CN107451148A - Video classification method and device and electronic equipment - Google Patents
Video classification method and device and electronic equipment Download PDFInfo
- Publication number
- CN107451148A CN107451148A CN201610375160.3A CN201610375160A CN107451148A CN 107451148 A CN107451148 A CN 107451148A CN 201610375160 A CN201610375160 A CN 201610375160A CN 107451148 A CN107451148 A CN 107451148A
- Authority
- CN
- China
- Prior art keywords
- video
- classification
- sort key
- key word
- phrase
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/70—Information retrieval; Database structures therefor; File system structures therefor of video data
- G06F16/78—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/7867—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a video classification method, a video classification device and electronic equipment, relates to an information classification technology, and can improve video classification efficiency. The video classification method comprises the following steps: acquiring a video to be classified; extracting key words in the video to be classified to obtain key word groups; matching the keyword sets with preset classification keyword sets of each video classification mapping in sequence to respectively obtain classification keyword sets matched with the classification keyword sets; acquiring browsing parameters of the video to be classified, and calculating a classification quality score of the video to be classified according to the browsing parameters and the classification key phrase; and if the classification quality score exceeds a preset classification quality score threshold, classifying the video to be classified into the classification video corresponding to the classification key phrase. The method is suitable for classifying the video category and quality.
Description
Technical field
The present invention relates to information classification technology, more particularly to a kind of video classification methods, device and electronic equipment.
Background technology
Today's society, information have penetrated into each industry and operation function field, turn into the important factor of production.People
Excavation and utilization for mass data, imply that new ripple increase in productivity and the arrival of consumer surplus's tide, wherein,
To the Accurate classification of magnanimity information, be advantageous to people and reduce the time browsed needed for magnanimity information, relevant industries can be obtained more
For accurate information, so as to the experience that is accurate and browsing information according to the lifting decision-making of accurate information.
Due to including numerous video websites on internet, and each video website includes numerous videos, for the ease of
User obtains the higher video of quality, it is necessary to which each video website carries out classification ranking to video therein from video website.Mesh
Preceding video classification methods, labeling is mainly carried out for video automatically by user and uploads to video website, video network
The operation personnel to stand is classified for video again according to video tab and combination video content, is wrapped for example, being uploaded for user
Containing video tab be comedy, amusement, humour video, operation personnel can be ranged to make laughs and regarded after analyzing again
Frequency or non-funny video.After sorting, the temperature in conjunction with the video and recommendation carry out ranking, and according to rank order in net
Recommended on standing, for example, by some classify under video recommendations in the top to video website homepage, it is more to obtain
The concern of people, be advantageous to the popularization of homepage video.
But existing video classification methods, because manual type of the video website based on operation personnel is divided video
Class is, it is necessary to expend substantial amounts of manpower to carry out video screening and classification so that visual classification is less efficient.Further, due to
When the operation personnel of each video website is to visual classification, classify according to subjective judgement so that video classification methods vary, no
Same video may be classified into different classification, cause each classification video to include institute according to different operation personnel
There is video so that visual classification is more chaotic, so as to influence the video tastes of user.
The content of the invention
In view of this, the embodiment of the present invention provides a kind of video classification methods, device and electronic equipment, can lift video
Classification effectiveness, solve the problems, such as that existing video classification methods need visual classification caused by manually being classified less efficient.
In a first aspect, the embodiment of the present invention provides a kind of video classification methods, including:
Obtain video to be sorted;
The keyword in the video to be sorted is extracted, obtains crucial phrase;
The sort key word set that the crucial phrase maps with each visual classification pre-set successively is matched, point
The sort key phrase to match with each sort key word set is not obtained;
The parameter that browses of the video to be sorted is obtained, parameter is browsed according to described in and the sort key phrase calculates
The classification quality fraction of the video to be sorted;
If the classification quality fraction exceedes the classification quality score threshold pre-set, the video to be sorted is returned
Class is in video of classifying corresponding to the sort key phrase.
With reference in a first aspect, in the first embodiment of first aspect, by the crucial phrase successively with setting in advance
The sort key word set for each visual classification mapping put is matched, and respectively obtains the classification to match with each sort key word set
Crucial phrase includes:
Concentrated in the mapping relations of visual classification and sort key word set, extract the first sort key word set;
The crucial phrase is matched with the first sort key word set, obtain in the crucial phrase with it is described
Each keyword that first sort key word set matches, obtain the first sort key phrase;
Judge whether the sort key word set of the mapping relations concentration is extracted to finish, if it is, terminate flow, if
It is no, next second sort key word set is extracted, the crucial phrase is matched with the second sort key word set, is obtained
Each keyword to match in the crucial phrase with the second sort key word set, the second sort key phrase is obtained, directly
The sort key word set concentrated to the mapping relations, which is extracted, to be finished.
With reference to the first embodiment of first aspect, in second of embodiment of first aspect, the acquisition regards
The mapping relations collection of frequency division class and sort key word set includes:
Obtain Sample video collection;
According to the visual classification pre-set, each Sample video concentrated to the Sample video is classified, obtained
Classification samples video group corresponding to each visual classification;
The keyword that each classification samples video bag contains in a classification samples video group is extracted, obtains sample crucial phrase;
Word frequency statisticses are carried out to the keyword in the sample crucial phrase;
The keyword of word frequency top N is write to the sort key word set of the classification samples video group mapping, builds video
Classification and the mapping relations of the sort key word set;
Judge whether classification samples video group is extracted to finish, if not, performing in one classification samples video group of the extraction
The step of keyword that each classification samples video bag contains, if it is, the mapping relations according to structure form mapping relations collection.
With reference to the first or second of embodiment of first aspect, first aspect, in the third implementation of first aspect
In mode, the classification quality fraction is calculated using following formula:
F=ξ1*lg(brows)+ξ2*lg10((likes-dislikes)*(likes/(dislikes+1)))+ξ3*lg
(date)+ξ4*key-words
In formula,
F is classification quality fraction;
ξ1For pageview weight coefficient;
Brows is pageview;
ξ2To like several weight coefficients;
Likes is to like several;
Dislikes is not like number;
ξ3For date weight coefficient;
Date is number of days of the issue date apart from current date;
ξ4For sort key phrase weight coefficient;
Key-words is the sort key word number that sort key phrase includes.
With reference to the third embodiment of first aspect, in the 4th kind of embodiment of first aspect, the pageview
Weight coefficient is 0.25, described to like several weight coefficients be 0.0.375, the date weight coefficient is 0.1, and, described point
Class keywords group weight coefficient is 0.025.
With reference to the first or second of embodiment of first aspect, first aspect, in the 5th kind of implementation of first aspect
In mode, methods described also includes:
Classify corresponding to the sort key phrase in classification in video, arranged according to the classification quality fraction
Sequence, video corresponding to the classification quality fraction of M positions before sorting is obtained, is recommended to the homepage of the video website pre-set.
With reference to the 5th kind of embodiment of first aspect, in the 6th kind of embodiment of first aspect, methods described is also
Including:
The user's collection for registering classification video corresponding to the sort key phrase is obtained, by the classification quality of M positions before sequence
Each user that video corresponding to fraction is concentrated to the user of acquisition pushes.
Second aspect, the embodiment of the present invention provide a kind of visual classification device, including:Video acquiring module, keyword carry
Modulus block, matching module, classification quality fraction computing module and visual classification module, wherein,
Video acquiring module, for obtaining video to be sorted;
Keyword extracting module, for extracting the keyword in the video to be sorted, obtain crucial phrase;
Matching module, for the sort key word for mapping the crucial phrase with each visual classification pre-set successively
Collection is matched, and respectively obtains the sort key phrase to match with each sort key word set;
Classification quality fraction computing module, for obtaining the parameter that browses of the video to be sorted, ginseng is browsed according to described in
Several and described sort key phrase calculates the classification quality fraction of the video to be sorted;
Visual classification module, if the classification quality fraction exceedes the classification quality score threshold pre-set, by institute
State video to be sorted and range classification video corresponding to the sort key phrase.
With reference to second aspect, in the first embodiment of second aspect, the matching module includes:Sort key word
Collect extraction unit, matching unit and Traversal Unit, wherein,
Sort key word set extraction unit, concentrated for the mapping relations in visual classification and sort key word set, extraction
First sort key word set;
Matching unit, for the crucial phrase to be matched with the first sort key word set, obtain the pass
Each keyword to match in keyword group with the first sort key word set, obtain the first sort key phrase;
Traversal Unit, whether the sort key word set for judging the mapping relations concentration, which is extracted, finishes, if it is, knot
Line journey, if not, the next second sort key word set of extraction, the crucial phrase and the second sort key word set are entered
Row matching, obtains each keyword to match in the crucial phrase with the second sort key word set, obtains the second classification
Crucial phrase, until the sort key word set that the mapping relations are concentrated is extracted and finished.
With reference to the first embodiment of second aspect, in second of embodiment of second aspect, the classification is closed
Keyword collection extraction unit includes:Sample video collection subelement, sample classification subelement, sample keyword subelement, word frequency statisticses
Subelement, mapping relations structure subelement and mapping relations collection subelement, wherein,
Sample video collection subelement, for obtaining Sample video collection;
Sample classification subelement, for according to the visual classification pre-set, that is concentrated to the Sample video to be every the same
This video is classified, and obtains classification samples video group corresponding to each visual classification;
Sample keyword subelement, for extracting the key that each classification samples video bag contains in a classification samples video group
Word, obtain sample crucial phrase;
Word frequency statisticses subelement, for carrying out word frequency statisticses to the keyword in the sample crucial phrase;
Mapping relations build subelement, are mapped for the keyword of word frequency top N to be write into the classification samples video group
Sort key word set, the mapping relations of structure visual classification and the sort key word set;
Mapping relations collection subelement, is used for, and judges whether classification samples video group is extracted and finishes, if not, notice sample
Keyword subelement, if it is, the mapping relations according to structure form mapping relations collection.
With reference to the first or second of embodiment of second aspect, second aspect, in the third implementation of second aspect
In mode, the classification quality fraction is calculated using following formula:
F=ξ1*lg(brows)+ξ2*lg10((likes-dislikes)*(likes/(dislikes+1)))+ξ3*lg
(date)+ξ4*key-words
In formula,
F is classification quality fraction;
ξ1For pageview weight coefficient;
Brows is pageview;
ξ2To like several weight coefficients;
Likes is to like several;
Dislikes is not like number;
ξ3For date weight coefficient;
Date is number of days of the issue date apart from current date;
ξ4For sort key phrase weight coefficient;
Key-words is the sort key word number that sort key phrase includes.
With reference to the third embodiment of second aspect, in the 4th kind of embodiment of second aspect, the pageview
Weight coefficient is 0.25, described to like several weight coefficients be 0.0.375, the date weight coefficient is 0.1, and, described point
Class keywords group weight coefficient is 0.025.
With reference to the first or second of embodiment of second aspect, second aspect, in the 5th kind of implementation of second aspect
In mode, described device also includes:
Sort recommendations module, for corresponding to the sort key phrase in classification classify video in, according to described point
Class mass fraction is ranked up, and is obtained video corresponding to the classification quality fraction of M positions before sorting, is recommended to the video pre-set
The homepage of website.
With reference to the first or second of embodiment of second aspect, second aspect, in the 6th kind of implementation of second aspect
In mode, described device also includes:
Pushing module, the user's collection for video of classifying corresponding to the sort key phrase is registered for obtaining, by M before sequence
Each user that video corresponding to the classification quality fraction of position is concentrated to the user of acquisition pushes.
The third aspect, the embodiment of the present invention provide a kind of electronic equipment, and the electronic equipment includes:Housing, processor, deposit
Reservoir, circuit board and power circuit, wherein, circuit board is placed in the interior volume that housing surrounds, and processor and memory are set
On circuit boards;Power circuit, for each circuit or the device power supply for above-mentioned electronic equipment;Memory is used to store and can hold
Line program code;The executable program code that processor is stored by reading in memory is run and executable program code pair
The program answered, for performing foregoing any described video classification methods.
A kind of video classification methods, device and electronic equipment provided in an embodiment of the present invention, by obtaining video to be sorted;
The keyword in the video to be sorted is extracted, obtains crucial phrase;The crucial phrase is respectively regarded with what is pre-set successively
The sort key word set of frequency classification map is matched, and respectively obtains the sort key word to match with each sort key word set
Group;The parameter that browses of the video to be sorted is obtained, is browsed according to described in described in parameter and sort key phrase calculating
The classification quality fraction of video to be sorted;If the classification quality fraction exceedes the classification quality score threshold pre-set,
The video to be sorted is ranged into video of classifying corresponding to the sort key phrase.Visual classification efficiency is lifted, is solved existing
The problem of visual classification caused by having video classification methods to need artificial classified is less efficient.
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
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the video classification methods schematic flow sheet of embodiments of the invention one;
Fig. 2 is the video classification methods schematic flow sheet of embodiments of the invention two;
Fig. 3 is the visual classification apparatus structure schematic diagram of embodiments of the invention three;
Fig. 4 is the structural representation of electronic equipment one embodiment of the present invention.
Embodiment
The embodiment of the present invention is described in detail below in conjunction with the accompanying drawings.
It will be appreciated that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.Base
Embodiment in the present invention, those of ordinary skill in the art obtained under the premise of creative work is not made it is all its
Its embodiment, belongs to the scope of protection of the invention.
Fig. 1 is the video classification methods schematic flow sheet of embodiments of the invention one, as shown in figure 1, the method for the present embodiment
It can include:
Step 101, video to be sorted is obtained.
In this step, as an alternative embodiment, video to be sorted can be grabbed from the video website pre-set
Video, including one or more videos.
As an alternative embodiment, the video website pre-set includes:Yoqoo, potato net, LeEco net, iqiyi.com
Net, cruel 6 net, Tengxun's video, Sohu's video, pptv, pps, YouTube, Baidu are audio-visual etc..
As another alternative embodiment, video to be sorted can be the video for carrying video tab that user uploads.
Step 102, the keyword in the video to be sorted is extracted, obtains crucial phrase.
In this step, as an alternative embodiment, by dividing the title in video to be sorted, brief introduction, comment etc.
Word, cutting and redundancy processing, can obtain each keyword included in the video to be sorted, form crucial phrase.
Step 103, the sort key word set that the crucial phrase maps with each visual classification pre-set successively is entered
Row matching, respectively obtains the sort key phrase to match with each sort key word set.
In this step, as an alternative embodiment, the crucial phrase is reflected with each visual classification pre-set successively
The sort key word set penetrated is matched, and respectively obtaining the sort key phrase to match with each sort key word set includes:
A11, concentrated in the mapping relations of visual classification and sort key word set, extract the first sort key word set;
A12, the crucial phrase is matched with the first sort key word set, obtain in the crucial phrase with
Each keyword that the first sort key word set matches, obtain the first sort key phrase;
A13, judge whether the sort key word set of the mapping relations concentration is extracted and finish, if it is, terminating flow, such as
Fruit is no, extracts next second sort key word set, the crucial phrase is matched with the second sort key word set, obtained
Each keyword to match in the crucial phrase with the second sort key word set is taken, obtains the second sort key phrase,
Until the sort key word set that the mapping relations are concentrated is extracted and finished.
In the present embodiment, mapping relations are concentrated, and include one or more visual classifications, and each visual classification maps one point
Class keywords collection, by the way that the crucial phrase that the keyword extracted from video to be sorted obtains and each sort key word set are entered
Row matching, can obtain the one or more keywords to match with corresponding visual classification, and it is corresponding to form the visual classification
Sort key phrase.
In the present embodiment, as an alternative embodiment, visual classification and the mapping relations Ji Bao of sort key word set are obtained
Include:
B11, obtain Sample video collection;
In the present embodiment, Sample video collection can be the crawl or from user from the video website pre-set
Collected in the video of upload, the present embodiment is not construed as limiting to this.
B12, according to the visual classification pre-set, each Sample video concentrated to the Sample video is classified,
Obtain classification samples video group corresponding to each visual classification;
In the present embodiment, video can be classified in advance, for example, can be by visual classification:Funny video, red-letter day
Video etc. being blessed, wherein, for funny video, can be divided into again:Sport category funny video, life kind funny video, amusement class
Funny video etc., the present embodiment is not construed as limiting to this.
In the present embodiment, the analysis of each Sample video is concentrated to Sample video by experienced operation personnel, by it
Range in corresponding visual classification, wherein, same Sample video, different classification videos can be attributed to, for example, for a certain
Video, if by analysis, can both range funny video, red-letter day blessing video can also be ranged.
B13, the keyword that each classification samples video bag contains in a classification samples video group is extracted, obtains sample keyword
Group;
In the present embodiment, by the title in each classification samples video in classification samples video group, brief introduction, comment etc.
Segmented, the processing of cutting and redundancy, each pass that each classification samples video bag contains in the classification samples video group can be obtained
Keyword, form sample crucial phrase.In sample crucial phrase, same keyword is remembered respectively corresponding to different classifications Sample video
Record, same keyword corresponding to same classification samples video also record respectively, for example, for a certain classification samples video, extraction
Keyword may be:UEFA Champions League, make laughs, the collection of choice specimens, make laughs, Rhoneldo, make laughs, the collection of choice specimens.
B14, word frequency statisticses are carried out to the keyword in the sample crucial phrase;
In the present embodiment, in sample crucial phrase, the word frequency of each keyword is counted.If for example, sample keyword
The keyword that group includes is:UEFA Champions League, make laughs, the collection of choice specimens, make laughs, Rhoneldo, make laughs, the collection of choice specimens, then the word frequency of UEFA Champions League is 1, is made laughs
Word frequency be 3, the word frequency of the collection of choice specimens is 2, and the word frequency of Rhoneldo is 1.
B15, the keyword of word frequency top N is write to the sort key word set of the classification samples video group mapping, structure
Visual classification and the mapping relations of the sort key word set;
In the present embodiment, the keyword of the higher top N of word frequency is won out, the classification as corresponding visual classification is closed
Keyword collection.For example, for funny video, count what title, brief introduction, comment etc. in each video being referred in funny video included
Keyword, the word frequency of each keyword is counted, the higher keyword of word frequency is won out, form classification Video Key word set,
And then build funny video with classify Video Key word set mapping relations, for example, by the classification Video Key word set of formation with
Make laughs as label, for example, the classification Video Key word set of formation can be named as funny video keyword set, so as to
Build mapping relations.For another example for red-letter day bless video, statistics be referred to red-letter day blessing video in each video in title,
The keyword that brief introduction, comment etc. include, the word frequency of each keyword is counted, the higher keyword of word frequency is won out, named
Video Key word set is blessed for red-letter day.
B16, judge whether classification samples video group is extracted and finish, if not, step B13 is performed, if it is, according to structure
Mapping relations formed mapping relations collection.
Step 104, the parameter that browses of the video to be sorted is obtained, parameter and the sort key are browsed according to described in
Phrase calculates the classification quality fraction of the video to be sorted;
In the present embodiment, browsing parameter includes:Pageview, like number, do not like number, issue date etc..Browsing parameter can
By analyzing video acquisition to be sorted.
In the present embodiment, it is contemplated that the pageview of video, like number, do not like number meet long-tail model, using pageview as
Example, i.e., pageview can explode in a short time, tend towards stability afterwards gentle.Thus, as an alternative embodiment, utilize following formula meter
Point counting class mass fraction:
F=ξ1*lg(brows)+ξ2*lg10((likes-dislikes)*(likes/(dislikes+1)))+ξ3*lg
(date)+ξ4*key-words
In formula,
F is classification quality fraction, if for example, sort key phrase is keyword corresponding to funny video, F is to make laughs
Video quality fraction, if sort key phrase is to bless keyword corresponding to video red-letter day, F is to bless video quality in red-letter day
Fraction.
ξ1For pageview weight coefficient;
Brows is pageview;
ξ2To like several weight coefficients;
Likes is to like several;
Dislikes is not like number;
ξ3For date weight coefficient;
Date is number of days of the issue date apart from current date;
ξ4For sort key phrase weight coefficient;
Key-words is the sort key word number that sort key phrase includes.
In the present embodiment, each coefficient can be set according to being actually needed.As an alternative embodiment, in the present embodiment, pass through
Statistic and analysis, it can set:
ξ1=0.125, ξ2=0.0375, ξ3=0.1, ξ4=0.025.
Certainly, in practical application, classification quality fraction can also be normalized.
Step 105, if the classification quality fraction exceedes the classification quality score threshold that pre-sets, treated described point
Class video ranges classification video corresponding to the sort key phrase.
In this step, classification quality score threshold can be set according to being actually needed, for different classification videos, Ke Yishe
Different classification quality score thresholds is put, identical classification quality score threshold can also be set.
In the present embodiment, if for example, by Keywords matching, the sort key phrase of video to be sorted includes making laughs regarding
Frequency crucial phrase and red-letter day blessing Video Key phrase, then be applied to above-mentioned formula by funny video crucial phrase, done
Laugh at video quality fraction, occurrence 2;It will bless that Video Key phrase is applied to above-mentioned formula red-letter day, and obtain red-letter day blessing and regard
Frequency mass fraction, occurrence 1.5.
If funny video quality score thresholds and red-letter day blessing video quality score threshold are 1.8, this is treated point
Class video is attributed to funny video;If funny video quality score thresholds are arranged to 1.8, red-letter day blessing video quality score threshold
1.2 are arranged to, then the video to be sorted is attributed to funny video respectively and red-letter day blesses video.
The video classification methods of the present embodiment one, by obtaining video to be sorted;Extract the pass in the video to be sorted
Keyword, obtain crucial phrase;The sort key word set that the crucial phrase is mapped with each visual classification pre-set successively
Matched, respectively obtain the sort key phrase to match with each sort key word set;Obtain the clear of the video to be sorted
Look at parameter, parameter is browsed according to described in and the sort key phrase calculates the classification quality fraction of the video to be sorted;
If the classification quality fraction exceedes the classification quality score threshold pre-set, the video to be sorted is ranged described
Classification video corresponding to sort key phrase.So, the video to be sorted using the sort key word set pre-set to extraction
In crucial phrase matched, and combine the video to be sorted and browse parameter, carrying out mass fraction to video to be sorted comments
Estimate, realize the automatic classification to video to be sorted according to the mass fraction assessed, the time needed for visual classification is short, visual classification
Efficiency high.Further, it is unified according to the classification quality fraction for browsing parameter and sort key phrase calculating video to be sorted
Video classification methods, so as to improving the video tastes of user.
Fig. 2 is the video classification methods schematic flow sheet of the embodiment of the present invention two, as shown in Fig. 2 the method for the present embodiment can
With including:
Step 201, video to be sorted is obtained;
Step 202, the keyword in the video to be sorted is extracted, obtains crucial phrase;
Step 203, the sort key word set that the crucial phrase maps with each visual classification pre-set successively is entered
Row matching, respectively obtains the sort key phrase to match with each sort key word set;
Step 204, the parameter that browses of the video to be sorted is obtained, parameter and the sort key are browsed according to described in
Phrase calculates the classification quality fraction of the video to be sorted;
Step 205, if the classification quality fraction exceedes the classification quality score threshold that pre-sets, treated described point
Class video ranges classification video corresponding to the sort key phrase;
In the present embodiment, the process of step 201 to step 205 respectively with the step 101 of above method embodiment one to step
Rapid 105 is similar, and here is omitted.
Step 206, in classification video corresponding to the sort key phrase in classification, according to the classification quality fraction
It is ranked up, obtains video corresponding to the classification quality fraction of M positions before sorting, recommend to the head of the video website pre-set
Page;
Step 207, the user's collection for registering classification video corresponding to the sort key phrase is obtained, by M positions before sequence
Each user that video corresponding to classification quality fraction is concentrated to the user of acquisition pushes.
In the embodiment of the present invention, using video classification methods, the classification video of high quality can be obtained, so as to continually
Provide a user make laughs, red-letter day blessing etc. particular category the short-sighted frequency of high quality.It is much to use due to watching short-sighted frequency of making laughs
The daily hobby at family, viewing red-letter day blessing video can increase festive air, so as to adjust the Working Life of people.
The video classification methods of inventive embodiments two, by being ranked up and recommending to video website to classification quality fraction
Homepage or push to user, can endlessly provide a user high-quality video, meet user's request.
Fig. 3 is the visual classification apparatus structure schematic diagram of embodiments of the invention three, as shown in figure 3, the device of the present embodiment
It can include:Video acquiring module 31, keyword extracting module 32, matching module 33, classification quality fraction computing module 34 with
And visual classification module 35, wherein,
Video acquiring module 31, for obtaining video to be sorted;
In the present embodiment, as an alternative embodiment, video to be sorted can be from the video website crawl pre-set
To one or more videos or user upload the one or more videos for carrying video tab.
Keyword extracting module 32, for extracting the keyword in the video to be sorted, obtain crucial phrase;
In the present embodiment, by being segmented to the title in video to be sorted, brief introduction, comment etc., cutting and redundancy
Processing, each keyword included in the video to be sorted can be obtained, form crucial phrase.
Matching module 33, for the sort key for mapping the crucial phrase with each visual classification pre-set successively
Word set is matched, and respectively obtains the sort key phrase to match with each sort key word set;
In the present embodiment, as an alternative embodiment, matching module 33 includes:Sort key word set extraction unit, matching
Unit and Traversal Unit (not shown), wherein,
Sort key word set extraction unit, concentrated for the mapping relations in visual classification and sort key word set, extraction
First sort key word set;
In the present embodiment, mapping relations are concentrated, and include one or more visual classifications, and each visual classification maps one point
Class keywords collection.
In the present embodiment, as an alternative embodiment, sort key word set extraction unit includes:Sample video collected works list
Member, sample classification subelement, sample keyword subelement, word frequency statisticses subelement, mapping relations structure subelement and mapping
Set of relations subelement, wherein,
Sample video collection subelement, for obtaining Sample video collection;
Sample classification subelement, for according to the visual classification pre-set, that is concentrated to the Sample video to be every the same
This video is classified, and obtains classification samples video group corresponding to each visual classification;
In the present embodiment, video can be classified in advance, for example, can be by visual classification:Funny video, red-letter day
Video etc. being blessed, wherein, for funny video, can be divided into again:Sport category funny video, life kind funny video, amusement class
Funny video etc., same Sample video can be attributed to different classification videos, and the present embodiment is not construed as limiting to this.
Sample keyword subelement, for extracting the key that each classification samples video bag contains in a classification samples video group
Word, obtain sample crucial phrase;
In the present embodiment, by the title in each classification samples video in classification samples video group, brief introduction, comment etc.
Segmented, the processing of cutting and redundancy, each pass that each classification samples video bag contains in the classification samples video group can be obtained
Keyword, form sample crucial phrase.
Word frequency statisticses subelement, for carrying out word frequency statisticses to the keyword in the sample crucial phrase;
Mapping relations build subelement, are mapped for the keyword of word frequency top N to be write into the classification samples video group
Sort key word set, the mapping relations of structure visual classification and the sort key word set;
In the present embodiment, the keyword of the higher top N of word frequency is won out, the classification as corresponding visual classification is closed
Keyword collection.
Mapping relations collection subelement, is used for, and judges whether classification samples video group is extracted and finishes, if not, notice sample
Keyword subelement, if it is, the mapping relations according to structure form mapping relations collection.
Matching unit, for the crucial phrase to be matched with the first sort key word set, obtain the pass
Each keyword to match in keyword group with the first sort key word set, obtain the first sort key phrase;
Traversal Unit, whether the sort key word set for judging the mapping relations concentration, which is extracted, finishes, if it is, knot
Line journey, if not, the next second sort key word set of extraction, the crucial phrase and the second sort key word set are entered
Row matching, obtains each keyword to match in the crucial phrase with the second sort key word set, obtains the second classification
Crucial phrase, until the sort key word set that the mapping relations are concentrated is extracted and finished.
Classification quality fraction computing module 34, for obtaining the parameter that browses of the video to be sorted, browsed according to described in
Parameter and the sort key phrase calculate the classification quality fraction of the video to be sorted;
In the present embodiment, browsing parameter includes:Pageview, like number, do not like number, issue date etc..Browsing parameter can
By analyzing video acquisition to be sorted.
In the present embodiment, as an alternative embodiment, the classification quality fraction is calculated using following formula:
F=ξ1*lg(brows)+ξ2*lg10((likes-dislikes)*(likes/(dislikes+1)))+ξ3*lg
(date)+ξ4*key-words
In formula,
F is classification quality fraction;
ξ1For pageview weight coefficient;
Brows is pageview;
ξ2To like several weight coefficients;
Likes is to like several;
Dislikes is not like number;
ξ3For date weight coefficient;
Date is number of days of the issue date apart from current date;
ξ4For sort key phrase weight coefficient;
Key-words is the sort key word number that sort key phrase includes.
In the present embodiment, as an alternative embodiment, pageview weight coefficient is 0.25, described likes several weight coefficients to be
0.0.375, the date weight coefficient is 0.1, and, the sort key phrase weight coefficient is 0.025.
Visual classification module 35, will if the classification quality fraction exceedes the classification quality score threshold pre-set
The video to be sorted ranges classification video corresponding to the sort key phrase.
In the present embodiment, classification quality score threshold can be set according to being actually needed, can be with for different classification videos
Different classification quality score thresholds is set, identical classification quality score threshold can also be set.
In the present embodiment, as an alternative embodiment, the device can also include:
Sort recommendations module 36, for corresponding to the sort key phrase in classification classify video in, according to described
Classification quality fraction is ranked up, and is obtained video corresponding to the classification quality fraction of M positions, recommendation before sorting and is regarded to what is pre-set
The homepage of frequency website.
As another alternative embodiment, the device can also include:
Pushing module 37, the user's collection for video of classifying corresponding to the sort key phrase is registered for obtaining, will be sorted
Each user that video corresponding to the classification quality fraction of preceding M positions is concentrated to the user of acquisition pushes.
The device of the present embodiment, it can be used for the technical scheme for performing embodiment of the method shown in Fig. 1 and Fig. 2, it realizes former
Reason is similar with technique effect, and here is omitted.
It should be noted that herein, such as first and second or the like relational terms are used merely to a reality
Body or operation make a distinction with another entity or operation, and not necessarily require or imply and deposited between these entities or operation
In any this actual relation or order.Moreover, term " comprising ", "comprising" or its any other variant are intended to
Nonexcludability includes, so that process, method, article or equipment including a series of elements not only will including those
Element, but also the other element including being not expressly set out, or it is this process, method, article or equipment also to include
Intrinsic key element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that
Other identical element also be present in process, method, article or equipment including the key element.
Each embodiment in this specification is described by the way of related, identical similar portion between each embodiment
Divide mutually referring to what each embodiment stressed is the difference with other embodiment.
For device embodiment, because it is substantially similar to embodiment of the method, so the comparison of description is simple
Single, the relevent part can refer to the partial explaination of embodiments of method.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system including the system of processor or other can be held from instruction
The system of row system, device or equipment instruction fetch and execute instruction) use, or combine these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium " can any can be included, store, communicate, propagate or pass
Defeated program is for instruction execution system, device or equipment or the dress used with reference to these instruction execution systems, device or equipment
Put.The more specifically example (non-exhaustive list) of computer-readable medium includes following:Electricity with one or more wiring
Connecting portion (electronic installation), portable computer diskette box (magnetic device), random access memory (RAM), read-only storage
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device, and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium, which can even is that, to print the paper of described program thereon or other are suitable
Medium, because can then enter edlin, interpretation or if necessary with it for example by carrying out optical scanner to paper or other media
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.
In the above-described embodiment, multiple steps or method can use storage to be performed in memory and by suitable instruction
The software or firmware that system performs are realized.If for example, being realized with hardware, with another embodiment, can use
Any one of following technology well known in the art or their combination are realized:With for realizing logic work(to data-signal
The discrete logic of the logic gates of energy, there is the application specific integrated circuit of suitable combinational logic gate circuit, programmable gate
Array (PGA), field programmable gate array (FPGA) etc..
The embodiment of the present invention also provides a kind of electronic equipment, and the electronic equipment includes the dress described in foregoing any embodiment
Put.
Fig. 4 is the structural representation of electronic equipment one embodiment of the present invention, it is possible to achieve is implemented shown in Fig. 1-3 of the present invention
The flow of example, as shown in figure 4, above-mentioned electronic equipment can include:Housing 41, processor 42, memory 43, circuit board 44 and electricity
Source circuit 45, wherein, circuit board 44 is placed in the interior volume that housing 41 surrounds, and processor 42 and memory 43 are arranged on circuit
On plate 44;Power circuit 45, for each circuit or the device power supply for above-mentioned electronic equipment;Memory 43 is used to store and can hold
Line program code;Processor 42 is run and executable program generation by reading the executable program code stored in memory 43
Program corresponding to code, for performing the video classification methods described in foregoing any embodiment.
Processor 42 to the specific implementation procedures of above-mentioned steps and processor 42 by run executable program code come
The step of further performing, the description of Fig. 1-3 illustrated embodiments of the present invention is may refer to, will not be repeated here.
The electronic equipment exists in a variety of forms, includes but is not limited to:
(1) mobile communication equipment:The characteristics of this kind equipment is that possess mobile communication function, and to provide speech, data
Communicate as main target.This Terminal Type includes:Smart mobile phone (such as iPhone), multimedia handset, feature mobile phone, and it is low
Hold mobile phone etc..
(2) super mobile personal computer equipment:This kind equipment belongs to the category of personal computer, there is calculating and processing work(
Can, typically also possess mobile Internet access characteristic.This Terminal Type includes:PDA, MID and UMPC equipment etc., such as iPad.
(3) portable entertainment device:This kind equipment can show and play content of multimedia.The kind equipment includes:Audio,
Video player (such as iPod), handheld device, e-book, and intelligent toy and portable car-mounted navigation equipment.
(4) server:The equipment for providing the service of calculating, the composition of server are total including processor, hard disk, internal memory, system
Line etc., server is similar with general computer architecture, but due to needing to provide highly reliable service, therefore in processing energy
Power, stability, reliability, security, scalability, manageability etc. require higher.
(5) other electronic equipments with data interaction function.
Those skilled in the art are appreciated that to realize all or part of step that above-described embodiment method carries
Suddenly it is that by program the hardware of correlation can be instructed to complete, described program can be stored in a kind of computer-readable storage medium
In matter, the program upon execution, including one or a combination set of the step of embodiment of the method.
For convenience of description, it is to be divided into various units/modules with function to describe respectively to describe apparatus above.Certainly, exist
The function of each unit/module can be realized in same or multiple softwares and/or hardware when implementing of the invention.
As seen through the above description of the embodiments, those skilled in the art can be understood that this
Invention can add the mode of required general hardware platform to realize by software.Based on such understanding, the present invention
The part that is substantially contributed in other words to prior art of technical scheme can be embodied in the form of software product, should
Computer software product can be stored in storage medium, such as ROM/RAM, magnetic disc, CD, including some instructions are causing
One computer equipment (can be personal computer, server, or network equipment etc.) perform each embodiment of the present invention or
Method described in some parts of person's embodiment.
The foregoing is only a specific embodiment of the invention, but protection scope of the present invention is not limited thereto, any
Those familiar with the art the invention discloses technical scope in, the change or replacement that can readily occur in, all should
It is included within the scope of the present invention.Therefore, protection scope of the present invention should be defined by scope of the claims.
Claims (10)
- A kind of 1. video classification methods, it is characterised in that including:Obtain video to be sorted;The keyword in the video to be sorted is extracted, obtains crucial phrase;The sort key word set that the crucial phrase maps with each visual classification pre-set successively is matched, respectively To the sort key phrase to match with each sort key word set;The parameter that browses of the video to be sorted is obtained, is browsed according to described in described in parameter and sort key phrase calculating The classification quality fraction of video to be sorted;If the classification quality fraction exceedes the classification quality score threshold pre-set, the video to be sorted is ranged Classification video corresponding to the sort key phrase.
- 2. video classification methods according to claim 1, it is characterised in that it is described by the crucial phrase successively with advance The sort key word set of each visual classification mapping set is matched, and respectively obtains point to match with each sort key word set Class keywords group includes:Concentrated in the mapping relations of visual classification and sort key word set, extract the first sort key word set;The crucial phrase is matched with the first sort key word set, obtained in the crucial phrase with described first Each keyword that sort key word set matches, obtain the first sort key phrase;Judge whether the sort key word set of the mapping relations concentration is extracted to finish, if it is, terminating flow, if not, carrying One second sort key word set is removed, the crucial phrase is matched with the second sort key word set, described in acquisition Each keyword to match in crucial phrase with the second sort key word set, the second sort key phrase is obtained, until institute The sort key word set for stating mapping relations concentration is extracted and finished.
- 3. video classification methods according to claim 2, it is characterised in that the acquisition visual classification and sort key word The mapping relations collection of collection includes:Obtain Sample video collection;According to the visual classification pre-set, each Sample video concentrated to the Sample video is classified, obtained each Classification samples video group corresponding to visual classification;The keyword that each classification samples video bag contains in a classification samples video group is extracted, obtains sample crucial phrase;Word frequency statisticses are carried out to the keyword in the sample crucial phrase;The keyword of word frequency top N is write to the sort key word set of the classification samples video group mapping, builds visual classification With the mapping relations of the sort key word set;Judge whether classification samples video group is extracted to finish, if not, performing each in one classification samples video group of the extraction The step of keyword that classification samples video bag contains, if it is, the mapping relations according to structure form mapping relations collection.
- 4. according to the video classification methods described in any one of claims 1 to 3, it is characterised in that calculate described point using following formula Class mass fraction:F=ξ1*lg(brows)+ξ2*lg10((likes-dislikes)*(likes/(dislikes+1)))+ξ3*lg(date) +ξ4*key-wordsIn formula,F is classification quality fraction;ξ1For pageview weight coefficient;Brows is pageview;ξ2To like several weight coefficients;Likes is to like several;Dislikes is not like number;ξ3For date weight coefficient;Date is number of days of the issue date apart from current date;ξ4For sort key phrase weight coefficient;Key-words is the sort key word number that sort key phrase includes.
- 5. video classification methods according to claim 4, it is characterised in that the pageview weight coefficient is 0.25, institute State that to like several weight coefficients be 0.0.375, the date weight coefficient is 0.1, and, the sort key phrase weight coefficient For 0.025.
- 6. according to the video classification methods described in any one of claims 1 to 3, it is characterised in that methods described also includes:Classify corresponding to the sort key phrase in classification in video, be ranked up, obtain according to the classification quality fraction Video corresponding to the classification quality fraction of M positions before sorting is taken, is recommended to the homepage of the video website pre-set.
- 7. video classification methods according to claim 6, it is characterised in that methods described also includes:The user's collection for registering classification video corresponding to the sort key phrase is obtained, by the classification quality fraction of M positions before sequence Each user that corresponding video is concentrated to the user of acquisition pushes.
- A kind of 8. visual classification device, it is characterised in that including:Video acquiring module, keyword extracting module, matching module, Classification quality fraction computing module and visual classification module, wherein,Video acquiring module, for obtaining video to be sorted;Keyword extracting module, for extracting the keyword in the video to be sorted, obtain crucial phrase;Matching module, the sort key word set for the crucial phrase to be mapped with each visual classification pre-set successively are entered Row matching, respectively obtains the sort key phrase to match with each sort key word set;Classification quality fraction computing module, for obtaining the parameter that browses of the video to be sorted, browsed according to described in parameter with And the sort key phrase calculates the classification quality fraction of the video to be sorted;Visual classification module, if the classification quality fraction exceedes the classification quality score threshold pre-set, treated described Classification video ranges classification video corresponding to the sort key phrase.
- 9. visual classification device according to claim 8, it is characterised in that the matching module includes:Sort key word Collect extraction unit, matching unit and Traversal Unit, wherein,Sort key word set extraction unit, concentrated for the mapping relations in visual classification and sort key word set, extraction first Sort key word set;Matching unit, for the crucial phrase to be matched with the first sort key word set, obtain the keyword Each keyword to match in group with the first sort key word set, obtain the first sort key phrase;Traversal Unit, whether the sort key word set for judging the mapping relations concentration, which is extracted, finishes, if it is, terminating stream Journey, if not, the next second sort key word set of extraction, by the crucial phrase and the second sort key word set progress Match somebody with somebody, obtain each keyword to match in the crucial phrase with the second sort key word set, obtain the second sort key Phrase, until the sort key word set that the mapping relations are concentrated is extracted and finished.
- 10. a kind of electronic equipment, it is characterised in that the electronic equipment includes:Housing, processor, memory, circuit board and electricity Source circuit, wherein, circuit board is placed in the interior volume that housing surrounds, and processor and memory are set on circuit boards;Power supply Circuit, for each circuit or the device power supply for above-mentioned electronic equipment;Memory is used to store executable program code;Processing The executable program code that device is stored by reading in memory runs program corresponding with executable program code, for holding Video classification methods described in the foregoing any claim 1-7 of row.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610375160.3A CN107451148A (en) | 2016-05-31 | 2016-05-31 | Video classification method and device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610375160.3A CN107451148A (en) | 2016-05-31 | 2016-05-31 | Video classification method and device and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107451148A true CN107451148A (en) | 2017-12-08 |
Family
ID=60485903
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610375160.3A Pending CN107451148A (en) | 2016-05-31 | 2016-05-31 | Video classification method and device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107451148A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108804598A (en) * | 2018-05-29 | 2018-11-13 | 王妃 | Cloud atlas distributed video sorting technique |
CN108966007A (en) * | 2018-09-03 | 2018-12-07 | 青岛海信电器股份有限公司 | A kind of method and device for distinguishing video scene at HDMI |
CN109145151A (en) * | 2018-06-20 | 2019-01-04 | 北京达佳互联信息技术有限公司 | A kind of the emotional semantic classification acquisition methods and device of video |
CN109344908A (en) * | 2018-10-30 | 2019-02-15 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating model |
CN109522450A (en) * | 2018-11-29 | 2019-03-26 | 腾讯科技(深圳)有限公司 | A kind of method and server of visual classification |
CN110366043A (en) * | 2019-08-20 | 2019-10-22 | 北京字节跳动网络技术有限公司 | Method for processing video frequency, device, electronic equipment and readable medium |
CN110619061A (en) * | 2018-12-25 | 2019-12-27 | 北京时光荏苒科技有限公司 | Video classification method and device, electronic equipment and readable medium |
CN111246257A (en) * | 2020-03-17 | 2020-06-05 | 百度在线网络技术(北京)有限公司 | Video recommendation method, device, equipment and storage medium |
CN111385659A (en) * | 2018-12-29 | 2020-07-07 | 广州市百果园信息技术有限公司 | Video recommendation method, device, equipment and storage medium |
CN112016613A (en) * | 2020-08-26 | 2020-12-01 | 广州市百果园信息技术有限公司 | Training method and device for video content classification model, computer equipment and medium |
CN112084370A (en) * | 2020-09-10 | 2020-12-15 | 维沃移动通信有限公司 | Video processing method and device and electronic equipment |
CN114782797A (en) * | 2022-06-21 | 2022-07-22 | 深圳市万物云科技有限公司 | House scene classification method, device and equipment and readable storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102945246A (en) * | 2012-09-28 | 2013-02-27 | 北界创想(北京)软件有限公司 | Method and device for processing network information data |
CN103186550A (en) * | 2011-12-27 | 2013-07-03 | 盛乐信息技术(上海)有限公司 | Method and system for generating video-related video list |
CN104199896A (en) * | 2014-08-26 | 2014-12-10 | 海信集团有限公司 | Video similarity determining method and video recommendation method based on feature classification |
CN104219575A (en) * | 2013-05-29 | 2014-12-17 | 酷盛(天津)科技有限公司 | Related video recommending method and system |
US20150186503A1 (en) * | 2012-10-12 | 2015-07-02 | Tencent Technology (Shenzhen) Company Limited | Method, system, and computer readable medium for interest tag recommendation |
CN105512331A (en) * | 2015-12-28 | 2016-04-20 | 海信集团有限公司 | Video recommending method and device |
-
2016
- 2016-05-31 CN CN201610375160.3A patent/CN107451148A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103186550A (en) * | 2011-12-27 | 2013-07-03 | 盛乐信息技术(上海)有限公司 | Method and system for generating video-related video list |
CN102945246A (en) * | 2012-09-28 | 2013-02-27 | 北界创想(北京)软件有限公司 | Method and device for processing network information data |
US20150186503A1 (en) * | 2012-10-12 | 2015-07-02 | Tencent Technology (Shenzhen) Company Limited | Method, system, and computer readable medium for interest tag recommendation |
CN104219575A (en) * | 2013-05-29 | 2014-12-17 | 酷盛(天津)科技有限公司 | Related video recommending method and system |
CN104199896A (en) * | 2014-08-26 | 2014-12-10 | 海信集团有限公司 | Video similarity determining method and video recommendation method based on feature classification |
CN105512331A (en) * | 2015-12-28 | 2016-04-20 | 海信集团有限公司 | Video recommending method and device |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108804598A (en) * | 2018-05-29 | 2018-11-13 | 王妃 | Cloud atlas distributed video sorting technique |
CN109145151A (en) * | 2018-06-20 | 2019-01-04 | 北京达佳互联信息技术有限公司 | A kind of the emotional semantic classification acquisition methods and device of video |
CN108966007A (en) * | 2018-09-03 | 2018-12-07 | 青岛海信电器股份有限公司 | A kind of method and device for distinguishing video scene at HDMI |
CN108966007B (en) * | 2018-09-03 | 2021-08-31 | 海信视像科技股份有限公司 | Method and device for distinguishing video scenes under HDMI |
CN109344908A (en) * | 2018-10-30 | 2019-02-15 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating model |
CN109522450A (en) * | 2018-11-29 | 2019-03-26 | 腾讯科技(深圳)有限公司 | A kind of method and server of visual classification |
US11741711B2 (en) | 2018-11-29 | 2023-08-29 | Tencent Technology (Shenzhen) Company Limited | Video classification method and server |
CN110619061A (en) * | 2018-12-25 | 2019-12-27 | 北京时光荏苒科技有限公司 | Video classification method and device, electronic equipment and readable medium |
CN111385659A (en) * | 2018-12-29 | 2020-07-07 | 广州市百果园信息技术有限公司 | Video recommendation method, device, equipment and storage medium |
CN110366043A (en) * | 2019-08-20 | 2019-10-22 | 北京字节跳动网络技术有限公司 | Method for processing video frequency, device, electronic equipment and readable medium |
CN111246257A (en) * | 2020-03-17 | 2020-06-05 | 百度在线网络技术(北京)有限公司 | Video recommendation method, device, equipment and storage medium |
CN111246257B (en) * | 2020-03-17 | 2022-07-26 | 百度在线网络技术(北京)有限公司 | Video recommendation method, device, equipment and storage medium |
CN112016613A (en) * | 2020-08-26 | 2020-12-01 | 广州市百果园信息技术有限公司 | Training method and device for video content classification model, computer equipment and medium |
CN112084370A (en) * | 2020-09-10 | 2020-12-15 | 维沃移动通信有限公司 | Video processing method and device and electronic equipment |
CN114782797A (en) * | 2022-06-21 | 2022-07-22 | 深圳市万物云科技有限公司 | House scene classification method, device and equipment and readable storage medium |
CN114782797B (en) * | 2022-06-21 | 2022-09-20 | 深圳市万物云科技有限公司 | House scene classification method, device and equipment and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107451148A (en) | Video classification method and device and electronic equipment | |
CN110704674B (en) | Video playing integrity prediction method and device | |
CN107944913B (en) | High-potential user purchase intention prediction method based on big data user behavior analysis | |
CN104899302B (en) | Recommend the method and apparatus of music to user | |
CN105138670B (en) | Audio file label generating method and system | |
CN106294425B (en) | The automatic image-text method of abstracting and system of commodity network of relation article | |
CN110532479A (en) | A kind of information recommendation method, device and equipment | |
CN103810162B (en) | The method and system of recommendation network information | |
US20150205580A1 (en) | Method and System for Sorting Online Videos of a Search | |
CN106339502A (en) | Modeling recommendation method based on user behavior data fragmentation cluster | |
CN103116588A (en) | Method and system for personalized recommendation | |
CN108595660A (en) | Label information generation method, device, storage medium and the equipment of multimedia resource | |
CN108304399A (en) | The recommendation method and device of Web content | |
Katz et al. | Using Wikipedia to boost collaborative filtering techniques | |
CN110222260A (en) | A kind of searching method, device and storage medium | |
US20160012454A1 (en) | Database systems for measuring impact on the internet | |
CN104834739B (en) | Internet information storage system | |
Zhou et al. | Relevance feature mapping for content-based multimedia information retrieval | |
CN108280124A (en) | Product classification method and device, ranking list generation method and device, electronic equipment | |
CN104615741A (en) | Cloud computing based cold start item recommending method and device | |
CN104199838B (en) | A kind of user model constructing method based on label disambiguation | |
CN107341199A (en) | A kind of recommendation method based on documentation & info general model | |
CN107316229A (en) | A kind of method and device of online displaying performance project information | |
Kim | Recommender system based on product taxonomy in e-commerce sites. | |
CN108733669A (en) | A kind of personalized digital media content recommendation system and method based on term vector |
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: 20171208 |
|
RJ01 | Rejection of invention patent application after publication |