CN107392666A - Advertisement data processing method, device and advertisement placement method and device - Google Patents
Advertisement data processing method, device and advertisement placement method and device Download PDFInfo
- Publication number
- CN107392666A CN107392666A CN201710607076.4A CN201710607076A CN107392666A CN 107392666 A CN107392666 A CN 107392666A CN 201710607076 A CN201710607076 A CN 201710607076A CN 107392666 A CN107392666 A CN 107392666A
- Authority
- CN
- China
- Prior art keywords
- data
- video
- advertisement
- video paragraph
- scene
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000003672 processing method Methods 0.000 title claims abstract description 26
- 238000013507 mapping Methods 0.000 claims abstract description 24
- 230000011218 segmentation Effects 0.000 claims abstract description 24
- 238000009826 distribution Methods 0.000 claims abstract description 18
- 241001269238 Data Species 0.000 claims abstract description 14
- 239000000284 extract Substances 0.000 claims description 11
- 238000005520 cutting process Methods 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 230000000149 penetrating effect Effects 0.000 claims 1
- 238000010586 diagram Methods 0.000 description 10
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000013145 classification model Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0276—Advertisement creation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0277—Online advertisement
Landscapes
- Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Development Economics (AREA)
- Finance (AREA)
- Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a kind of advertisement data processing method, device and advertisement placement method and device, the advertisement data processing method includes：Scene segmentation is carried out to video data, obtains at least one video paragraph data, and is the unique video paragraph mark of each video paragraph data distribution；For each video paragraph data, scene classification is carried out to video paragraph data and multiple ad datas, and add corresponding scene class label；For video paragraph data distribution scene type label identical ad data, and the similarity score between video paragraph data and each ad data distributed is calculated respectively；Each ad data to be distributed distributes unique advertisement and identifier, and generates the ad score list of video paragraph data；The mapping relations established between the video paragraph mark of video paragraph data and ad score list.This has just obtained the correlation degree between video paragraph and each advertisement, and guarantee is provided subsequently to launch the advertisement of the high degree of association.
Description
Technical field
The present invention relates to technical field of network video, more specifically to a kind of advertisement data processing method, device and
Advertisement placement method and device.
Background technology
With the continuous development of video display arts, viewing Internet video has become the important composition of people's daily entertainment
Part, therefore, Internet video also turn into the major fields that advertisement is launched by advertiser.
The broadcasting form of network video advertisement is many at present, occurs such as before Internet video broadcasting, in broadcasting pause, and
And network video advertisement is shown in a manner of pop-up window etc. outside full frame covering, part of screen covering or screen substantially.However, by
Substantially unrelated with the content of Internet video in the content of these network video advertisements, therefore, network video advertisement is substantially impossible
The interest of user is transferred, or even user experience can be reduced.
The content of the invention
In view of this, the present invention provide a kind of network video advertisement data processing method, device and advertisement placement method and
Device, to solve the interest that existing network video advertisement can not possibly transfer user substantially, or even user experience can be reduced.Skill
Art scheme is as follows：
A kind of advertisement data processing method, including：
Scene segmentation is carried out to video data, obtains at least one video paragraph data, and be each described video-frequency band
Fall the unique video paragraph mark of data distribution；
For video paragraph data each described, field is carried out respectively to the video paragraph data and multiple ad datas
Scape is classified, and adds corresponding scene type label；
For ad data described in the video paragraph data distribution scene type label identical, and regard described in calculating respectively
Similarity score between frequency paragraph data and each ad data distributed；
Each ad data to be distributed distributes unique advertisement and identifier, and generates the video paragraph data
Ad score list；Wherein, the ad score list include the advertisement and identifier of each ad data that has distributed with
And corresponding similarity score；
The mapping relations established between the video paragraph mark of the video paragraph data and the ad score list.
Preferably, it is described to video data progress scene segmentation, including：
Video lens cutting is carried out to video data, obtains at least one shot sequence, each described shot sequence bag
Containing at least one frame of video；
Selecting video key frame at least one frame of video included respectively from each shot sequence；
At least one key frame of video progress scene clustering obtained according to default clustering rule to selection, and according to
Scene clustering result is segmented.
Preferably, it is described that scene classification is carried out respectively to the video paragraph data and multiple ad datas, including：
The first audio coefficients feature of speech data in the video paragraph data is extracted, meanwhile, extract multiple advertisement numbers
According to the second audio system characteristic of middle speech data；
It is special to the first audio coefficients feature and multiple second audio coefficients respectively according to default scene type
Sign is classified.
Preferably, the phase calculated respectively between the video paragraph data and each ad data for being distributed
Like degree fraction, including：
The the first audio coefficients feature and the first text data of speech data in the video paragraph data are extracted, meanwhile,
Extract the second audio coefficients feature and the second text data of speech data in each ad data distributed；
For each the described ad data distributed, according to the first audio coefficients feature and corresponding second audio
Coefficient characteristics calculate audio similarity fraction, meanwhile, according to first text data and the corresponding second text data meter
Calculate text similarity fraction；
According to the audio similarity fraction and the text similarity fraction calculate the video paragraph data with it is described
Similarity score between ad data.
A kind of ad data processing unit, including：Scene segmentation module, scene classification module, matching primitives module, list
Module is established in generation module and mapping；
The scene segmentation module, for carrying out scene segmentation to video data, at least one video paragraph data is obtained,
And it is each unique video paragraph mark of video paragraph data distribution；
The scene classification module, for for video paragraph data each described, to the video paragraph data and
Multiple ad datas carry out scene classification respectively, and add corresponding scene type label；
The matching primitives module, for for advertisement described in the video paragraph data distribution scene type label identical
Data, and the similarity score for calculating the video paragraph data respectively between each ad data for being distributed；
The List Generating Module, for the unique advertisement and identifier of each ad data distribution to be distributed, and
Generate the ad score list of the video paragraph data；Wherein, the ad score list includes each institute distributed
State the advertisement and identifier of ad data and corresponding similarity score；
Module is established in the mapping, and the video paragraph for establishing the video paragraph data identifies and the ad score
Mapping relations between list.
Preferably, for the scene segmentation module to video data progress scene segmentation, it is specifically used for：
Video lens cutting is carried out to video data, obtains at least one shot sequence, each described shot sequence bag
Containing at least one frame of video；Selecting video closes at least one frame of video included respectively from each shot sequence
Key frame；At least one key frame of video obtained according to default clustering rule to selection carries out scene clustering, and according to field
Scape cluster result is segmented.
Preferably, for carrying out the scene of scene classification respectively to the video paragraph data and multiple ad datas
Sort module, it is specifically used for：
The first audio coefficients feature of speech data in the video paragraph data is extracted, meanwhile, extract multiple advertisement numbers
According to the second audio system characteristic of middle speech data；According to default scene type respectively to the first audio coefficients feature and
Multiple second audio coefficients features are classified.
Preferably, for calculating the phase between the video paragraph data and each ad data for being distributed respectively
Like the matching primitives module of degree fraction, it is specifically used for：
The the first audio coefficients feature and the first text data of speech data in the video paragraph data are extracted, meanwhile,
Extract the second audio coefficients feature and the second text data of speech data in each ad data distributed；For institute
Each described ad data of distribution, according to the first audio coefficients feature and corresponding second audio coefficients feature calculation sound
Frequency similarity score, meanwhile, text similarity point is calculated according to first text data and corresponding second text data
Number；The video paragraph data and the advertisement number are calculated according to the audio similarity fraction and the text similarity fraction
Similarity score between.
A kind of advertisement placement method, including：
When receiving the advertisement playing request for carrying current video paragraph mark, according to the video paragraph pre-established
Mapping relations between mark and ad score list, search Current ad fraction row corresponding to the current video paragraph mark
Table, wherein, the ad score list includes the advertisements for each ad data that the video paragraph data is matched
Knowledge and corresponding similarity score, the mapping relations between the video paragraph mark and ad score list are according to above-mentioned
What the advertisement data processing method described in a technical scheme of anticipating was established；
Maximum similarity fraction is chosen from the Current ad fraction list, and the maximum similarity fraction is corresponding
Advertisement and identifier be advertisement and identifier to be played；
Obtain ad data to be played corresponding to the advertisement and identifier to be played and feed back.
A kind of advertisement delivery device, including：List lookup module, choose determining module and acquisition feedback module；
The list lookup module, for when receive carry current video paragraph mark advertisement playing request when,
According to the mapping relations between the video paragraph mark pre-established and ad score list, the current video paragraph mark is searched
Current ad fraction list corresponding to knowledge, wherein, the ad score list include the video paragraph data matched it is each
The advertisement and identifier of the individual ad data and corresponding similarity score, between the video paragraph mark and ad score list
Mapping relations be to be established according to the advertisement data processing method described in above-mentioned any one technical scheme；
The selection determining module, for choosing maximum similarity fraction from the Current ad fraction list, and will
Advertisement and identifier corresponding to the maximum similarity fraction is advertisement and identifier to be played；
The acquisition feedback module, for obtaining ad data to be played corresponding to the advertisement and identifier to be played and anti-
Feedback.
Compared to prior art, what the present invention realized has the beneficial effect that：
Advertisement data processing method, device and advertisement placement method and device provided by the invention above, the ad data
Processing method is by for video paragraph Data Matching scene type label identical ad data, tentatively ensure that advertisement and video
Relevance between paragraph, further, ad score list corresponding to video paragraph data is generated, this has just obtained video paragraph
With the correlation degree between each advertisement, guarantee is provided subsequently to launch the advertisement of the high degree of association.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
The embodiment of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing of offer obtains other accompanying drawings.
Fig. 1 is the method flow diagram of advertisement data processing method provided in an embodiment of the present invention；
Fig. 2 is advertisement data processing method step S101 provided in an embodiment of the present invention method flow diagram；
Fig. 3 is advertisement data processing method step S102 provided in an embodiment of the present invention method flow diagram；
Fig. 4 is advertisement data processing method step S103 provided in an embodiment of the present invention method flow diagram；
Fig. 5 is the structural representation of ad data processing unit provided in an embodiment of the present invention；
Fig. 6 is the method flow diagram of advertisement placement method provided in an embodiment of the present invention；
Fig. 7 is the structural representation of advertisement delivery device provided in an embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other under the premise of creative work is not made
Embodiment, belong to the scope of protection of the invention.
The embodiment of the present invention discloses a kind of advertisement data processing method, and method flow diagram is as shown in figure 1, including following step
Suddenly：
S101, scene segmentation is carried out to video data, obtains at least one video paragraph data, and be each video-frequency band
Fall the unique video paragraph mark of data distribution；
During step S101 is performed, video data can be the data in the video database pre-set, video
The type of data include but is not limited to PGC (Professional Generated Content, professional production content) or
UGC (User Generated Content, user's original content), for PGC, this is a kind of compared to UGC for this kind of video data
Video data, because duration is longer, scene is also more, therefore, after carrying out scene segmentation, this kind of video of PGC can obtain multiple
Video paragraph data, and the available video paragraph data of this kind of video data of UGC；
Following step can specifically be used by " carrying out scene segmentation to video data " during specific implementation, in step S101
Suddenly, method flow diagram is as shown in Figure 2：
S1001, video lens cutting is carried out to video data, obtains at least one shot sequence, each shot sequence
Include at least one frame of video；
During step S1001 is performed, video data is considered as the sequence of a continuous static image construction, often
One still image is referred to as a frame；And what a shot sequence in video data was made up of the frame of video of Time Continuous, it is right
Answer the start stop operation that video camera once records；The process that video lens cutting is carried out to video data exactly presses video data
At least one shot sequence is divided into according to the start-stop time.
S1002, selecting video key frame at least one frame of video included respectively from each shot sequence；
, can be according to the video content represented by each shot sequence, from what is included during step S1002 is performed
Representative key frame of video is chosen at least one frame of video.
S1003, scene clustering is carried out at least one key frame of video that selection obtains according to default clustering rule, and pressed
It is segmented according to scene clustering result；
During step S1003 is performed, default clustering rule be pre-set include definition, captions,
The rule of the cluster features such as similarity, to obtain a video scene, during being segmented according to scene clustering result,
For example, the key frame of video of shot sequence 1,2 and 3 can be combined cluster, the key frame of video of shot sequence 4 and 5 can be carried out
Combination cluster, therefore, can be combined into a video paragraph data, shot sequence by the total data in shot sequence 1,2 and 3
Total data in 4 and 5 is combined into a video paragraph data.
S102, for each video paragraph data, scene is carried out respectively to video paragraph data and multiple ad datas
Classification, and add corresponding scene type label；
During specific implementation, " scene is carried out respectively to video paragraph data and multiple ad datas in step S102
Classification " can specifically use following steps, and method flow diagram is as shown in Figure 3：
S1004, the first audio coefficients feature of speech data in video paragraph data is extracted, meanwhile, extract multiple advertisements
Second audio system characteristic of speech data in data；
S1005, according to default scene type respectively to the first audio coefficients feature and multiple second audio coefficients features
Classified；
During step S1005 is performed, including but not limited to Jing Yin scene type, pure voice, non-pure language are preset
Sound, music and ring sound, can be according to specific setting be actually needed, and the present embodiment does not do any restriction；
The first audio coefficients feature and multiple second audio coefficients features are divided respectively according to default scene type
During class, at least one video data and ad data can be chosen first as sample；Included in extraction sample each
The audio coefficients feature of data, and scene type label is added according to default scene type；According to default algorithm of support vector machine,
Each data and its corresponding scene type label generation scene classification model in sample；Using scene classification model to extraction
Audio coefficients feature is classified；
Wherein, audio coefficients feature can be audio MFCC coefficients (Mel-Frequency Cepstral
Coefficients, mel cepstrum coefficients), audio MFCC coefficients are the cepstrum parameters extracted in Mel scale frequencies domain,
Mel scales describe the nonlinear characteristic of human ear frequency.
S103, it is video paragraph data distribution scene type label identical ad data, and calculates video paragraph respectively
Similarity score between data and each ad data distributed；
During step S103 is performed, video paragraph data and ad data can be calculated according to default similarity algorithm
Between similarity score；
During specific implementation, " video paragraph data and each advertisement number distributed are calculated respectively in step S103
Similarity score between " can specifically use following steps, and method flow diagram is as shown in Figure 4：
S1006, the first audio coefficients feature and the first text data of speech data in video paragraph data are extracted, together
When, extract the second audio coefficients feature and the second text data of speech data in each ad data for being distributed；
During step S1006 is performed, the first audio coefficients feature and the second audio coefficients feature can be audio
MFCC coefficients (Mel-Frequency Cepstral Coefficients, mel cepstrum coefficients), and ASR can be passed through
(Automatic Speech Recognition, automatic speech recognition technology) extracts video paragraph data and ad data respectively
In speech data, and be converted into corresponding text data.
S1007, for each ad data distributed, according to the first audio coefficients feature and corresponding second audio system
Number feature calculation audio similarity fraction, meanwhile, it is similar to corresponding second text data calculating text according to the first text data
Spend fraction；
During step S1007 is performed, video paragraph data and advertisement number can be calculated according to cosine similarity algorithm
, can be according to TF-IDF (term frequency-inverse document according to the audio similarity fraction in terms of audio
Frequency, term frequency-inverse document frequency) video paragraph data and the ad data text similarity in terms of text point
Number；
Specifically, TF-IDF is a kind of statistical method, to assess a words for a file set or a language material
The significance level of a copy of it file in storehouse.The directly proportional increase of number that the importance of words occurs hereof with it,
But the frequency that can occur with it in corpus is inversely proportional decline simultaneously.The various forms of TF-IDF weightings is often searched engine
Using as the measurement of degree of correlation or grading between file and user's inquiry.
S1008, according to audio similarity fraction and text similarity fraction calculate video paragraph data and ad data it
Between similarity score；
During step S1008 is performed, similarity score can be calculated according to equation below (1)：
V=A*a+B*b (1)
Wherein, V is similarity score, and A is audio similarity fraction, and a is weighted value, B corresponding to audio similarity fraction
For text similarity fraction, b is weighted value corresponding to text similarity fraction, and a+b=1.
S104, each ad data to be distributed distributes unique advertisement and identifier, and generates the wide of video paragraph data
Accuse fraction list；Wherein, the ad score list includes the advertisement and identifier of each ad data that has distributed and corresponding
Similarity score；
S105, the mapping relations established between the video paragraph mark of video paragraph data and ad score list.
It should be noted that the present embodiment merely provides a kind of embodiment for calculating similarity score, specifically, can
To be actually needed the sequencing that selection calculates audio similarity fraction and text similarity fraction, also, to reduce amount of calculation,
For example, audio similarity fraction carries out threshold value comparison, it is similar that audio can also be chosen to the similarity score preferentially calculated
Spend the calculating that fraction carries out follow-up text similarity fraction and similarity score higher than the ad data of threshold value.
Above step S1001~step S1003 is only " to video data in the step S101 that the embodiment of the present application discloses
A kind of preferable implementation of progress scene segmentation " process, the specific implementation about this process can be according to the need of oneself
Any setting is asked, is not limited herein.
Above step S1004~step S1005 is only " to video paragraph in the step S102 that the embodiment of the present application discloses
Data and multiple ad datas carry out scene classification respectively " a kind of preferable implementation of process, about the specific of this process
Implementation can arbitrarily be set according to the demand of oneself, not limited herein.
Above step S1006~step S1008 is only " to calculate regard respectively in the step S103 that the embodiment of the present application discloses
A kind of preferable implementation of similarity score between frequency paragraph data and each ad data distributed " process, has
Closing the specific implementation of this process can arbitrarily be set according to the demand of oneself, not limit herein.
Advertisement data processing method provided in an embodiment of the present invention, by for video paragraph Data Matching scene type label
Identical ad data, the relevance between advertisement and video paragraph is tentatively ensure that, further, generate video paragraph data pair
The ad score list answered, this has just obtained the correlation degree between video paragraph and each advertisement, and high association is launched to be follow-up
The advertisement of degree provides guarantee.
The advertisement data processing method provided based on above-described embodiment, it is above-mentioned wide that the embodiment of the present invention then corresponds to offer execution
Accuse data processing method device, its structural representation as shown in figure 5, including：Scene segmentation module 101, scene classification module
102nd, module 105 is established in matching primitives module 103, List Generating Module 104 and mapping；
Scene segmentation module 101, for carrying out scene segmentation to video data, at least one video paragraph data is obtained,
And it is the unique video paragraph mark of each video paragraph data distribution；
Scene classification module 102, for for each video paragraph data, to video paragraph data and multiple advertisement numbers
According to carrying out scene classification respectively, and add corresponding scene type label；
Matching primitives module 103, for for video paragraph data distribution scene type label identical ad data, and point
Similarity score that Ji Suan be between video paragraph data and each ad data for being distributed；
List Generating Module 104, for the unique advertisement and identifier of each ad data distribution to be distributed, and generate and regard
The ad score list of frequency paragraph data；Wherein, the ad score list includes the wide of each ad data distributed
Accuse mark and corresponding similarity score；
Module 105 is established in mapping, and the video paragraph for establishing video paragraph data is identified between ad score list
Mapping relations.
Optionally, for the scene segmentation module 101 to video data progress scene segmentation, it is specifically used for：
Video lens cutting is carried out to video data, obtains at least one shot sequence, each shot sequence includes extremely
A few frame of video；Selecting video key frame at least one frame of video included respectively from each shot sequence；According to pre-
If clustering rule carries out scene clustering at least one key frame of video that selection obtains, and is divided according to scene clustering result
Section.
Optionally, for carrying out the scene classification of scene classification respectively to the video paragraph data and multiple ad datas
Module 102, is specifically used for：
The first audio coefficients feature of speech data in the video paragraph data is extracted, meanwhile, extract multiple advertisement numbers
According to the second audio system characteristic of middle speech data；According to default scene type respectively to the first audio coefficients feature and
Multiple second audio coefficients features are classified.
Optionally, for the similarity score between each ad data for calculating video paragraph data respectively and being distributed
Matching primitives module 103, be specifically used for：
The the first audio coefficients feature and the first text data of speech data in the video paragraph data are extracted, meanwhile,
Extract the second audio coefficients feature and the second text data of speech data in each ad data distributed；For institute
Each described ad data of distribution, according to the first audio coefficients feature and corresponding second audio coefficients feature calculation sound
Frequency similarity score, meanwhile, text similarity point is calculated according to first text data and corresponding second text data
Number；The video paragraph data and the advertisement number are calculated according to the audio similarity fraction and the text similarity fraction
Similarity score between.
Ad data processing unit provided in an embodiment of the present invention, by for video paragraph Data Matching scene type label
Identical ad data, the relevance between advertisement and video paragraph is tentatively ensure that, further, generate video paragraph data pair
The ad score list answered, this has just obtained the correlation degree between video paragraph and each advertisement, and high association is launched to be follow-up
The advertisement of degree provides guarantee.
The advertisement data processing method and ad data processing unit provided based on above-described embodiment, the embodiment of the present invention is also
A kind of advertisement placement method is provided, method flow diagram is as shown in fig. 6, comprise the following steps：
S201, when receiving the advertisement playing request for carrying current video paragraph mark, regarded according to what is pre-established
Frequency range falls the mapping relations between mark and ad score list, searches Current ad fraction corresponding to current video paragraph mark
List, wherein, the ad score list includes the advertisement and identifier for each ad data that the video paragraph data is matched
And corresponding similarity score, the mapping relations between the video paragraph mark and ad score list are according to above-mentioned any
What the advertisement data processing method described in one technical scheme was established；
S202, maximum similarity fraction is chosen from Current ad fraction list, and by corresponding to maximum similarity fraction
Advertisement and identifier is advertisement and identifier to be played；
S203, obtain ad data to be played corresponding to advertisement and identifier to be played and feed back.
Advertisement placement method provided in an embodiment of the present invention, identified according to current video paragraph to player and feed back the degree of correlation
Fraction highest ad data, the high similarity for launching advertisement and video is realized, to lift the smooth degree of viewing, improve and use
Experience at family.
The advertisement placement method provided based on above-described embodiment, the embodiment of the present invention, which then corresponds to provide, performs above-mentioned advertisement throwing
Put the device of method, its structural representation as shown in fig. 7, comprises：List lookup module 201, choose determining module 202 and obtain
Feedback module 203；
List lookup module 201, for when receive carry current video paragraph mark advertisement playing request when, root
According to the mapping relations between the video paragraph mark pre-established and ad score list, it is corresponding to search current video paragraph mark
Current ad fraction list, wherein, the ad score list includes each advertisement that the video paragraph data is matched
The advertisement and identifier of data and corresponding similarity score, the mapping relations between the video paragraph mark and ad score list
It is to be established according to the advertisement data processing method described in above-mentioned any one technical scheme；
Determining module 202 is chosen, for choosing maximum similarity fraction from Current ad fraction list, and by maximum phase
The advertisement and identifier like corresponding to degree fraction is advertisement and identifier to be played；
Acquisition feedback module 203, for obtaining ad data to be played corresponding to advertisement and identifier to be played and feeding back.
Advertisement delivery device provided in an embodiment of the present invention, identified according to current video paragraph to player and feed back the degree of correlation
Fraction highest ad data, the high similarity for launching advertisement and video is realized, to lift the smooth degree of viewing, improve and use
Experience at family.
A kind of advertisement data processing method, device and advertisement placement method provided by the present invention and device are carried out above
It is discussed in detail, specific case used herein is set forth to the principle and embodiment of the present invention, above example
Explanation be only intended to help understand the present invention method and its core concept；Meanwhile for those of ordinary skill in the art,
According to the thought of the present invention, there will be changes in specific embodiments and applications, in summary, in this specification
Appearance should not be construed as limiting the invention.
It should be noted that each embodiment in this specification is described by the way of progressive, each embodiment weight
Point explanation is all difference with other embodiment, between each embodiment identical similar part mutually referring to.
For device disclosed in embodiment, because it is corresponded to the method disclosed in Example, so fairly simple, the phase of description
Part is closed referring to method part illustration.
It should also be noted that, herein, such as first and second or the like relational terms are used merely to one
Entity or operation make a distinction with another entity or operation, and not necessarily require or imply between these entities or operation
Any this actual relation or order be present.Moreover, term " comprising ", "comprising" or its any other variant are intended to contain
Lid nonexcludability includes, so that the key element that process, method, article or equipment including a series of elements are intrinsic,
Either also include for these processes, method, article or the intrinsic key element of equipment.In the absence of more restrictions,
The key element limited by sentence "including a ...", it is not excluded that in the process including the key element, method, article or equipment
In other identical element also be present.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (10)
- A kind of 1. advertisement data processing method, it is characterised in that including：Scene segmentation is carried out to video data, obtains at least one video paragraph data, and be each described video paragraph number Identified according to unique video paragraph is distributed；For video paragraph data each described, scene point is carried out respectively to the video paragraph data and multiple ad datas Class, and add corresponding scene type label；For ad data described in the video paragraph data distribution scene type label identical, and the video-frequency band is calculated respectively The similarity score for falling data between each ad data for being distributed；Each ad data to be distributed distributes unique advertisement and identifier, and generates the advertisement of the video paragraph data Fraction list；Wherein, the ad score list includes the advertisement and identifier and phase of each ad data distributed The similarity score answered；The mapping relations established between the video paragraph mark of the video paragraph data and the ad score list.
- 2. according to the method for claim 1, it is characterised in that it is described to video data progress scene segmentation, including：Video lens cutting is carried out to video data, obtains at least one shot sequence, each described shot sequence includes extremely A few frame of video；Selecting video key frame at least one frame of video included respectively from each shot sequence；At least one key frame of video obtained according to default clustering rule to selection carries out scene clustering, and according to scene Cluster result is segmented.
- 3. according to the method for claim 1, it is characterised in that described to the video paragraph data and multiple ad datas Scene classification is carried out respectively, including：The first audio coefficients feature of speech data in the video paragraph data is extracted, meanwhile, extract in multiple ad datas Second audio system characteristic of speech data；The first audio coefficients feature and multiple second audio coefficients features are entered respectively according to default scene type Row classification.
- 4. according to the method for claim 1, it is characterised in that described to calculate the video paragraph data respectively with being distributed Each ad data between similarity score, including：The the first audio coefficients feature and the first text data of speech data in the video paragraph data are extracted, meanwhile, extraction The the second audio coefficients feature and the second text data of speech data in each ad data distributed；For each the described ad data distributed, according to the first audio coefficients feature and corresponding second audio coefficients Feature calculation audio similarity fraction, meanwhile, text is calculated according to first text data and corresponding second text data This similarity score；The video paragraph data and the advertisement are calculated according to the audio similarity fraction and the text similarity fraction Similarity score between data.
- A kind of 5. ad data processing unit, it is characterised in that including：Scene segmentation module, scene classification module, matching primitives Module is established in module, List Generating Module and mapping；The scene segmentation module, for carrying out scene segmentation to video data, at least one video paragraph data is obtained, and be Each described unique video paragraph mark of video paragraph data distribution；The scene classification module, for for video paragraph data each described, to the video paragraph data and multiple Ad data carries out scene classification respectively, and adds corresponding scene type label；The matching primitives module, for for advertisement number described in the video paragraph data distribution scene type label identical According to, and the similarity score for calculating the video paragraph data respectively between each ad data for being distributed；The List Generating Module, for the unique advertisement and identifier of each ad data distribution to be distributed, and generate The ad score list of the video paragraph data；Wherein, the ad score list is each described wide comprising what is distributed Accuse the advertisement and identifier of data and corresponding similarity score；Module is established in the mapping, and the video paragraph for establishing the video paragraph data identifies and the ad score list Between mapping relations.
- 6. device according to claim 5, it is characterised in that for carrying out the scene of scene segmentation to video data Segmentation module, it is specifically used for：Video lens cutting is carried out to video data, obtains at least one shot sequence, each described shot sequence includes extremely A few frame of video；Selecting video is crucial at least one frame of video included respectively from each shot sequence Frame；At least one key frame of video obtained according to default clustering rule to selection carries out scene clustering, and according to scene Cluster result is segmented.
- 7. device according to claim 5, it is characterised in that for the video paragraph data and multiple ad datas The scene classification module of scene classification is carried out respectively, is specifically used for：The first audio coefficients feature of speech data in the video paragraph data is extracted, meanwhile, extract in multiple ad datas Second audio system characteristic of speech data；According to default scene type respectively to the first audio coefficients feature and multiple The second audio coefficients feature is classified.
- 8. device according to claim 5, it is characterised in that for calculating the video paragraph data respectively with being distributed Each ad data between similarity score the matching primitives module, be specifically used for：The the first audio coefficients feature and the first text data of speech data in the video paragraph data are extracted, meanwhile, extraction The the second audio coefficients feature and the second text data of speech data in each ad data distributed；For being distributed Each described ad data, according to the first audio coefficients feature and corresponding second audio coefficients feature calculation audio phase Like degree fraction, meanwhile, text similarity fraction is calculated according to first text data and corresponding second text data；According to According to the audio similarity fraction and the text similarity fraction calculate the video paragraph data and the ad data it Between similarity score.
- A kind of 9. advertisement placement method, it is characterised in that including：When receiving the advertisement playing request for carrying current video paragraph mark, identified according to the video paragraph pre-established With the mapping relations between ad score list, Current ad fraction list corresponding to the current video paragraph mark is searched, Wherein, the ad score list include the advertisement and identifier of each ad data that the video paragraph data is matched with And corresponding similarity score, the mapping relations between the video paragraph mark and ad score list are according to claim 1 What the advertisement data processing method described in~4 any one was established；Maximum similarity fraction is chosen from the Current ad fraction list, and will be wide corresponding to the maximum similarity fraction Announcement is identified as advertisement and identifier to be played；Obtain ad data to be played corresponding to the advertisement and identifier to be played and feed back.
- A kind of 10. advertisement delivery device, it is characterised in that including：List lookup module, choose determining module and obtain feedback mould Block；The list lookup module, for when receive carry current video paragraph mark advertisement playing request when, according to Mapping relations between the video paragraph mark pre-established and ad score list, search the current video paragraph mark pair The Current ad fraction list answered, wherein, the ad score list includes each institute that the video paragraph data is matched State the advertisement and identifier of ad data and corresponding similarity score, reflecting between the video paragraph mark and ad score list The relation of penetrating is established according to the advertisement data processing method described in Claims 1 to 4 any one；The selection determining module, for choosing maximum similarity fraction from the Current ad fraction list, and by described in Advertisement and identifier corresponding to maximum similarity fraction is advertisement and identifier to be played；The acquisition feedback module, for obtaining ad data to be played corresponding to the advertisement and identifier to be played and feeding back.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710607076.4A CN107392666A (en) | 2017-07-24 | 2017-07-24 | Advertisement data processing method, device and advertisement placement method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710607076.4A CN107392666A (en) | 2017-07-24 | 2017-07-24 | Advertisement data processing method, device and advertisement placement method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107392666A true CN107392666A (en) | 2017-11-24 |
Family
ID=60337429
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710607076.4A Pending CN107392666A (en) | 2017-07-24 | 2017-07-24 | Advertisement data processing method, device and advertisement placement method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107392666A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107948729A (en) * | 2017-12-13 | 2018-04-20 | 广东欧珀移动通信有限公司 | Rich Media's processing method, device, storage medium and electronic equipment |
| CN107977244A (en) * | 2017-12-18 | 2018-05-01 | 北京奇虎科技有限公司 | The treating method and apparatus of picture is opened in a kind of application |
| CN108174309A (en) * | 2018-01-16 | 2018-06-15 | 深圳市瑞致达科技有限公司 | Barrage advertisement broadcast method, barrage advertisement play back device and readable storage medium storing program for executing |
| CN109068188A (en) * | 2018-09-25 | 2018-12-21 | 深圳市瑞致达科技有限公司 | The method and device, terminal and medium of local advertising are shown in multimedia |
| CN110276633A (en) * | 2019-05-09 | 2019-09-24 | 威比网络科技(上海)有限公司 | Advertisement placement method, system, equipment and storage medium based on online education |
| CN110399489A (en) * | 2019-07-08 | 2019-11-01 | 厦门市美亚柏科信息股份有限公司 | A kind of chat data segmentation method, device and storage medium |
| CN110475121A (en) * | 2018-05-10 | 2019-11-19 | 腾讯科技（深圳）有限公司 | A kind of video data handling procedure, device and relevant device |
| CN110611834A (en) * | 2019-08-28 | 2019-12-24 | 杭州当虹科技股份有限公司 | Streaming media advertisement interactive association accurate delivery method |
| CN110769291A (en) * | 2019-11-18 | 2020-02-07 | 上海极链网络科技有限公司 | Video processing method and device, electronic equipment and storage medium |
| WO2020168606A1 (en) * | 2019-02-22 | 2020-08-27 | 深圳前海微众银行股份有限公司 | Advertisement video optimising method, apparatus and device and computer readable storage medium |
| CN111629273A (en) * | 2020-04-14 | 2020-09-04 | 北京奇艺世纪科技有限公司 | Video management method, device, system and storage medium |
| CN113283821A (en) * | 2021-07-22 | 2021-08-20 | 腾讯科技（深圳）有限公司 | Virtual scene processing method and device, electronic equipment and computer storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1883198A (en) * | 2003-11-17 | 2006-12-20 | 皇家飞利浦电子股份有限公司 | Commercial insertion into video streams based on surrounding program content |
| US20080208849A1 (en) * | 2005-12-23 | 2008-08-28 | Conwell William Y | Methods for Identifying Audio or Video Content |
| CN101573726A (en) * | 2006-12-28 | 2009-11-04 | 夏普株式会社 | Advertisement distribution system, advertisement distribution server, advertisement distribution method, program, and recording medium |
| CN104902345A (en) * | 2015-05-26 | 2015-09-09 | 多维新创（北京）技术有限公司 | Method and system for realizing interactive advertising and marketing of products |
| CN106484736A (en) * | 2015-09-01 | 2017-03-08 | 上海亿动信息技术有限公司 | A kind of utilization mobile terminal identifies the information issuing method of audio/video file, apparatus and system |
| CN106686404A (en) * | 2016-12-16 | 2017-05-17 | 中兴通讯股份有限公司 | Video analysis platform, matching method, accurate advertisement delivery method and system |
-
2017
- 2017-07-24 CN CN201710607076.4A patent/CN107392666A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1883198A (en) * | 2003-11-17 | 2006-12-20 | 皇家飞利浦电子股份有限公司 | Commercial insertion into video streams based on surrounding program content |
| US20080208849A1 (en) * | 2005-12-23 | 2008-08-28 | Conwell William Y | Methods for Identifying Audio or Video Content |
| CN101573726A (en) * | 2006-12-28 | 2009-11-04 | 夏普株式会社 | Advertisement distribution system, advertisement distribution server, advertisement distribution method, program, and recording medium |
| CN104902345A (en) * | 2015-05-26 | 2015-09-09 | 多维新创（北京）技术有限公司 | Method and system for realizing interactive advertising and marketing of products |
| CN106484736A (en) * | 2015-09-01 | 2017-03-08 | 上海亿动信息技术有限公司 | A kind of utilization mobile terminal identifies the information issuing method of audio/video file, apparatus and system |
| CN106686404A (en) * | 2016-12-16 | 2017-05-17 | 中兴通讯股份有限公司 | Video analysis platform, matching method, accurate advertisement delivery method and system |
Non-Patent Citations (3)
| Title |
|---|
| 付畅俭: "《基于内容的视频结构挖掘》", 31 August 2016, 湘潭大学出版社 * |
| 朱明: "《数据挖掘（第2版）》", 30 November 2008, 中国科学技术大学出版社 * |
| 陆枫 等: ""基于细粒度标签的在线视频广告投放机制研究"", 《计算机研究与发展》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107948729A (en) * | 2017-12-13 | 2018-04-20 | 广东欧珀移动通信有限公司 | Rich Media's processing method, device, storage medium and electronic equipment |
| CN107948729B (en) * | 2017-12-13 | 2020-03-27 | Oppo广东移动通信有限公司 | Rich media processing method and device, storage medium and electronic equipment |
| CN107977244A (en) * | 2017-12-18 | 2018-05-01 | 北京奇虎科技有限公司 | The treating method and apparatus of picture is opened in a kind of application |
| CN108174309A (en) * | 2018-01-16 | 2018-06-15 | 深圳市瑞致达科技有限公司 | Barrage advertisement broadcast method, barrage advertisement play back device and readable storage medium storing program for executing |
| CN110475121B (en) * | 2018-05-10 | 2022-02-11 | 腾讯科技（深圳）有限公司 | Video data processing method and device and related equipment |
| CN110475121A (en) * | 2018-05-10 | 2019-11-19 | 腾讯科技（深圳）有限公司 | A kind of video data handling procedure, device and relevant device |
| CN109068188A (en) * | 2018-09-25 | 2018-12-21 | 深圳市瑞致达科技有限公司 | The method and device, terminal and medium of local advertising are shown in multimedia |
| WO2020168606A1 (en) * | 2019-02-22 | 2020-08-27 | 深圳前海微众银行股份有限公司 | Advertisement video optimising method, apparatus and device and computer readable storage medium |
| CN110276633A (en) * | 2019-05-09 | 2019-09-24 | 威比网络科技(上海)有限公司 | Advertisement placement method, system, equipment and storage medium based on online education |
| CN110399489A (en) * | 2019-07-08 | 2019-11-01 | 厦门市美亚柏科信息股份有限公司 | A kind of chat data segmentation method, device and storage medium |
| CN110399489B (en) * | 2019-07-08 | 2022-06-17 | 厦门市美亚柏科信息股份有限公司 | Chat data segmentation method, device and storage medium |
| CN110611834A (en) * | 2019-08-28 | 2019-12-24 | 杭州当虹科技股份有限公司 | Streaming media advertisement interactive association accurate delivery method |
| CN110769291A (en) * | 2019-11-18 | 2020-02-07 | 上海极链网络科技有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN111629273B (en) * | 2020-04-14 | 2022-02-11 | 北京奇艺世纪科技有限公司 | Video management method, device, system and storage medium |
| CN111629273A (en) * | 2020-04-14 | 2020-09-04 | 北京奇艺世纪科技有限公司 | Video management method, device, system and storage medium |
| CN113283821B (en) * | 2021-07-22 | 2021-10-29 | 腾讯科技（深圳）有限公司 | Virtual scene processing method and device, electronic equipment and computer storage medium |
| CN113283821A (en) * | 2021-07-22 | 2021-08-20 | 腾讯科技（深圳）有限公司 | Virtual scene processing method and device, electronic equipment and computer storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107392666A (en) | Advertisement data processing method, device and advertisement placement method and device | |
| CN105095508B (en) | A kind of multimedia content recommended method and multimedia content recommendation apparatus | |
| WO2022095380A1 (en) | Ai-based virtual interaction model generation method and apparatus, computer device and storage medium | |
| CN102405495B (en) | Audio classification for information retrieval using sparse features | |
| CN105138670B (en) | Audio file label generating method and system | |
| US20140164507A1 (en) | Media content portions recommended | |
| US20080109391A1 (en) | Classifying content based on mood | |
| CN109165302A (en) | Multimedia file recommendation method and device | |
| CN110209869A (en) | A kind of audio file recommended method, device and storage medium | |
| CN106776528B (en) | Information processing method and device | |
| CN103678668A (en) | Prompting method of relevant search result, server and system | |
| MX2012000724A (en) | Estimating and displaying social interest in time-based media. | |
| CN103942328B (en) | A kind of video retrieval method and video-unit | |
| CN104268279B (en) | The querying method and device of corpus data | |
| US20160381433A1 (en) | Information distribution device, information distribution method, and program | |
| CN108334518A (en) | A kind of advertisement loading method and device | |
| KR20170107868A (en) | Method and system to recommend music contents by database composed of user's context, recommended music and use pattern | |
| CN110287788A (en) | A kind of video classification methods and device | |
| CN109460503A (en) | Answer input method, device, storage medium and electronic equipment | |
| CN110020200A (en) | A kind of personalized recommendation method and system based on history material | |
| JP5367872B2 (en) | How to provide users with selected content items | |
| CN111259245A (en) | Work pushing method and device and storage medium | |
| CN109977294A (en) | Information/query processing device, query processing/text query method, storage medium | |
| JP7207543B2 (en) | Information recommendation device, information recommendation system, information recommendation method, and information recommendation program | |
| US11410706B2 (en) | Content pushing method for display device, pushing device and display device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171124 |
|
| RJ01 | Rejection of invention patent application after publication |