user, determines the user characteristics of each user;
Based on machine learning algorithm model, provided according to the user characteristics of each user, modified multimedia to be pushed The statistical nature of multimedia resource each to be pushed and machine learning algorithm model parameter in the table of source obtain modified wait push away Multimedia resource each to be pushed by favorable rating value in the multimedia resource table sent;
By multimedia resource each to be pushed in modified multimedia resource table to be pushed by favorable rating value It is descending to be ranked up;
To be pushed multimedia resource of the selected and sorted before the second default value is determined as telecommunications network and sends classification;
When the multimedia resource wait push belongs to telecommunications network transmission classification, multimedia resource to be pushed is passed through Telecommunications network IP plug-flow platform pushes to ground base station push front end, further includes:
To be pushed multimedia resource of the sequence before the second default value is pushed to by telecommunications network IP plug-flow platform Ground base station pushes front end;
Wherein, the machine learning algorithm model parameter is determined by machine learning algorithm model.
In one embodiment, the machine learning algorithm model determines as follows:
By the statistical nature of each of user characteristics, modified multimedia resource table multimedia resource to be pushed and User matches the program request situation of each multimedia resource to be pushed, and obtains matching result;
According to matching result, the multimedia resource to be pushed of user's program request is formed into program request set, by the non-program request of user Multimedia resource to be pushed form non-program request set;
Program request set and non-program request set are merged into comprehensive collection, comprehensive collection is divided into training set and test set;
Verifying is trained to training set and test set based on machine learning algorithm, determines machine learning algorithm model.
It in one embodiment, will be to be pushed more when the multimedia resource wait push belongs to radio network and sends classification Media resource pushes front end push by DVB radio network, comprising:
When the multimedia resource wait push belongs to radio network and sends classification, each multimedia resource to be pushed is divided into Multistage obtains multiple multimedia resource sections to be pushed;
For each multimedia resource section to be pushed, one hash check code is set;
Multiple multimedia resource sections to be pushed comprising hash check code are pushed into front end push by DVB radio network.
In one embodiment, this method further include:
The not complete received multimedia resource to be pushed that user terminal is sent is received by telecommunications network IP plug-flow platform Information and the request repeat instruction of section;The information of the not complete received multimedia resource section to be pushed, by user terminal It is determined after being verified to multiple multimedia resource sections wait push comprising hash check code;
According to the information and request repeat instruction of the not complete received multimedia resource section to be pushed, judge whether Have and does not receive the not complete received multimedia resource section to be pushed completely greater than the user of third default value, if so, Then indicate that the not complete received multimedia resource section to be pushed is pushed to user terminal by ground base station push front end;
If whether having at the other user terminals that no, judgement is connected on telecommunications network described not complete received wait push away The multimedia resource section sent, if so, then the other user terminals that are connected on telecommunications network of instruction will it is described it is complete it is received to The multimedia resource section of push pushes to user terminal;
If no, by telecommunications network IP plug-flow platform directly by the not complete received multimedia resource to be pushed Section pushes to user terminal.
In embodiments of the present invention, classifying content recommended engine according to user watch multimedia resource relevant information and to Multimedia resource to be pushed is divided into radio network and sends classification and telecommunications network transmission by the description information of the multimedia resource of push Classification pushes front end by DVB radio network and the multimedia resource to be pushed for belonging to radio network transmission classification is pushed to ground Base station pushes front end or user terminal, and more matchmakers to be pushed of telecommunications network transmission classification will be belonged to by telecommunications network IP plug-flow platform Body resource supplying can have the selection of foundation in this way due to being classified to multimedia resource to ground base station push front end It is to be pushed using radio network, or pushed using telecommunications network, to reach resource optimal allocation, so that network bandwidth Utilization rate is higher, and cost is lower.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with It obtains other drawings based on these drawings.
Fig. 1 is multimedia resource supplying system architecture diagram in the embodiment of the present invention;
Fig. 2 is the practical application schematic diagram of multimedia resource supplying system in the embodiment of the present invention;
Fig. 3 is the specific network topology structure schematic diagram of multimedia resource supplying system in the embodiment of the present invention;
Fig. 4 is user information collection device in the embodiment of the present invention, multimedia resource information collection apparatus and classifying content The concrete function schematic diagram of recommended engine;
Fig. 5 is a kind of specific multimedia resource classification method flow chart in the embodiment of the present invention;
Fig. 6 is a kind of specific multimedia resource classification method flow chart in the embodiment of the present invention;
Fig. 7 is the architecture diagram of user terminal in the embodiment of the present invention;
Fig. 8 is to be pushed out the mechanism for correcting errors schematic diagram staggered the time in the embodiment of the present invention to broadcasting network using telecommunication network;
Fig. 9 is the structural schematic diagram of DVB radio network push front end in the embodiment of the present invention;
Figure 10 is multimedia resource method for pushing flow chart in the embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that the described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.Based on this Embodiment in invention, every other reality obtained by those of ordinary skill in the art without making creative efforts Example is applied, shall fall within the protection scope of the present invention.
Existing video push technology, it is only simple that telecommunications network is come to pushing video in conjunction with radio network, but do not have Classify to different videos, not can determine which video is pushed using radio network, which video using telecommunications network into Row push, so as to reach resource optimal allocation.For above-mentioned problems of the prior art, mentioned in the embodiment of the present invention A kind of multimedia resource supplying system and method are gone out, different multimedia resources are classified, can selectively be made Multimedia resource push is carried out with radio network or telecommunications network, has reached resource optimal allocation, while network bandwidth can also be made Utilization rate is higher, and cost is lower.
Fig. 1 is multimedia resource supplying system architecture diagram in the embodiment of the present invention, as shown in Figure 1, the system includes one User information collection device 101, a multimedia resource information collection apparatus 102, a classifying content recommended engine 103, one A DVB radio network pushes front end 104, a telecommunications network IP plug-flow platform 105.
Wherein, user information collection device 101 watches the relevant information of multimedia resource for collecting user;
Multimedia resource information collection apparatus 102, for collecting the description information of multimedia resource to be pushed;
Classifying content recommended engine 103, for watching the relevant information of multimedia resource and to be pushed more according to user The description information of media resource determines the classification of multimedia resource to be pushed;
DVB radio network pushes front end 104, for inciting somebody to action when the multimedia resource wait push belongs to radio network and sends classification Multimedia resource to be pushed pushes to ground base station push front end or user terminal;
Telecommunications network IP plug-flow platform 105 will be to for when the multimedia resource wait push belongs to telecommunications network and sends classification The multimedia resource of push pushes to ground base station push front end;Wherein, ground base station push front end, for will be to be pushed it is more Media resource pushes to user terminal.
Fig. 2 is the practical application schematic diagram of multimedia resource supplying system in the embodiment of the present invention, as shown in Fig. 2, at this In example, which can also include a media asset management device 107, gather around for being responsible for management whole system Some multimedia resources (video resource, audio resource etc.) for needing to be pushed.DVB radio network push front end and telecommunications network IP are pushed away Levelling platform obtains multimedia resource to be pushed accordingly from media asset management device and is pushed.
As shown in Fig. 2, in this example, classifying content recommended engine 103 is provided with user information collection device 101 and multimedia The information that source information collection device 102 is collected into is input, so that the multimedia resource for treating push is classified.Classifying rules It can be configured according to actual needs, such as can be the more matchmakers for treating push according to temperature according to different countries and regions Body resource is ranked up, and selects all favorite multimedia resource of a part of various countries various regions user, pushes front end using DVB radio network 104 are sent directly to user terminal 106 (home media center) by satellite (DVB-S) or wired (DVB-C);Select a part For all favorite multimedia resource of some region of user, by satellite relay or made using DVB radio network push front end 104 After being transferred to ground base station push front end by cable broadband/optical fiber with telecommunications network IP plug-flow platform 105, pushed by ground base station Front end is broadcast to the user terminal 106 (home media center) of its covering by terrestrial wireless (DVB-T);Select last part For the multimedia resource of each user individual, telecommunication network (such as wireless network is utilized using telecommunications network IP plug-flow platform 105 3G, 4G-LTE, WIFI, cable broadband/optical fiber etc.) idle flow is pushed to user terminal 106 (home media center).Wherein, User terminal 106 (home media center) can receive DVB radio network signal and telecommunications network IP signal simultaneously.
Fig. 3 is the specific network topology structure schematic diagram of multimedia resource supplying system in the embodiment of the present invention, such as Fig. 3 institute Show, in this example, multimedia resource can be directly pushed in the media of user via satellite by DVB radio network push front end The heart, also can use satellite relay be sent to ground base station (after A to D) by ground base station broadcast be pushed to user.Meanwhile telecommunications Net IP plug-flow platform cable broadband can be used to connect with ground base station, carry out multimedia resource push.In addition, between ground base station Can also multimedia resource transmission be carried out using cable broadband connection mutually.
Fig. 4 is user information collection device in the embodiment of the present invention, multimedia resource information collection apparatus and classifying content The concrete function schematic diagram of recommended engine, as shown in figure 4, in this example, it is more that user information collection device 101 collects user's viewing The relevant information of media resource, the relevant information that user therein watches multimedia resource includes that user watches multimedia resource The information such as behavioural information, the attribute of user, hobby;Such as can primarily focus on the information of three aspects: DVB radio network TV is seen See the audience ratings of user, which uses the history program request behavior of system, and user is on the third-party platforms such as social networks, forum Information of interest.The phase that the user information collection device watches multimedia resource per collecting and updating user at regular intervals Information is closed, and is stored in the database of system, which is the database in classifying content recommended engine, and the database is same When be also stored with the description information of following described multimedia resource.
In this example, multimedia resource information collection apparatus 102 collects the description information of multimedia resource to be pushed, In multimedia resource to be pushed description information be multimedia resource to be pushed label, such as: language, director, drill Member, photographing region, subject matter etc., third party website is to the scoring of the multimedia resource and box-office income/album pin of film The information such as amount.
User is obtained to watch after the relevant information of multimedia resource and the description information of multimedia resource to be pushed, Classifying content recommended engine 103 can classify according to the multimedia resource that the two information treat push, and classification method has more Kind, it is a kind of specific example of multimedia resource classification method as shown in Figure 5, in this example, classifying content recommended engine 103 has Body as follows classifies to multimedia resource:
Step 501: when there is multiple ground base stations to cover multiple users, all users that each ground base station is covered are true It is set to a user group;
Step 502: watching the relevant information of multimedia resource according to user, determine the population characteristic of each user group;
Step 503: according to the description information of each multimedia resource to be pushed, determining multimedia money each to be pushed The statistical nature in source;
Step 504: the statistical nature of the population characteristic of each user group and multimedia resource each to be pushed is carried out Matching, obtains each user group to the matching degree numerical value of each multimedia resource to be pushed;
Step 505: it is ranked up matching degree numerical value that each user group treats the multimedia resource of push is descending, To be pushed multimedia resource of the selected and sorted before the first default value, generates the multimedia resource matching degree of each user group Sequencing table;
Step 506: determining multimedia each to be pushed in the multimedia resource matching degree sequencing table of multiple user groups The proportion of resource;
Step 507: multimedia resource to be pushed that proportion is reached the first preset ratio, proportion are between Multimedia resource to be pushed between one preset ratio and the second preset ratio is determined as radio network and sends classification;By institute's accounting Example is determined as telecommunications network lower than the multimedia resource to be pushed of the second preset ratio and sends classification;Wherein, the first preset ratio Greater than the second preset ratio;
DVB radio network push front end 104 is specifically used for:
The multimedia resource to be pushed that proportion reaches the first preset ratio is pushed into user terminal;
By to be pushed multimedia resource push of the proportion between the first preset ratio and the second preset ratio Front end is pushed to ground base station;
Telecommunications network IP plug-flow platform 105 is specifically used for:
Proportion is pushed into ground base station push front end lower than the multimedia resource to be pushed of the second preset ratio.
When it is implemented, as shown in figure 4, watching the relevant information and multimedia to be pushed of multimedia resource for user The description information of resource, classifying content recommended engine 103 do the pretreatment of data first, extract feature according to information above.It is right In the extracting method of single user's feature be diversified, such as: by sport, music, family, news, 5 yuan or less, 5-10 Member, English, multimedia resource label to be pushed as Portuguese are set as an one-dimensional vector, if User likes sport category, then the position for indicating sport is set 1, otherwise sets 0, will finally obtain one one-dimensional 0/1 of the user to Amount.For multimedia resource to be pushed, its statistical nature can also be determined using the method, also obtain one it is one-dimensional Vector.Pretreatment enters the substantial period of algorithm after finishing.
In order to distinguish aforementioned three kinds of described situations (nearly all user likes that regional total user is all liked, and User individual selection), it is useful with the institute of a ground base station covering first when there is multiple ground base stations to cover multiple users Family is a user group, gives a certain number of multimedia resources to be pushed;The correlation of multimedia resource is watched according to user Information determines the population characteristic of each user group, the population characteristic extraction method of user group and the extraction side of single user's feature Method is similar, but population characteristic indicate be the multimedia resource label to be pushed to each percentage, numerical value higher generation Such resource (such as sport 80%, family 40%, English 70% etc.) is more liked by table this area, and the final population characteristic is one one Dimensional vector, length is identical as the length of the one-dimensional vector for the statistical nature for indicating multimedia resource to be pushed, but value from Non-zero i.e. 1 becomes a decimal between 0-1.
After the statistical nature of the population characteristic and multimedia resource each to be pushed that have obtained each user group, The statistical nature of the population characteristic of each user group and multimedia resource each to be pushed is matched, matching process is more Kind multiplicity, such as two one-dimensional vectors can be done into inner product operation, each user group is obtained to each multimedia to be pushed The matching degree numerical value of resource.For each user group, each user group is calculated to all multimedia resources to be pushed Matching degree numerical value, by these, matching degree numerical value is descending is ranked up, (n+m) % (the first default value) matching degree before selecting Highest multimedia resource to be pushed forms the multimedia resource matching degree sequencing table of each user group.Wherein n indicates final It is desirable for the percentage for the multimedia resource that DVB radio network push front end is pushed, representative value is 20 or 30, and m indicates one A compensation amount of relaxing, representative value 5.
For multiple user groups, multiple multimedia resource matching degree sequencing tables can be equally generated, multiple user groups pair are counted The each multimedia resource in multiple multimedia resource matching degree sequencing tables answered sorts in all multimedia resource matching degrees The proportion of table, if nearly all user group all has demand, such as percentage (the first default ratio to the multimedia resource Example) reach 95% or more, then user directly is pushed directly to using satellite and DVB-S agreement using DVB radio network push front end Terminal (home media center).If being not that nearly all user group has demand, but still has the (allusion quotation of k of k% or more Offset is that 25) user group has demand, and DVB radio network push front end may be used by satellite relay and be pushed to ground base station push Front end is connect using DVB-T protocol broadcast to its user covered after having the ground base station push front end receiver of push demand It receives.If the user group only less than k% has demand, then telecommunications network IP plug-flow platform can be used by way of cable broadband Corresponding multimedia resource is transmitted to ground base station by IP agreement and pushes front end, without the use of satellite.Ground base station push Front end is equally pushed after receiving multimedia resource.
Illustrate the classification and result of multimedia resource below by the mode that list data is enumerated.
Select multimedia resource to be pushed a total of 21, respectively video 1- video 21, it is assumed that each user's Signature analysis is ready for, and can have each user to like one of each video the list of degree.Such as a certain User to video 1- video 21 to like degree as shown in table 1.
Table 1
Assuming that thinking that the user likes more than or equal to 0.5, think that the user does not like less than 0.5, then it is available A table as shown in Table 2.
Table 2
To identical video summation in this table (table 2) of user all inside a user group, then used divided by total Amount, distribution of the available user group to the favorable rating of a certain video.
Be set with 7 user groups, respectively user group 1 to user group 7, then according to the method described above determine user group 1 to User group 7 is to the respective matching degree numerical value (could also say that favorable rating) of video 1- video 21, as shown in table 3:
Table 3
Then, if the first default value is 25%, i.e. n is 20, m 5,21*25%=5.25, exactly takes sequence preceding 5 A video, user group 1 are as shown in table 4 to the corresponding multimedia resource matching degree sequencing table of user group 7:
Table 4
User group 1 User group 2 User group 3 User group 4 User group 5 User group 6 User group 7
Video 1 Video 1 Video 1 Video 1 Video 1 Video 1 Video 1
Video 9 Video 9 Video 8 Video 6 Video 6 Video 2 Video 4
Video 14 Video 5 Video 9 Video 17 Video 14 Video 16 Video 10
Video 2 Video 15 Video 13 Video 13 Video 5 Video 6 Video 19
Video 8 Video 3 Video 19 Video 21 Video 13 Video 10 Video 8
In user group 1 into the corresponding multimedia resource matching degree sequencing table of user group 7, each video institute accounting is determined Example, being respectively as follows: video 1 is 100%, and video 2 is about 28.6% (2/7), and video 3 is about 14.3% (1/7), and video 4 is about 14.3% (1/7), video 5 are about 28.6% (2/7), and video 6 is about 42.9% (3/7), and video 8 is about 42.9% (3/7), depending on Frequently 9 about 42.9% (3/7), video 10 are about 28.6% (2/7), and video 13 is about 42.9% (3/7), and video 14 is about 28.6% (2/7), video 15 are about 14.3% (1/7), and video 16 is about 14.3% (1/7), and video 17 is about 14.3% (1/ 7), video 19 is about 28.6% (2/7), and video 21 is about 14.3% (1/7).
Furthermore, it is assumed that the first preset ratio is 95%, and the second preset ratio is 25%.Up to or over 95% only Video 1, it is determined that video 1 is that radio network sends classification, which pushes front end by DVB radio network and directly push to user Terminal.
Video between 25% to 95% has: video 2, video 5, video 6, video 8, video 9, video 10, video 13, video 14, video 19 determine that these videos equally also belong to radio network and send classification, but these video councils are wide by DVB It broadcasts net push front end and pushes to ground base station push front end, be then pushed to user terminal again.
Video lower than 25% has: video 3, video 4, video 15, video 16, video 17, video 21 determine these videos Belong to telecommunications network and send classification, these video councils push to ground base station by telecommunications network IP plug-flow platform and push front end, then It is pushed to user terminal again.
After the above logic judgment finishes, each user terminal (home media center) is had been allocated that Multimedia resource to be pushed.Other than the above multimedia resource, classifying content recommended engine is also needed according to a certain user Interest carry out personalized push.In this example, classifying content recommended engine can be according to multimedia resource classification side shown in fig. 6 Method is classified:
Step 601: a multimedia resource table to be pushed is established for each user, in multimedia resource table to be pushed Including needing the multimedia resource pushed;
Step 602: the multimedia resource table for treating push is modified, and the received multimedia to be pushed of user is provided It deletes in source;
Step 603: watching the relevant information of multimedia resource according to user, determine the user characteristics of each user;
Step 604: machine learning algorithm model is based on, according to the user characteristics of each user, modified to be pushed The statistical nature of multimedia resource each to be pushed and machine learning algorithm model parameter, are modified in multimedia resource table Multimedia resource each to be pushed by favorable rating value in the multimedia resource table of afterwards to be pushed;
Step 605: by being liked for multimedia resource each to be pushed in modified multimedia resource table to be pushed Like that degree value is descending to be ranked up;
Step 606: to be pushed multimedia resource of the selected and sorted before the second default value is determined as telecommunications network transmission Classification;
Telecommunications network IP plug-flow platform also particularly useful for:
To be pushed multimedia resource of the sequence before the second default value is pushed into ground base station push front end;
Wherein, machine learning algorithm model parameter is determined by machine learning algorithm model.
When it is implemented, in order to from modified multimedia resource table to be pushed sequence obtain user and most like to obtain Multimedia resource, need the algorithm using machine learning.Firstly the need of determination machine learning algorithm mould used in the present invention Type, the machine learning algorithm model determine that steps are as follows:
Firstly, by the statistics of each of user characteristics, modified multimedia resource table multimedia resource to be pushed Feature and the user to the program request situation (finally whether having selected the multimedia resource) of each multimedia resource to be pushed into Row matching, obtains matching result;
Then, according to matching result, the multimedia resource to be pushed of user's program request is formed into program request set, not by user The multimedia resource to be pushed of program request forms non-program request set;Program request set and non-program request set are merged, obtain one A comprehensive collection, is then divided into test set and training set for comprehensive collection.For example, can choose the comprehensive content for concentrating 60% is used as instruction Practice collection, selects the content of comprehensive concentration 40% as test set.
Then, choose typical machine learning algorithm (such as SVM, decision tree, artificial neural network, deep learning etc.) to On training set and the data of test set be trained and do cross validation, can obtain for the user machine learning model and Its machine learning algorithm model parameter.The machine learning algorithm model parameter and user's pairing are stored in database profession.
After machine learning algorithm model used in the present invention has been determined, it is directed to modified more matchmakers to be pushed every time The multimedia resource increased newly in body resource table, according to its feature, the machine learning algorithm of user characteristics and preservation in the database Model parameter calls same machine learning algorithm model to be calculated, can obtain the multimedia resource by favorable rating value; This new multimedia resource is added in modified multimedia resource table to be pushed and others are to be pushed in table Multimedia resource sorts together, and the multimedia resource that can choose f% (artificial regulation) uses the idle flow pair of telecommunication network User pushes.
It illustrates.For some user A in user group 1, multimedia resource table to be pushed includes video 1 to video 21, after above-mentioned push, user group 1 has received 5 videos, respectively video 1, video 9, video 14, video 2, view Frequently 8.These videos are deleted from multimedia resource table to be pushed, include in modified multimedia resource table to be pushed 16 videos are respectively as follows: video 3, video 4, video 5, video 6, video 7, video 10, video 11, video 12, video 13, view Frequently 15, video 16, video 17, video 18, video 19, video 20, video 21.It calculates user A and journey is liked to this 16 videos Degree, is ranked up it, if last ranking results are as follows: video 17, video 3, video 7, video 4, video 15, video 18, view Frequently 5, video 6, video 12, video 13, video 19, video 20, video 16, video 21, video 10, video 11, select 25% Multimedia resource is exactly 16*25%=4 video, i.e. video 17, video 3, video 7, video 4 before selecting, passes through telecommunication network Idle flow user is pushed.
When it is implemented, Fig. 7 is the hardware architecture diagram of user terminal (home media center), as shown in fig. 7, in this example In, user terminal (home media center) 106 includes a CPU, memory, audio/video decoding chip, descrambling module and one Hard disk.In terms of interface, it includes telecommunication network interface, such as a WiFi chip, an Ethernet interface, a 3G/4G-LTE Modem modem (or interface for supporting external 3G/4G-LTE modem, such as USB).It is also wrapped simultaneously The interface of broadcasting network containing DVB, such as DVB-C tuner, DVB-S tuner, DVB-T tuner.
Fig. 8 is that the mechanism for correcting errors staggered the time is pushed out to broadcasting network using telecommunication network.As shown in figure 8, in this example, it is first First multimedia resource to be pushed can be segmented and school is set when using DVB radio network push front end push multimedia resource Test code.For the above, it may include: segmentation module 901 that DVB radio network, which pushes front end 104, for that will belong to radio network hair The multimedia resource for sending each of classification to be pushed is divided into multistage (such as mono- section of 4KB), obtains multiple multimedia resources to be pushed Section;Check code setup module 902 is such as utilized for a hash check code to be arranged for multimedia resource section each to be pushed Sha1 or MD5 etc., these hash check codes form a hash and verify code character, can be placed in video file head;Pushing module 903, for will include that multiple multimedia resource sections to be pushed of hash check code push to user terminal, as shown in Figure 9.
After DVB radio network push front end push, user terminal can be received comprising the multiple wait push away of hash check code The multimedia resource section sent, then verifies it, determines whether each multimedia resource section receives completely, finally obtains one Which multimedia resource section a table, which represents, completely receives, which multimedia resource section does not receive completely.If in the presence of not having Complete received multimedia resource section can detect whether that telecommunications network IP plug-flow platform can be connected by telecommunication network, if so, by The information of the multimedia resource section of complete downloading (or perhaps downloading successfully) or the multimedia resource for not downloading (reception) completely Information and request repeat the instruction report of section are waited to connect to telecommunications network IP plug-flow platform if not, request is put into queue Telecommunications network retransmits after finding telecommunications network IP plug-flow platform.Then, telecommunications network IP plug-flow platform can according to it is complete it is received to Whether the information and request repeat of the multimedia resource section of push instruct, first determine whether to have at ground base station push front end not complete Received multimedia resource section to be pushed, if so, judging whether that the area of some ground base station covering is more than or equal to one again The same multimedia resource section of the user of fixed number amount (k%) reception failure (loss) simultaneously, if so, being pushed using ground base station Not complete received multimedia resource section to be pushed is pushed to user terminal by front end;If ground base station push does not have at front end It is same that the area of not complete received multimedia resource section to be pushed or the covering of a ground base station is less than a certain number of users When the same multimedia resource section of reception failure, then it is complete received to be pushed front end push or not by ground base station Multimedia resource section, the other users for needing to be connected to using P2P (Peer to Peer) protocol detection on telecommunications network at this time are whole The multimedia resource section (not complete received multimedia resource section to be pushed) for whether thering is it to need at end, if so, then utilizing P2P agreement obtains not complete received multimedia resource section to be pushed from the other user terminals being connected on telecommunications network; If the telecommunications network IP plug-flow platform that no, then be used directly is complete received to be pushed more for the point-to-point push of each user Media resource section.Two kinds of sending methods can make user terminal rejudge whether multimedia resource section receives completely, and after Continue subsequent process, terminates after completely receiving entire multimedia resource.
Based on the same inventive concept, a kind of multimedia resource method for pushing is additionally provided in the embodiment of the present invention, it is such as following Embodiment described in.Since the principle that multimedia resource method for pushing solves the problems, such as is similar to multimedia resource supplying system, because The implementation of this multimedia resource method for pushing may refer to the implementation of multimedia resource supplying system, and overlaps will not be repeated. Terms used above " unit " or " module " can be the combination for realizing the software and/or hardware of predetermined function.Although Device described in above embodiments is preferably realized with software, but the combined realization of hardware or software and hardware And may and be contemplated.
Figure 10 is multimedia resource method for pushing flow chart in the embodiment of the present invention, as shown in Figure 10, the multimedia resource Method for pushing includes:
Step 1001: collecting the relevant information that user watches multimedia resource;
Step 1002: collecting the description information of multimedia resource to be pushed;
Step 1003: being believed according to the description that user watches the relevant information and multimedia resource to be pushed of multimedia resource Breath determines the classification of multimedia resource to be pushed;
Step 1004: when the multimedia resource wait push belongs to radio network and sends classification, multimedia to be pushed being provided Source pushes front end by DVB radio network and pushes to ground base station push front end or user terminal;
When the multimedia resource wait push belongs to telecommunications network and sends classification, multimedia resource to be pushed is passed through into telecommunications Net IP plug-flow platform pushes to ground base station push front end;
Step 1005: multimedia resource to be pushed being pushed into front end by ground base station and pushes to user terminal.
When it is implemented, step 1003 can be divided according to the multimedia resource that method and step shown in fig. 5 treats push Class specifically includes:
When there is multiple ground base stations to cover multiple users, all users that each ground base station is covered are determined as one User group;
The relevant information that multimedia resource is watched according to user, determines the population characteristic of each user group;
According to the description information of each multimedia resource to be pushed, the statistics of multimedia resource each to be pushed is determined Feature;
The statistical nature of the population characteristic of each user group and multimedia resource each to be pushed is matched, is obtained Matching degree numerical value of each user group to each multimedia resource to be pushed;
It is ranked up matching degree numerical value that each user group treats the multimedia resource of push is descending, selected and sorted Multimedia resource to be pushed before the first default value generates the multimedia resource matching degree sequencing table of each user group;
Determine the institute of multimedia resource each to be pushed in the multimedia resource matching degree sequencing table of multiple user groups Accounting example;
Proportion is reached into the multimedia resource to be pushed of the first preset ratio, proportion is preset between first and compared Multimedia resource to be pushed between example and the second preset ratio is determined as radio network and sends classification;By proportion lower than the The multimedia resource to be pushed of two preset ratios is determined as telecommunications network and sends classification;Wherein, the first preset ratio is greater than second Preset ratio;
When the multimedia resource wait push belongs to radio network and sends classification, multimedia resource to be pushed is passed through into DVB Radio network push front end pushes to ground base station push front end or user terminal, comprising:
Front end is pushed by DVB radio network to push away the multimedia resource to be pushed that proportion reaches the first preset ratio It send to user terminal;
By DVB radio network push front end by proportion between the first preset ratio and the second preset ratio to The multimedia resource of push pushes to ground base station push front end;
When the multimedia resource wait push belongs to telecommunications network and sends classification, multimedia resource to be pushed is passed through into telecommunications Net IP plug-flow platform pushes to ground base station push front end, comprising:
Multimedia resource to be pushed by telecommunications network IP plug-flow platform by proportion lower than the second preset ratio pushes away It send to ground base station and pushes front end.
When it is implemented, step 1003 can also carry out personalized point to multimedia resource according to method and step shown in fig. 6 Class may include:
A multimedia resource table to be pushed is established for each user, includes all in multimedia resource table to be pushed Multimedia resource to be pushed;
The multimedia resource table for treating push is modified, and the received multimedia resource to be pushed of user is deleted;
The relevant information that multimedia resource is watched according to user, determines the user characteristics of each user;
Based on machine learning algorithm model, provided according to the user characteristics of each user, modified multimedia to be pushed The statistical nature of multimedia resource each to be pushed and machine learning algorithm model parameter in the table of source obtain modified wait push away Multimedia resource each to be pushed by favorable rating value in the multimedia resource table sent;
By multimedia resource each to be pushed in modified multimedia resource table to be pushed by favorable rating value It is descending to be ranked up;
To be pushed multimedia resource of the selected and sorted before the second default value is determined as telecommunications network and sends classification;
When the multimedia resource wait push belongs to telecommunications network and sends classification, multimedia resource to be pushed is passed through into telecommunications Net IP plug-flow platform pushes to ground base station push front end, can also include:
To be pushed multimedia resource of the sequence before the second default value is pushed to by telecommunications network IP plug-flow platform Ground base station pushes front end;
Wherein, machine learning algorithm model parameter is determined by machine learning algorithm model.
When it is implemented, machine learning algorithm model determines as follows:
By the statistical nature of each of user characteristics, modified multimedia resource table multimedia resource to be pushed and User matches the program request situation of each multimedia resource to be pushed, and obtains matching result;
According to matching result, the multimedia resource to be pushed of user's program request is formed into program request set, by the non-program request of user Multimedia resource to be pushed form non-program request set;
Program request set and non-program request set are merged into comprehensive collection, comprehensive collection is divided into training set and test set;
Verifying is trained to training set and test set based on machine learning algorithm, determines machine learning algorithm model.
When it is implemented, when pushing to multimedia resource, which can also include:
When the multimedia resource wait push belongs to radio network and sends classification, each multimedia resource to be pushed is divided into Multistage obtains multiple multimedia resource sections to be pushed;
For each multimedia resource section to be pushed, one hash check code is set;
Multiple multimedia resource sections to be pushed comprising hash check code are pushed into front end push by DVB radio network.
When it is implemented, after receiving multimedia resource, which can also include:
The not complete received multimedia resource to be pushed that user terminal is sent is received by telecommunications network IP plug-flow platform Information and the request repeat instruction of section;The information of the not complete received multimedia resource section to be pushed, by user terminal It is determined after being verified to multiple multimedia resource sections wait push comprising hash check code;
According to the information and request repeat instruction of not complete received multimedia resource section to be pushed, judge whether there is big The not complete received multimedia resource section to be pushed is not received completely in the user of third default value, if so, then referring to Show that ground base station push front end will be as corresponding not complete received multimedia resource section to be pushed pushes to user terminal;
If whether having at the other user terminals that no, judgement is connected on telecommunications network not complete received to be pushed Multimedia resource section, if so, then the other user terminals that are connected on telecommunications network of instruction will be complete received to be pushed more Media resource section pushes to user terminal;
If no, directly not complete received multimedia resource section to be pushed is pushed away by telecommunications network IP plug-flow platform It send to user terminal.
In conclusion the present invention has the advantage that
1, compared with traditional broadcast network push vod scheme, system proposed by the present invention can for different users into Row personalized recommendation.
2, with traditional broadcast net compared with the scheme of net cast or push is done in telecommunications network fusion, system proposed by the present invention Multimedia resource can be divided into 3 classes and be pushed using different technologies by system, so that network bandwidth utilization factor is higher, cost is lower.
3, system proposed by the present invention does not need user and reserves in advance, and then waiting could watch, current system support User is wanted that the video seen exists in advance on the hard disk of media center by the interest of look-ahead user.Even if in addition, if video Not on hard disk, it can directly pass through telecommunication network streaming media.
4, for DVB broadcasting network push failure the case where (for example the electricity went off to half user man by push) or part are believed Breath can be re-broadcast the remaining file of push using ground base station with intelligent selection or be utilized telecommunication network due to error code reception failure Point-to-point push.
5, system proposed by the present invention can be accustomed to recommending corresponding video according to the practical viewing in user's use, make to recommend It is more accurate.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more, The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates, Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one The step of function of being specified in a box or multiple boxes.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field For art personnel, the embodiment of the present invention can have various modifications and variations.All within the spirits and principles of the present invention, made Any modification, equivalent substitution, improvement and etc. should all be included in the protection scope of the present invention.

Claims (10)

1. a kind of multimedia resource supplying system characterized by comprising user information collection device, multimedia resource information Collection device, classifying content recommended engine, DVB radio network push front end, telecommunications network IP plug-flow platform;
The user information collection device watches the relevant information of multimedia resource for collecting user;
The multimedia resource information collection apparatus, for collecting the description information of multimedia resource to be pushed;
The classifying content recommended engine, for watching the relevant information and multimedia to be pushed of multimedia resource according to user The description information of resource determines the classification of multimedia resource to be pushed;
The DVB radio network pushes front end, will be wait push away for when the multimedia resource wait push belongs to radio network and sends classification The multimedia resource sent pushes to ground base station push front end or user terminal;
The telecommunications network IP plug-flow platform, for the multimedia resource wait push belong to telecommunications network send classification when, will be wait push away The multimedia resource sent pushes to ground base station push front end;Wherein, the ground base station pushes front end, for will be to be pushed Multimedia resource pushes to user terminal;
The classifying content recommended engine is specifically used for:
When there is multiple ground base stations to cover multiple users, all users that each ground base station is covered are determined as a user Group;
The relevant information that multimedia resource is watched according to user, determines the population characteristic of each user group;
According to the description information of each multimedia resource to be pushed, determine that the statistics of multimedia resource each to be pushed is special Sign;
The statistical nature of the population characteristic of each user group and multimedia resource each to be pushed is matched, is obtained each Matching degree numerical value of the user group to each multimedia resource to be pushed;
It is ranked up matching degree numerical value that each user group treats the multimedia resource of push is descending, selected and sorted is Multimedia resource to be pushed before one default value generates the multimedia resource matching degree sequencing table of each user group;
Determine institute's accounting of multimedia resource each to be pushed in the multimedia resource matching degree sequencing table of multiple user groups Example;
By proportion reach the multimedia resource to be pushed of the first preset ratio, proportion between the first preset ratio and Multimedia resource to be pushed between second preset ratio is determined as radio network and sends classification;Proportion is pre- lower than second If the multimedia resource to be pushed of ratio, which is determined as telecommunications network, sends classification;Wherein, first preset ratio is greater than second Preset ratio;
DVB radio network push front end is specifically used for:
The multimedia resource to be pushed that proportion reaches the first preset ratio is pushed into user terminal;
To be pushed multimedia resource of the proportion between the first preset ratio and the second preset ratio is pushed to ground Face base station pushes front end;
The telecommunications network IP plug-flow platform is specifically used for:
Proportion is pushed into ground base station push front end lower than the multimedia resource to be pushed of the second preset ratio.
2. multimedia resource supplying system as described in claim 1, which is characterized in that the classifying content recommended engine also has Body is used for:
A multimedia resource table to be pushed is established for each user, includes all in the multimedia resource table to be pushed Multimedia resource to be pushed;
The multimedia resource table for treating push is modified, and the received multimedia resource to be pushed of user is deleted;
The relevant information that multimedia resource is watched according to user, determines the user characteristics of each user;
Based on machine learning algorithm model, according to the user characteristics of each user, modified multimedia resource table to be pushed In each multimedia resource to be pushed statistical nature and machine learning algorithm model parameter, obtain modified to be pushed Multimedia resource each to be pushed by favorable rating value in multimedia resource table;
By multimedia resource each to be pushed in modified multimedia resource table to be pushed by favorable rating value by big It is ranked up to small;
To be pushed multimedia resource of the selected and sorted before the second default value is determined as telecommunications network and sends classification;
The telecommunications network IP plug-flow platform also particularly useful for:
To be pushed multimedia resource of the sequence before the second default value is pushed into ground base station push front end;
Wherein, the machine learning algorithm model parameter is determined by machine learning algorithm model.
3. multimedia resource supplying system as claimed in claim 2, which is characterized in that the classifying content recommended engine also has Body is used for:
Machine learning algorithm model is determined as follows:
By the statistical nature of each of user characteristics, modified multimedia resource table multimedia resource to be pushed and user The program request situation of each multimedia resource to be pushed is matched, matching result is obtained;
According to matching result, the multimedia resource to be pushed of user's program request is formed into program request set, by the non-program request of user to The multimedia resource of push forms non-program request set;
Program request set and non-program request set are merged into comprehensive collection, comprehensive collection is divided into training set and test set;
Verifying is trained to training set and test set based on machine learning algorithm, determines machine learning algorithm model.
4. multimedia resource supplying system as described in claim 1, which is characterized in that DVB radio network push front end packet Include segmentation module, check code setup module and pushing module;
The segmentation module is used for when the multimedia resource wait push belongs to radio network and sends classification, will be each to be pushed more Media resource is divided into multistage, obtains multiple multimedia resource sections to be pushed;
The check code setup module is used to that a hash check code to be arranged for multimedia resource section each to be pushed;
The pushing module is used to push multiple multimedia resource sections to be pushed comprising hash check code.
5. multimedia resource supplying system as claimed in claim 4, which is characterized in that the telecommunications network IP plug-flow platform is also used In:
Receive the information and request repeat instruction of the not complete received multimedia resource section to be pushed that user terminal is sent;Institute The information for stating not complete received multimedia resource section to be pushed, by user terminal to including the multiple wait push away of hash check code The multimedia resource section sent determines after being verified;
According to the information and request repeat instruction of the not complete received multimedia resource section to be pushed, judge whether there is big The not complete received multimedia resource section to be pushed is not received completely in the user of third default value, if so, then referring to Show that the not complete received multimedia resource section to be pushed is pushed to user terminal by ground base station push front end;
If whether having at the other user terminals that no, judgement is connected on telecommunications network described not complete received to be pushed Multimedia resource section, if so, the other user terminals that then instruction is connected on telecommunications network will be described not complete received wait push Multimedia resource section push to user terminal;
If the not complete received multimedia resource section to be pushed directly no, is pushed to user terminal.
6. a kind of multimedia resource method for pushing characterized by comprising
Collect the relevant information that user watches multimedia resource;
Collect the description information of multimedia resource to be pushed;
The relevant information of multimedia resource and the description information of multimedia resource to be pushed are watched according to user, are determined wait push Multimedia resource classification;
When the multimedia resource wait push belongs to radio network and sends classification, multimedia resource to be pushed is broadcasted by DVB Net push front end pushes to user terminal, or multimedia resource to be pushed is pushed front end by DVB radio network and pushes to ground Face base station pushes front end, pushes to user terminal by ground base station push front end;
When the multimedia resource wait push belongs to telecommunications network and sends classification, multimedia resource to be pushed is passed through into telecommunications network IP Plug-flow platform pushes to ground base station push front end, pushes to user terminal by ground base station push front end;
It is described according to user watch multimedia resource relevant information and multimedia resource to be pushed description information, determine to The classification of the multimedia resource of push, comprising:
When there is multiple ground base stations to cover multiple users, all users that each ground base station is covered are determined as a user Group;
The relevant information that multimedia resource is watched according to user, determines the population characteristic of each user group;
According to the description information of each multimedia resource to be pushed, determine that the statistics of multimedia resource each to be pushed is special Sign;
The statistical nature of the population characteristic of each user group and multimedia resource each to be pushed is matched, is obtained each Matching degree numerical value of the user group to each multimedia resource to be pushed;
It is ranked up matching degree numerical value that each user group treats the multimedia resource of push is descending, selected and sorted is Multimedia resource to be pushed before one default value generates the multimedia resource matching degree sequencing table of each user group;
Determine institute's accounting of multimedia resource each to be pushed in the multimedia resource matching degree sequencing table of multiple user groups Example;
By proportion reach the multimedia resource to be pushed of the first preset ratio, proportion between the first preset ratio and Multimedia resource to be pushed between second preset ratio is determined as radio network and sends classification;
Proportion is determined as telecommunications network lower than the multimedia resource to be pushed of the second preset ratio and sends classification;
Wherein, first preset ratio is greater than the second preset ratio;
When the multimedia resource wait push belongs to radio network transmission classification, multimedia resource to be pushed is passed through into DVB Radio network push front end pushes to ground base station push front end or user terminal, comprising:
Front end is pushed by DVB radio network to push to the multimedia resource to be pushed that proportion reaches the first preset ratio User terminal;
By DVB radio network push front end by proportion between the first preset ratio and the second preset ratio wait push Multimedia resource push to ground base station push front end;
When the multimedia resource wait push belongs to telecommunications network transmission classification, multimedia resource to be pushed is passed through into telecommunications Net IP plug-flow platform pushes to ground base station push front end, comprising:
Multimedia resource to be pushed by telecommunications network IP plug-flow platform by proportion lower than the second preset ratio pushes to Ground base station pushes front end.
7. multimedia resource method for pushing as claimed in claim 6, which is characterized in that described to watch multimedia money according to user The description information of the relevant information in source and multimedia resource to be pushed determines the classification of multimedia resource to be pushed, also wraps It includes:
A multimedia resource table to be pushed is established for each user, includes all in the multimedia resource table to be pushed Multimedia resource to be pushed;
The multimedia resource table for treating push is modified, and the received multimedia resource to be pushed of user is deleted;
The relevant information that multimedia resource is watched according to user, determines the user characteristics of each user;
Based on machine learning algorithm model, according to the user characteristics of each user, modified multimedia resource table to be pushed In each multimedia resource to be pushed statistical nature and machine learning algorithm model parameter, obtain modified to be pushed Multimedia resource each to be pushed by favorable rating value in multimedia resource table;
By multimedia resource each to be pushed in modified multimedia resource table to be pushed by favorable rating value by big It is ranked up to small;
To be pushed multimedia resource of the selected and sorted before the second default value is determined as telecommunications network and sends classification;
When the multimedia resource wait push belongs to telecommunications network transmission classification, multimedia resource to be pushed is passed through into telecommunications Net IP plug-flow platform pushes to ground base station push front end, further includes:
To be pushed multimedia resource of the sequence before the second default value is pushed into ground by telecommunications network IP plug-flow platform Base station pushes front end;
Wherein, the machine learning algorithm model parameter is determined by machine learning algorithm model.
8. multimedia resource method for pushing as claimed in claim 7, which is characterized in that the machine learning algorithm model according to As under type determines:
By the statistical nature of each of user characteristics, modified multimedia resource table multimedia resource to be pushed and user The program request situation of each multimedia resource to be pushed is matched, matching result is obtained;
According to matching result, the multimedia resource to be pushed of user's program request is formed into program request set, by the non-program request of user to The multimedia resource of push forms non-program request set;
Program request set and non-program request set are merged into comprehensive collection, comprehensive collection is divided into training set and test set;
Verifying is trained to training set and test set based on machine learning algorithm, determines machine learning algorithm model.
9. multimedia resource method for pushing as claimed in claim 6, which is characterized in that belong in multimedia resource to be pushed When radio network sends classification, multimedia resource to be pushed is pushed into front end push by DVB radio network, comprising:
When the multimedia resource wait push belongs to radio network and sends classification, each multimedia resource to be pushed is divided into more Section obtains multiple multimedia resource sections to be pushed;
For each multimedia resource section to be pushed, one hash check code is set;
Multiple multimedia resource sections to be pushed comprising hash check code are pushed into front end push by DVB radio network.
10. multimedia resource method for pushing as claimed in claim 9, which is characterized in that further include:
The not complete received multimedia resource section to be pushed that user terminal is sent is received by telecommunications network IP plug-flow platform Information and request repeat instruction;The information of the not complete received multimedia resource section to be pushed, by user terminal to packet Multiple multimedia resource sections wait push of the check code containing hash determine after being verified;
According to the information and request repeat instruction of the not complete received multimedia resource section to be pushed, judge whether there is big The not complete received multimedia resource section to be pushed is not received completely in the user of third default value, if so, then referring to Show that the not complete received multimedia resource section to be pushed is pushed to user terminal by ground base station push front end;
If whether having at the other user terminals that no, judgement is connected on telecommunications network described not complete received to be pushed Multimedia resource section, if so, the other user terminals that then instruction is connected on telecommunications network will be described not complete received wait push Multimedia resource section push to user terminal;
If no, directly the not complete received multimedia resource section to be pushed is pushed away by telecommunications network IP plug-flow platform It send to user terminal.
CN201610129990.8A 2016-03-08 2016-03-08 A kind of multimedia resource supplying system and method Active CN105791910B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610129990.8A CN105791910B (en) 2016-03-08 2016-03-08 A kind of multimedia resource supplying system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610129990.8A CN105791910B (en) 2016-03-08 2016-03-08 A kind of multimedia resource supplying system and method

Publications (2)

Publication Number Publication Date
CN105791910A CN105791910A (en) 2016-07-20
CN105791910B true CN105791910B (en) 2019-02-12

Family

ID=56388139

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610129990.8A Active CN105791910B (en) 2016-03-08 2016-03-08 A kind of multimedia resource supplying system and method

Country Status (1)

Country Link
CN (1) CN105791910B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106790275B (en) * 2017-02-20 2020-02-14 青岛海信电器股份有限公司 Multimedia data pushing method and device
CN106973315B (en) * 2017-04-07 2023-11-21 广西广播电视信息网络股份有限公司 Broadcast television network multichannel intelligent scheduling transmission system and method based on user requirements
US20180336645A1 (en) * 2017-05-22 2018-11-22 Google Inc. Using machine learning to recommend live-stream content
CN107277568A (en) * 2017-08-16 2017-10-20 广州市千钧网络科技有限公司 A kind of plug-flow configuration parameter dynamic adjusting method and device
CN110209920A (en) * 2018-05-02 2019-09-06 腾讯科技(深圳)有限公司 Treating method and apparatus, storage medium and the electronic device of media resource
CN110536157B (en) * 2018-05-23 2022-01-04 深圳Tcl数字技术有限公司 Method for automatically distributing television programs, storage medium and television terminal
CN108920385A (en) * 2018-09-18 2018-11-30 四川长虹电器股份有限公司 Matchmaker provides platform data automatization test system and test method
CN109474542B (en) * 2018-10-24 2022-05-13 平安科技(深圳)有限公司 Message push request flow control method, device and medium based on business rules
CN110750212A (en) * 2019-09-06 2020-02-04 中国平安财产保险股份有限公司 Article publishing method and device, computer equipment and storage medium
CN111277860B (en) * 2020-01-23 2021-04-06 北京邮电大学 Method, device and equipment for caching video in mobile edge network and readable medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102244811A (en) * 2011-07-20 2011-11-16 江苏省广播电视信息网络股份有限公司南京分公司 System and method for recommending television channels
CN102724179A (en) * 2012-05-25 2012-10-10 中国国际广播电台 Information dissemination system based on integration of broadcasting network and Internet
CN102802047A (en) * 2012-07-09 2012-11-28 北京迈伦斯科技有限公司 Information subscribing and information broadcasting method and system with cross-network collaboration

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101651283B1 (en) * 2009-11-13 2016-08-26 삼성전자 주식회사 Contents supply system and contents supply method thereof, display device and control method thereof
US9282354B2 (en) * 2011-10-28 2016-03-08 Qualcomm Incorporated Method and apparatus to detect a demand for and to establish demand-based multimedia broadcast multicast service

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102244811A (en) * 2011-07-20 2011-11-16 江苏省广播电视信息网络股份有限公司南京分公司 System and method for recommending television channels
CN102724179A (en) * 2012-05-25 2012-10-10 中国国际广播电台 Information dissemination system based on integration of broadcasting network and Internet
CN102802047A (en) * 2012-07-09 2012-11-28 北京迈伦斯科技有限公司 Information subscribing and information broadcasting method and system with cross-network collaboration

Also Published As

Publication number Publication date
CN105791910A (en) 2016-07-20

Similar Documents

Publication Publication Date Title
CN105791910B (en) A kind of multimedia resource supplying system and method
CN107205178B (en) Live broadcast room recommendation method and device
US8875186B2 (en) Apparatus and method of providing a recommended broadcast program
CN109429103B (en) Method and device for recommending information, computer readable storage medium and terminal equipment
CN103024555B (en) Based on television advertising detecting treatment system and the method for cloud computing
US10223359B2 (en) Determining recommended media programming from sparse consumption data
CN104837034B (en) A kind of information processing method, client and server
US20130227612A1 (en) System and method for organizing a media program guide according to popularity
US20170031916A1 (en) Methods and systems for searching for content items
CN109729390A (en) A kind of IPTV program monitoring method, apparatus and system
CN107070866A (en) The transmission method and device of a kind of flow data
CN106303591B (en) A kind of video recommendation method and device
US20210390210A1 (en) Privacy-aware content recommendations
CN102801539B (en) A kind of information issuing method and equipment, system
CN108093304A (en) A kind of intelligent recommendation method and system based on user's custom
US10701443B2 (en) System and method for recommending a content service to a content consumer
CN112150186A (en) Advertisement putting method, client device, server and system
CN103596022A (en) Method, device and system for content matching
KR20120071173A (en) System for providing additional service of vod content using sns message and method for providing additional service using the same
CN105979225A (en) Monitoring method and device for group video room
Khosroshahi et al. IPTV channel switching delay reduction through predicting subscribers’ behaviors and preferences
KR102155488B1 (en) Method for recommending contents and apparatus therefor
CN114501053A (en) Live stream acquisition method and device
Adeliyi et al. A meta-analysis of channel switching approaches for reducing zapping delay in internet protocol television
KR102130075B1 (en) Image learning support system for resolution restoration using user's preference information

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
TR01 Transfer of patent right

Effective date of registration: 20220118

Address after: 100010 8th floor, longfu building, No. 95, longfushi street, Dongcheng District, Beijing

Patentee after: Beijing State Owned Financial Leasing Co.,Ltd.

Address before: Room 423-1, 4th floor, building 2, No. 15, Wanyuan street, economic and Technological Development Zone, Daxing District, Beijing 100176

Patentee before: BEIJING STARTIMES SOFTWARE TECHNOLOGY Co.,Ltd.

TR01 Transfer of patent right