CN108898415A - A kind of the flow index of correlation prediction technique and device of video collection of drama - Google Patents
A kind of the flow index of correlation prediction technique and device of video collection of drama Download PDFInfo
- Publication number
- CN108898415A CN108898415A CN201810533410.0A CN201810533410A CN108898415A CN 108898415 A CN108898415 A CN 108898415A CN 201810533410 A CN201810533410 A CN 201810533410A CN 108898415 A CN108898415 A CN 108898415A
- Authority
- CN
- China
- Prior art keywords
- drama
- video collection
- data
- predicted
- characteristic
- 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
- 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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0203—Market surveys; Market polls
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Accounting & Taxation (AREA)
- Entrepreneurship & Innovation (AREA)
- Finance (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Game Theory and Decision Science (AREA)
- Data Mining & Analysis (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides a kind of flow index of correlation prediction technique of video collection of drama, device, electronic equipment and storage medium, this method to include：The information data of video collection of drama to be predicted is obtained, information data includes：Row's multicast data of video collection of drama to be predicted, video collection of drama to be predicted play the flow index of correlation data that public praise data relevant to video collection of drama to be predicted and video collection of drama to be predicted in preceding first preset time period play played video collection of drama relevant with video collection of drama to be predicted in preceding second preset time period；Information data is converted according to preset data form, obtains characteristic；By characteristic input prediction model, the flow index of correlation predicted value of video collection of drama to be predicted is obtained.Using the flow index of correlation prediction technique of video collection of drama provided by the invention, can before video collection of drama broadcasts the long period, video flow index of correlation data in a period of time after broadcast can be predicted, to provide higher reference value.
Description
Technical field
The present invention relates to multimedia technology field, in particular to the flow index of correlation prediction technique of a kind of video collection of drama,
Device, electronic equipment and storage medium.
Background technique
For video class enterprise, before new video collection of drama broadcasts, it usually needs it is pre- to do the relevant index of flow to it
It surveys, for example, predicting the video collection of drama in total playback volume of website.Carry out the prediction of flow relevant index can for programme machinating,
Buying, decision and later period finance etc. provide important reference.
The video traffic prediction carried out in the prior art is short-term forecast, and such as couple of days carries out before video collection of drama plays
Prediction, since predicted time is later, so that the reference value that prediction result can be provided may not be very high.
Summary of the invention
A kind of flow index of correlation prediction technique for being designed to provide video collection of drama of the embodiment of the present invention, device, electricity
Sub- equipment and storage medium predict flow index of correlation long-term in the progress of video collection of drama with realizing, to provide higher
Reference value.
Specific technical solution is as follows：
The embodiment of the invention provides a kind of flow index of correlation prediction technique of video collection of drama, the method includes：
The information data of video collection of drama to be predicted is obtained, the information data includes：The row of the video collection of drama to be predicted
Multicast data, the video collection of drama to be predicted play the interior public praise relevant to the video collection of drama to be predicted of preceding first preset time period
Data and the video collection of drama to be predicted play relevant to the video collection of drama to be predicted in preceding second preset time period
Play the flow index of correlation data of video collection of drama；
Feature extraction is carried out to the information data, obtains data to be converted, and by the data to be converted according to default
Data format is converted, the data after obtaining format conversion, the characteristic as the video collection of drama to be predicted；
By the characteristic input prediction model, the flow index of correlation prediction of the video collection of drama to be predicted is obtained
Value, the prediction model are acute according to the characteristic of the preset data form of Sample video collection of drama and the Sample video
What the known flow index of correlation data training of collection obtained, the Sample video collection of drama is played video collection of drama.
Optionally, described to convert the data to be converted according to preset data form, after obtaining format conversion
Data, as the characteristic of the video collection of drama to be predicted, including：
The video collection of drama to be predicted is played relevant to the video collection of drama to be predicted in preceding first preset time period
The data to be converted of public praise data and the video collection of drama to be predicted play preceding second preset time period it is interior with it is described to be predicted
The data to be converted of the flow index of correlation data of the relevant played video collection of drama of video collection of drama carry out logarithmic transformation；
By row's multicast data of the data obtained after the logarithmic transformation and the video collection of drama to be predicted wait turn
It changes data to be normalized, obtains the characteristic of the preset data form of the video collection of drama to be predicted.
Optionally, the prediction model is linear regression model (LRM).
Optionally, based on the following objective function training linear regression model (LRM)：
Wherein, n is the number of the Sample video collection of drama, and y is that the known flow correlation of the Sample video collection of drama refers to
Data are marked, f (w, x) is the linear regression model (LRM), andWherein k is characterized the number of data, xiIt is k
Ith feature data in characteristic；wiFor the parameter of ith feature data, L is regular terms；
Determine the process of the parameter value of the characteristic of the linear regression model (LRM), including：
By the k of the known flow index of correlation data of the Sample video collection of drama and the Sample video collection of drama
Characteristic substitutes into current objective function and is calculated, and obtains error amount；
Judge whether the error amount is less than preset error amount threshold value, if it is not, then adjusting the ginseng of the k characteristic
Numerical value, the objective function after being changed, and return the known flow index of correlation data of the Sample video collection of drama, with
And k characteristic of the Sample video collection of drama substitutes into the step of current objective function is calculated, obtains error amount；
If the error amount is less than preset error amount threshold value, completed the current parameter value of characteristic as training
The linear regression model (LRM) characteristic parameter value.
Optionally, described by the characteristic input prediction model, obtain the flow phase of the video collection of drama to be predicted
Index predicted value is closed, including：
By the characteristic xiIt is acute to obtain the video to be predicted by the linear regression model (LRM) f (w, x) that input training is completed
The flow index of correlation predicted value of collection.
The embodiment of the invention also provides a kind of flow index of correlation prediction meanss of video collection of drama, described device includes：
Information data obtains module, and for obtaining the information data of video collection of drama to be predicted, the information data includes：Institute
State row's multicast data of video collection of drama to be predicted, the video collection of drama to be predicted plays in preceding first preset time period with described to pre-
Survey the relevant public praise data of video collection of drama and the video collection of drama to be predicted play in preceding second preset time period and it is described to
Predict the flow index of correlation data of the relevant played video collection of drama of video collection of drama；
Characteristic extracting module obtains data to be converted for carrying out feature extraction to the information data, and will it is described to
Change data is converted according to preset data form, the data after obtaining format conversion, as the video collection of drama to be predicted
Characteristic；
Prediction module, for obtaining the flow of the video collection of drama to be predicted for the characteristic input prediction model
Index of correlation predicted value, the prediction model are according to the characteristic of the preset data form of Sample video collection of drama, Yi Jisuo
The known flow index of correlation data training for stating Sample video collection of drama obtains, and the Sample video collection of drama is played video
Collection of drama.
Optionally, the characteristic extracting module, is specifically used for：
The video collection of drama to be predicted is played relevant to the video collection of drama to be predicted in preceding first preset time period
The data to be converted of public praise data and the video collection of drama to be predicted play preceding second preset time period it is interior with it is described to be predicted
The data to be converted of the flow index of correlation data of the relevant played video collection of drama of video collection of drama carry out logarithmic transformation；
By row's multicast data of the data obtained after the logarithmic transformation and the video collection of drama to be predicted wait turn
It changes data to be normalized, obtains the characteristic of the preset data form of the video collection of drama to be predicted.
Optionally, the prediction model is linear regression model (LRM).
Optionally, described device further includes：
Training module, for based on the following objective function training linear regression model (LRM)：
Wherein, n is the number of the Sample video collection of drama, and y is that the known flow correlation of the Sample video collection of drama refers to
Data are marked, f (w, x) is the linear regression model (LRM), andWherein k is characterized the number of data, xiIt is k
Ith feature data in characteristic；wiFor the parameter of ith feature data, L is regular terms；
The training module is specifically used for, by the known flow index of correlation data of the Sample video collection of drama, and
K characteristic of the Sample video collection of drama substitutes into current objective function and is calculated, and obtains error amount；
Judge whether the error amount is less than preset error amount threshold value, if it is not, then adjusting the ginseng of the k characteristic
Numerical value, the objective function after being changed, and return the known flow index of correlation data of the Sample video collection of drama, with
And k characteristic of the Sample video collection of drama substitutes into the step of current objective function is calculated, obtains error amount；
If the error amount is less than preset error amount threshold value, completed the current parameter value of characteristic as training
The linear regression model (LRM) characteristic parameter value.
Optionally, the prediction module, is specifically used for：
By the characteristic xiIt is acute to obtain the video to be predicted by the linear regression model (LRM) f (w, x) that input training is completed
The flow index of correlation predicted value of collection.
The embodiment of the invention also provides a kind of electronic equipment, including processor, communication interface, memory and communication are total
Line, wherein the processor, the communication interface, the memory complete mutual communication by the communication bus；
The memory, for storing computer program；
The processor when for executing the program stored on the memory, realizes any of the above-described method and step.
The embodiment of the invention also provides a kind of computer readable storage medium, the computer readable storage medium memory
Computer program is contained, the computer program realizes any of the above-described method and step when being executed by processor.
Using the flow index of correlation prediction technique of video collection of drama provided in an embodiment of the present invention, device, electronic equipment with
And storage medium, the row that can obtain video collection of drama to be predicted broadcast public praise data before release data, broadcasting in preset time period
And the flow index of correlation data of other video collection of dramas relevant to the video collection of drama, instruction in advance is inputted after handling it
The prediction model perfected can broadcast the preceding long period in video collection of drama, can predict video a period of time after broadcast
Interior flow index of correlation data, to provide higher reference value.
Certainly, implement any of the products of the present invention or method it is not absolutely required at the same reach all the above excellent
Point.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described.
Fig. 1 is a kind of flow chart of the flow index of correlation prediction technique of video collection of drama provided in an embodiment of the present invention；
Fig. 2 is a kind of structural representation of the flow index of correlation prediction meanss of video collection of drama provided in an embodiment of the present invention
Figure；
Fig. 3 is a kind of structural schematic diagram of electronic equipment provided in an embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention is described.
The embodiment of the invention provides a kind of flow index of correlation prediction technique of video collection of drama, device, electronic equipment with
And storage medium, flow index of correlation long-term in the progress of video collection of drama is predicted with realizing, to provide higher reference price
Value.
Referring to Fig. 1, Fig. 1 is a kind of stream of the flow index of correlation prediction technique of video collection of drama provided in an embodiment of the present invention
Cheng Tu may comprise steps of：
Step S101：The information data of video collection of drama to be predicted is obtained, information data includes：The row of video collection of drama to be predicted
Multicast data, video collection of drama to be predicted play public praise data relevant to video collection of drama to be predicted in preceding first preset time period, with
And video collection of drama to be predicted plays played video collection of drama relevant to video collection of drama to be predicted in preceding second preset time period
Flow index of correlation data；
In embodiments of the present invention, the flow index of correlation of video collection of drama can be the video collection of drama in video website
Total playback volume is also possible to user and plays the flow etc. that the video collection of drama generates in total, do not limit this.
In embodiments of the present invention, video collection of drama to be predicted can be the video collection of drama that starts broadcasting after longer period, such as
The video collection of drama to start broadcasting after half a year, the prediction that flow index of correlation data are carried out to it can be after predicting that the video collection of drama starts broadcasting
Total broadcasting time in a period of time inherent video website.
Before the prediction for carrying out flow index of correlation to video collection of drama to be predicted, the row of the video collection of drama can be first obtained
The interior public praise data relevant to the video collection of drama of preset time period and the video collection of drama before multicast data, the video collection of drama play
The flow index of correlation data of the interior video collection of drama played relevant to the video collection of drama of preset time period before playing.Below
It is illustrated in conjunction with specific example.
If some video collection of drama to be predicted plays after half a year, for convenience, it is denoted as video collection of drama A to be predicted,
In the following embodiments, carry out generation with video collection of drama A to be predicted to refer to and start broadcasting after longer period, and need to carry out flow correlation
The video collection of drama of index prediction.
In order to predict the video collection of drama A a period of time, such as broadcasting time in three months after broadcast, can first obtain
Row's multicast data of video collection of drama A, wherein row's multicast data may include following data：
1), the subject matter data of video collection of drama A, the classification of related specific subject matter can carry out in advance it is customized, for example,
It is Metropolitan a, ancient costume class b that subject matter classification, which can be defined, passes through class c, and struggle against class d etc. in palace, can be corresponding by each subject matter classification
Value be set as 0 or 1, if the palace that the video collection of drama A is ancient costume is struggled against acute, the b in subject matter data, d can be determined as
1, a, c are determined as 0.
2), play video collection of drama A TV station's data, related TV station can carry out in advance it is customized, for example, can
To define the Chinese Central Television (CCTV) as a class, Hunan TV platform is b class, and Jiangsu TV station is c class, and Zhejiang Television Station is d class etc., can be with
The corresponding value of every one kind TV station is set as 0 or 1, for example, if video collection of drama A can be in the Chinese Central Television (CCTV) and Hunan electricity
Television stations plays out, then a, b in TV station's data can be determined as 1, c, d are determined as 0.
3) the video website data of video collection of drama A, are played, related video website can carry out customized, example in advance
Such as, can define iqiyi.com website is a class, and Sohu.com is b class, and youku.com website is c class etc., if video collection of drama A is only odd in love
Skill website is broadcasted, then a in video website data can be determined as 1, b, c are determined as 0.
4), the broadcast strategy data of video collection of drama A, wherein broadcast strategy can be customized in advance, for example, can define every
It plays 2 week to integrate as a class, plays 4 weekly and integrate as b class, play 1 daily and integrate as c class, play 2 daily and integrate as d class etc., if video is acute
Collecting A is to play 2 collection weekly, then a in broadcast strategy data can be determined as 1, b, c, d are determined as 0.
5), the collection number data of video collection of drama, can be customized according to the range progress of collection number, for example, collection number can be defined
10-20 is a class, integrates number 21-30 as b class, integrates number 31-40 as c class, integrate number 41-50 as d class etc., if the collection number of video collection of drama A is
36, then can by collect number data in c be determined as 1, a, b, d are determined as 0, it is of course also possible to not according to collection number range into
Row classification directly uses the numerical value of collection number of video collection of drama as above-mentioned collection number data.
6), video collection of drama, can be customized according to the range progress of charge number of days in the charge number of days data of video website,
For example, can define charge number of days 0 is a class, charge number of days 1-10 is b class, and charge number of days 11-20 is c class etc., if video collection of drama
A is 15 in the charge number of days of video website, then the c in number of days data of charging is determined as 1, a, b are determined as 0.Certainly, may be used
Not classified according to the range of charge number of days, directly use the numerical value of charge number of days as above-mentioned charge number of days data.
In embodiments of the present invention, can also obtain before video collection of drama A is played in preset time period with the video collection of drama A phase
The public praise data of pass, may include following data：
1), network influence of the actor or actress involved in video collection of drama A before the video collection of drama A is played in preset time period,
Wherein actor or actress may include the director of video collection of drama, performer, playwright, screenwriter, producer etc., wherein preset time period can be before playing
It plays in the previous year to six months, network influence can be some actor or actress network temperature index in the preset time period
Average value, median or maximum value etc..
2), network of the video display mechanism that video collection of drama A is related to before the video collection of drama A is played in preset time period influences
Power, wherein video display mechanism may include the production company of video collection of drama A, distributing and releasing corporation etc., wherein preset time period before playing
It can be to play in the previous year to six months, network influence can be production company network boom in the preset time period
Spend average value, median or the maximum value etc. of index.
If 3), video collection of drama A is serial works, serial works that video collection of drama A is related to can also be obtained in video collection of drama A
Network influence before playing in preset time period.Preset time period can be broadcasting the previous year to six months before wherein playing
It is interior, network influence can for other works in the series works in addition to video collection of drama A in the preset time period network
Average value, median or maximum value of temperature index etc..
4), if video collection of drama A is the reorganization of other types works, the other types that video collection of drama A is related to can also be obtained
Network influence of the works before video collection of drama A is played in preset time period.Preset time period can be broadcasting before wherein playing
In the previous year to six months, the other types works that network influence can be related to for video collection of drama A are in the preset time period
Average value, median or the maximum value etc. of interior network temperature index.
In embodiments of the present invention, can also obtain before video collection of drama A is played in preset time period with the video collection of drama A phase
The flow index of correlation data of other video collection of dramas played closed, can specifically include following data：
1), the flow with other video collection of dramas of video collection of drama A same type before video collection of drama A is played in preset time period
Index of correlation data, it is assumed that A is that the palace bucket of ancient costume is acute, then the available related data with the video collection of drama B of its same type, example
The average value that video collection of drama B plays the flow index of correlation data in the previous year to six months in video collection of drama A is such as obtained, in
Place value or maximum value etc..
2), other video collection of dramas relevant to the actor or actress that video collection of drama A is related to preset time period before video collection of drama A is played
Interior flow index of correlation data, for example, acting the leading role in video collection of drama A is Hu Ge, then other videos that available Hu Ge takes part in a performance
Collection of drama plays the average value of the flow index of correlation data in the previous year to half a year, median or maximum value in video collection of drama A
Deng.
3), other video collection of dramas relevant to the video display mechanism that video collection of drama A is related to are when default before video collection of drama A is played
Between flow index of correlation data in section, for example, distributing and releasing corporation is C in video collection of drama A, then available C company distribution
Other video collection of dramas video collection of drama A play the previous year to half a year in flow index of correlation data average value, median or
Person's maximum value etc..
After obtaining above- mentioned information data, data can be carried out to these information datas according to method in the prior art
Cleaning, such as completion have the information data of missing, remove the information data of format error, remove unwanted information data etc.,
Specific process is not described herein, and may refer to the prior art.Following step is the Information Number completed based on data cleansing
According to progress.
Step S102：Feature extraction is carried out to information data, obtains data to be converted, and by data to be converted according to default
Data format is converted, the data after obtaining format conversion, the characteristic as video collection of drama to be predicted；
In this step, validity feature can be extracted from the information data that above-mentioned cleaning is completed, for example, to be predicted
Video collection of drama plays in preceding first preset time period in public praise data relevant to the video collection of drama, and extracting can the representative registered permanent residence
The data of upright stone tablet are used for subsequent model training.
It should be noted that any mode that feature extraction may be implemented, can be applicable to above method stream of the invention
Cheng Zhong, to this, the present invention is not specifically limited.
It in embodiments of the present invention, can also be by the data after extraction according to preset data lattice after carrying out feature extraction
Formula is converted, the data after obtaining format conversion, the characteristic as video collection of drama to be predicted；
In embodiments of the present invention, the video collection of drama A to be predicted after progress feature extraction preceding first can be played to preset
Public praise data relevant to the video collection of drama A and video collection of drama A to be predicted are played in preceding second preset time period in period
The flow index of correlation data of played video collection of drama relevant to video collection of drama A to be predicted carry out logarithmic transformation, carry out logarithm
Method in the prior art can be used when transformation, this will not be repeated here.
Row's multicast data of video collection of drama A to be predicted after data and progress feature extraction after logarithmic transformation can be carried out
Normalized, specifically, minimax normalized can be carried out to it, so that these data are transformed into [0,1]
In range.The data obtained after being normalized are as the characteristic of video collection of drama A to be predicted.
Step S103：By characteristic input prediction model, the flow index of correlation prediction of video collection of drama to be predicted is obtained
Value, prediction model are according to the known of the characteristic of the preset data form of Sample video collection of drama and Sample video collection of drama
The training of flow index of correlation data obtain, Sample video collection of drama is played video collection of drama.
After the characteristic for obtaining video collection of drama A to be predicted, it is inputted prediction model, view to be predicted can be obtained
The flow index of correlation predicted value of frequency collection of drama A, such as total playback volume of the video collection of drama A after broadcast in three months.
In embodiments of the present invention, the relevant information that the video collection of drama broadcasted can be used instructs prediction model
Practice.
For example, some Sample video collection of drama C started broadcasting before three months, then the flow arrived in these three months of current time is related
Achievement data be it is known, the characteristic of the preset data form of video collection of drama C be also it is available arrive, can be according to
The identical method of characteristic of video collection of drama A is obtained to obtain the characteristic of video collection of drama C.
For example, row's multicast data of available video collection of drama C, may include the subject matter data of video collection of drama C, video is played
TV station's data of collection of drama C play the video website data of video collection of drama C, the broadcast strategy data of video collection of drama C, video play
Collect the collection number data of C, charge number of days data of the video collection of drama C in video website.
Video collection of drama C public praise data relevant to the video collection of drama C in preset time period before broadcasting can also be obtained,
In, play before preset time period can for play the previous year to six months, that is, 15 months away from current time to 9 months it
Between.
Video collection of drama C played video relevant to the video collection of drama C in preset time period before broadcasting can also be obtained
The flow index of correlation data of collection of drama.Wherein, preset time period can be broadcasting the previous year to six months before playing, that is,
Between current time 15 months to 9 months.
For in preset time period before broadcasting acquired in video collection of drama C public praise data and flow index of correlation data with
For in preset time period before broadcasting acquired in video collection of drama A public praise data and flow index of correlation data be correspond
, this will not be repeated here.
It can will be handled for information data acquired in video collection of drama C, obtain the preset data lattice of video collection of drama C
The characteristic of formula, the process of processing may include logarithmic transformation and minimax normalized, may refer to to be predicted
The process that video collection of drama A is handled, this will not be repeated here.
In embodiments of the present invention, other multiple samples can be obtained in the way of the characteristic for obtaining video collection of drama C
The characteristic of this collection of drama, known to characteristic and each sample data based on these sample collection of dramas at latter section of broadcasting
In flow index of correlation data train prediction model.
In embodiments of the present invention, used prediction model can be linear regression model (LRM).It can be with the following method
The training linear regression model (LRM)：
An objective function is defined first, which is used to indicate the error amount that training process generates, objective function
Value it is smaller, illustrate that trained result is better.
Objective function can be defined as function：
Wherein, n is the number of Sample video collection of drama, and y is the known flow index of correlation data of Sample video collection of drama, f
(w, x) is linear regression model (LRM), andWherein k is characterized the number of data, xiFor in k characteristic
Ith feature data；wiFor the parameter of ith feature data, L is regular terms；Wherein, L is one and wiRelevant function leads to
It often can be set as L1Or L2Two kinds of forms,Wherein k is characterized data
Number, α1And α2For preset regularization coefficient, | | wi| | for the absolute value of the parameter of ith feature data.
The process that linear regression model (LRM) f (w, x) is trained namely is determinedMiddle parameters wiValue mistake
Journey, specific training process are by each x of the y value of each Sample video collection of drama and each sampleiValue substitute into above-mentioned target
In function, error amount is obtained, whether error in judgement value is less than preset error amount threshold value, if error amount is not less than preset threshold
Value, then adjust the parameters value of characteristic, the objective function after being changed.Wherein, adjustment characteristic parameter value can
To use method in the prior art, for example, gradient descent method etc..
After the corresponding data of Sample video collection of drama is substituted into current objective function by certain, obtained result is less than default
Error amount, illustrate the trained completion of current linear regression model (LRM), the flow correlation of video collection of drama to be predicted can be referred to
Mark data are predicted.
In embodiments of the present invention, specific prediction process can be：By the relevant characteristic of video collection of drama to be predicted
xiThe linear regression model (LRM) f (w, x) that input training is completed, obtains the flow index of correlation predicted value of video collection of drama to be predicted.
Since the data used in training pattern are after logarithmic transformation, correspondingly, obtaining linear regression mould
After the result of type f (w, x) output, exponential transform can be carried out to the result, using the result after progress exponential transform as to pre-
Survey the flow index of correlation predicted value of video collection of drama.
As it can be seen that using the flow index of correlation prediction technique of video collection of drama provided in an embodiment of the present invention, can obtain to
The row of prediction video collection of drama broadcasts public praise data before release data, broadcasting in preset time period and relevant to the video collection of drama
The flow index of correlation data of other video collection of dramas input trained prediction model in advance, Neng Gou after handling it
Long period before video collection of drama broadcasts can predict video flow index of correlation data in a period of time after broadcast, from
And provide higher reference value.
Referring to fig. 2, Fig. 2 is a kind of knot of the flow index of correlation prediction meanss of video collection of drama provided in an embodiment of the present invention
Structure schematic diagram may include：
Information data obtains module 201, and for obtaining the information data of video collection of drama to be predicted, information data includes：To
It is interior with video collection of drama phase to be predicted to predict that row's multicast data of video collection of drama, video collection of drama to be predicted play preceding first preset time period
The public praise data of pass and video collection of drama to be predicted play relevant to video collection of drama to be predicted in preceding second preset time period
Play the flow index of correlation data of video collection of drama；
Characteristic extracting module 202 obtains data to be converted for carrying out feature extraction to the information data, and by institute
It states data to be converted to be converted according to preset data form, the data after obtaining format conversion, as the video to be predicted
The characteristic of collection of drama；
Prediction module 203, for by characteristic input prediction model, the flow correlation for obtaining video collection of drama to be predicted to refer to
Predicted value is marked, prediction model is the characteristic and Sample video collection of drama according to the preset data form of Sample video collection of drama
The training of known flow index of correlation data obtain, Sample video collection of drama is played video collection of drama.
In embodiments of the present invention, characteristic extracting module specifically can be used for：
Video collection of drama to be predicted is played into public praise data relevant to video collection of drama to be predicted in preceding first preset time period
Data to be converted and video collection of drama to be predicted to play preceding second preset time period interior relevant to video collection of drama to be predicted
The data to be converted for playing the flow index of correlation data of video collection of drama carry out logarithmic transformation；
By the data to be converted of row's multicast data of the data obtained after logarithmic transformation and video collection of drama to be predicted into
Row normalized obtains the characteristic of the preset data form of video collection of drama to be predicted.
In embodiments of the present invention, on the basis of the flow index of correlation prediction meanss of video collection of drama shown in Fig. 2, also
May include：
Training module, for based on following objective function training linear regression model (LRM)：
Wherein, n is the number of Sample video collection of drama, and y is the known flow index of correlation data of Sample video collection of drama, f
(w, x) is linear regression model (LRM), andWherein k is characterized the number of data, xiFor in k characteristic
Ith feature data；wiFor the parameter of ith feature data, L is regular terms；
The training module can be also used for determining the parameters value in linear regression model (LRM), determine in linear regression model (LRM)
The process of parameters value be specially：
By the known flow index of correlation data of Sample video collection of drama and k characteristic of Sample video collection of drama
It substitutes into current objective function to be calculated, obtains error amount；
Whether error in judgement value is less than preset error amount threshold value obtains if it is not, then adjusting the parameter value of k characteristic
Objective function after to change, and return the known flow index of correlation data and Sample video of Sample video collection of drama
K characteristic of collection of drama substitutes into the step of current objective function is calculated, obtains error amount；
If error amount is less than preset error amount threshold value, the line that the current parameter value of characteristic is completed as training
The parameter value of the characteristic of property regression model.
In embodiments of the present invention, prediction module specifically can be used for, by characteristic xiInput training is completed linear
Regression model f (w, x) obtains the flow index of correlation predicted value of video collection of drama to be predicted.
Since the data used in training pattern are after logarithmic transformation, correspondingly, obtaining linear regression mould
After the result of type f (w, x) output, exponential transform can be carried out to the result, using the result after progress exponential transform as to pre-
Survey the flow index of correlation predicted value of video collection of drama.
The embodiment of the invention discloses a kind of electronic equipment, as shown in Figure 3.Including processor 301, communication interface 302, deposit
Reservoir 303 and communication bus 304, wherein processor 301, communication interface 302, memory 303 are completed by communication bus 304
Mutual communication,
Memory 303, for storing computer program；
Processor 301 when for executing the program stored on memory 303, realizes any of the above-described method and step.
The communication bus that above-mentioned electronic equipment is mentioned can be Peripheral Component Interconnect standard (Peripheral Component
Interconnect, PCI) bus or expanding the industrial standard structure (Extended Industry Standard
Architecture, EISA) bus etc..The communication bus can be divided into address bus, data/address bus, control bus etc..For just
It is only indicated with a thick line in expression, figure, it is not intended that an only bus or a type of bus.
Communication interface is for the communication between above-mentioned electronic equipment and other equipment.
Memory may include random access memory (Random Access Memory, RAM), also may include non-easy
The property lost memory (Non-Volatile Memory, NVM), for example, at least a magnetic disk storage.Optionally, memory may be used also
To be storage device that at least one is located remotely from aforementioned processor.
Above-mentioned processor can be general processor, including central processing unit (Central Processing Unit,
CPU), network processing unit (Network Processor, NP) etc.；It can also be digital signal processor (Digital Signal
Processing, DSP), it is specific integrated circuit (Application Specific Integrated Circuit, ASIC), existing
It is field programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic device, discrete
Door or transistor logic, discrete hardware components.
The embodiment of the invention also provides a kind of computer readable storage medium, it is stored in computer readable storage medium
Computer program realizes any of the above-described method and step when computer program is executed by processor.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.Computer program product
Including one or more computer instructions.When loading on computers and executing computer program instructions, all or part of real estate
Raw process or function according to the embodiment of the present invention.Computer can be general purpose computer, special purpose computer, computer network,
Or other programmable devices.Computer instruction may be stored in a computer readable storage medium, or from a computer
Readable storage medium storing program for executing to another computer readable storage medium transmit, for example, computer instruction can from a web-site,
Computer, server or data center by wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (such as
Infrared, wireless, microwave etc.) mode transmitted to another web-site, computer, server or data center.Computer
Readable storage medium storing program for executing can be any usable medium or include one or more usable medium collection that computer can access
At the data storage devices such as server, data center.Usable medium can be magnetic medium, (for example, floppy disk, hard disk, magnetic
Band), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk Solid State Disk (SSD)) etc..
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any actual relationship or order or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to
Non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including element.
Each embodiment in this specification is all made of relevant mode and describes, same and similar portion between each embodiment
Dividing may refer to each other, and each embodiment focuses on the differences from other embodiments.Especially for device,
For electronic equipment, computer readable storage medium and computer program product embodiments, since it is substantially similar to method
Embodiment, so being described relatively simple, the relevent part can refer to the partial explaination of embodiments of method.
The above is merely preferred embodiments of the present invention, it is not intended to limit the scope of the present invention.It is all in this hair
Any modification, equivalent replacement, improvement and so within bright spirit and principle, are included within the scope of protection of the present invention.
Claims (11)
1. a kind of flow index of correlation prediction technique of video collection of drama, which is characterized in that the method includes：
The information data of video collection of drama to be predicted is obtained, the information data includes：The row of the video collection of drama to be predicted broadcasts number
Public praise number relevant to the video collection of drama to be predicted in preceding first preset time period is played according to, the video collection of drama to be predicted
According to and the video collection of drama to be predicted play that preceding second preset time period is interior relevant to the video collection of drama to be predicted have been broadcast
Put the flow index of correlation data of video collection of drama；
Feature extraction is carried out to the information data, obtains data to be converted, and by the data to be converted according to preset data
Format is converted, the data after obtaining format conversion, the characteristic as the video collection of drama to be predicted；
By the characteristic input prediction model, the flow index of correlation predicted value of the video collection of drama to be predicted, institute are obtained
State prediction model be according to the characteristic of the preset data form of Sample video collection of drama and the Sample video collection of drama
What the flow index of correlation data training known obtained, the Sample video collection of drama is played video collection of drama.
2. the method according to claim 1, wherein it is described by the data to be converted according to preset data form
It is converted, the data after obtaining format conversion, as the characteristic of the video collection of drama to be predicted, including：
The video collection of drama to be predicted is played into the interior public praise relevant to the video collection of drama to be predicted of preceding first preset time period
The data to be converted of data and the video collection of drama to be predicted play in preceding second preset time period and the video to be predicted
The data to be converted of the flow index of correlation data of the relevant played video collection of drama of collection of drama carry out logarithmic transformation；
By the data obtained after the logarithmic transformation and the number to be converted of row's multicast data of the video collection of drama to be predicted
According to being normalized, the characteristic of the preset data form of the video collection of drama to be predicted is obtained.
3. the method according to claim 1, wherein the prediction model is linear regression model (LRM).
4. according to the method described in claim 3, it is characterized in that, based on the following objective function training linear regression mould
Type：
Wherein, n is the number of the Sample video collection of drama, and y is the known flow index of correlation number of the Sample video collection of drama
According to, f (w, x) is the linear regression model (LRM), andWherein k is characterized the number of data, xiFor k spy
Levy ith feature data in data；wiFor the parameter of ith feature data, L is regular terms；
Determine the process of the parameter value of the characteristic of the linear regression model (LRM), including：
By the known flow index of correlation data of the Sample video collection of drama and the Sample video collection of dramakA characteristic
It is calculated according to the objective function for substituting into current, obtains error amount；
Judge whether the error amount is less than preset error amount threshold value, if it is not, then adjusting the parameter of the k characteristic
Value, the objective function after being changed, and return the known flow index of correlation data of the Sample video collection of drama, and
K characteristic of the Sample video collection of drama substitutes into the step of current objective function is calculated, obtains error amount；
If the error amount is less than preset error amount threshold value, the institute that the current parameter value of characteristic is completed as training
State the parameter value of the characteristic of linear regression model (LRM).
5. according to the method described in claim 4, obtaining it is characterized in that, described by the characteristic input prediction model
The flow index of correlation predicted value of the video collection of drama to be predicted, including：
By the characteristic xiThe linear regression model (LRM) f (w, x) that input training is completed, obtains the video collection of drama to be predicted
Flow index of correlation predicted value.
6. a kind of flow index of correlation prediction meanss of video collection of drama, which is characterized in that described device includes：
Information data obtains module, and for obtaining the information data of video collection of drama to be predicted, the information data includes：It is described to
Predict that row's multicast data of video collection of drama, the video collection of drama to be predicted plays in preceding first preset time period and the view to be predicted
The relevant public praise data of frequency collection of drama and the video collection of drama to be predicted play preceding second preset time period it is interior with it is described to be predicted
The flow index of correlation data of the relevant played video collection of drama of video collection of drama；
Characteristic extracting module obtains data to be converted, and will be described to be converted for carrying out feature extraction to the information data
Data are converted according to preset data form, the data after obtaining format conversion, the spy as the video collection of drama to be predicted
Levy data；
Prediction module, for by the characteristic input prediction model, the flow for obtaining the video collection of drama to be predicted to be related
Index predicted value, the prediction model are the characteristic and the sample according to the preset data form of Sample video collection of drama
What the known flow index of correlation data training of this video collection of drama obtained, the Sample video collection of drama is that played video is acute
Collection.
7. device according to claim 6, which is characterized in that the characteristic extracting module is specifically used for：
The video collection of drama to be predicted is played into the interior public praise relevant to the video collection of drama to be predicted of preceding first preset time period
The data to be converted of data and the video collection of drama to be predicted play in preceding second preset time period and the video to be predicted
The data to be converted of the flow index of correlation data of the relevant played video collection of drama of collection of drama carry out logarithmic transformation；
By the data obtained after the logarithmic transformation and the number to be converted of row's multicast data of the video collection of drama to be predicted
According to being normalized, the characteristic of the preset data form of the video collection of drama to be predicted is obtained.
8. device according to claim 6, which is characterized in that the prediction model is linear regression model (LRM).
9. device according to claim 8, which is characterized in that described device further includes：
Training module, for based on the following objective function training linear regression model (LRM)：
Wherein, n is the number of the Sample video collection of drama, and y is the known flow index of correlation number of the Sample video collection of drama
According to, f (w, x) is the linear regression model (LRM), andWherein k is characterized the number of data, xiFor k feature
Ith feature data in data；wiFor the parameter of ith feature data, L is regular terms；
The training module is specifically used for, by known flow index of correlation data of the Sample video collection of drama and described
K characteristic of Sample video collection of drama substitutes into current objective function and is calculated, and obtains error amount；
Judge whether the error amount is less than preset error amount threshold value, if it is not, then adjusting the parameter of the k characteristic
Value, the objective function after being changed, and return the known flow index of correlation data of the Sample video collection of drama, and
K characteristic of the Sample video collection of drama substitutes into the step of current objective function is calculated, obtains error amount；
If the error amount is less than preset error amount threshold value, the institute that the current parameter value of characteristic is completed as training
State the parameter value of the characteristic of linear regression model (LRM).
10. device according to claim 9, which is characterized in that the prediction module is specifically used for：
By the characteristic xiThe linear regression model (LRM) f (w, x) that input training is completed, obtains the video collection of drama to be predicted
Flow index of correlation predicted value.
11. a kind of electronic equipment, which is characterized in that including processor, communication interface, memory and communication bus, wherein described
Processor, the communication interface, the memory complete mutual communication by the communication bus；
The memory, for storing computer program；
The processor when for executing the program stored on the memory, realizes any side claim 1-5
Method step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810533410.0A CN108898415A (en) | 2018-05-29 | 2018-05-29 | A kind of the flow index of correlation prediction technique and device of video collection of drama |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810533410.0A CN108898415A (en) | 2018-05-29 | 2018-05-29 | A kind of the flow index of correlation prediction technique and device of video collection of drama |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108898415A true CN108898415A (en) | 2018-11-27 |
Family
ID=64343346
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810533410.0A Pending CN108898415A (en) | 2018-05-29 | 2018-05-29 | A kind of the flow index of correlation prediction technique and device of video collection of drama |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108898415A (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109451352A (en) * | 2018-12-11 | 2019-03-08 | 北京奇艺世纪科技有限公司 | A kind of video playing method for predicting and device |
| CN109587515A (en) * | 2018-12-11 | 2019-04-05 | 北京奇艺世纪科技有限公司 | A kind of video playing method for predicting and device |
| CN109963174A (en) * | 2019-01-28 | 2019-07-02 | 北京奇艺世纪科技有限公司 | Flow index of correlation predictor method, device and computer readable storage medium |
| CN110225407A (en) * | 2019-04-23 | 2019-09-10 | 北京奇艺世纪科技有限公司 | The acquisition methods and device and computer readable storage medium of a kind of played data |
| CN110830809A (en) * | 2019-11-20 | 2020-02-21 | 咪咕动漫有限公司 | Video content heat determination method, electronic device and storage medium |
| CN110933492A (en) * | 2019-12-10 | 2020-03-27 | 北京爱奇艺科技有限公司 | Method and device for predicting playing time |
| CN111062527A (en) * | 2019-12-10 | 2020-04-24 | 北京爱奇艺科技有限公司 | Video collection flow prediction method and device |
| CN111597383A (en) * | 2020-04-24 | 2020-08-28 | 北京奇艺世纪科技有限公司 | Video heat level prediction method and device |
| CN112511901A (en) * | 2020-12-07 | 2021-03-16 | 北京秒针人工智能科技有限公司 | Method and system for predicting comprehensive drama playing amount, computer device and storage medium |
| CN113052375A (en) * | 2021-03-19 | 2021-06-29 | 上海森宇文化传媒股份有限公司 | Method and device for predicting play volume of episode |
| CN113313511A (en) * | 2021-04-30 | 2021-08-27 | 北京奇艺世纪科技有限公司 | Video traffic prediction method, device, electronic equipment and medium |
| CN113347464A (en) * | 2021-04-30 | 2021-09-03 | 北京奇艺世纪科技有限公司 | Video playing flow prediction method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104035994A (en) * | 2014-06-11 | 2014-09-10 | 华东师范大学 | Prediction method of television play on-demand amount based on network data |
| CN105635762A (en) * | 2016-01-15 | 2016-06-01 | 深圳大学 | Video heat prediction method based on deep belief networks and system thereof |
| CN107133699A (en) * | 2017-05-12 | 2017-09-05 | 百度国际科技（深圳）有限公司 | Box office receipts trend forecasting method and device, equipment, storage medium |
| CN107547154A (en) * | 2016-06-23 | 2018-01-05 | 华为技术有限公司 | A kind of method and device for establishing video traffic prediction model |
| CN107707964A (en) * | 2016-08-08 | 2018-02-16 | 华为软件技术有限公司 | The method and apparatus for predicting video content temperature |
-
2018
- 2018-05-29 CN CN201810533410.0A patent/CN108898415A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104035994A (en) * | 2014-06-11 | 2014-09-10 | 华东师范大学 | Prediction method of television play on-demand amount based on network data |
| CN105635762A (en) * | 2016-01-15 | 2016-06-01 | 深圳大学 | Video heat prediction method based on deep belief networks and system thereof |
| CN107547154A (en) * | 2016-06-23 | 2018-01-05 | 华为技术有限公司 | A kind of method and device for establishing video traffic prediction model |
| CN107707964A (en) * | 2016-08-08 | 2018-02-16 | 华为软件技术有限公司 | The method and apparatus for predicting video content temperature |
| CN107133699A (en) * | 2017-05-12 | 2017-09-05 | 百度国际科技（深圳）有限公司 | Box office receipts trend forecasting method and device, equipment, storage medium |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109587515A (en) * | 2018-12-11 | 2019-04-05 | 北京奇艺世纪科技有限公司 | A kind of video playing method for predicting and device |
| CN109451352A (en) * | 2018-12-11 | 2019-03-08 | 北京奇艺世纪科技有限公司 | A kind of video playing method for predicting and device |
| CN109963174B (en) * | 2019-01-28 | 2021-06-22 | 北京奇艺世纪科技有限公司 | Flow related index estimation method and device and computer readable storage medium |
| CN109963174A (en) * | 2019-01-28 | 2019-07-02 | 北京奇艺世纪科技有限公司 | Flow index of correlation predictor method, device and computer readable storage medium |
| CN110225407A (en) * | 2019-04-23 | 2019-09-10 | 北京奇艺世纪科技有限公司 | The acquisition methods and device and computer readable storage medium of a kind of played data |
| CN110225407B (en) * | 2019-04-23 | 2021-09-03 | 北京奇艺世纪科技有限公司 | Method and device for acquiring playing data and computer readable storage medium |
| CN110830809A (en) * | 2019-11-20 | 2020-02-21 | 咪咕动漫有限公司 | Video content heat determination method, electronic device and storage medium |
| CN111062527A (en) * | 2019-12-10 | 2020-04-24 | 北京爱奇艺科技有限公司 | Video collection flow prediction method and device |
| CN110933492A (en) * | 2019-12-10 | 2020-03-27 | 北京爱奇艺科技有限公司 | Method and device for predicting playing time |
| CN111062527B (en) * | 2019-12-10 | 2023-12-05 | 北京爱奇艺科技有限公司 | Video traffic collection prediction method and device |
| CN111597383A (en) * | 2020-04-24 | 2020-08-28 | 北京奇艺世纪科技有限公司 | Video heat level prediction method and device |
| CN112511901A (en) * | 2020-12-07 | 2021-03-16 | 北京秒针人工智能科技有限公司 | Method and system for predicting comprehensive drama playing amount, computer device and storage medium |
| CN113052375A (en) * | 2021-03-19 | 2021-06-29 | 上海森宇文化传媒股份有限公司 | Method and device for predicting play volume of episode |
| CN113313511A (en) * | 2021-04-30 | 2021-08-27 | 北京奇艺世纪科技有限公司 | Video traffic prediction method, device, electronic equipment and medium |
| CN113347464A (en) * | 2021-04-30 | 2021-09-03 | 北京奇艺世纪科技有限公司 | Video playing flow prediction method and device, electronic equipment and storage medium |
| CN113347464B (en) * | 2021-04-30 | 2022-09-30 | 北京奇艺世纪科技有限公司 | Video playing flow prediction method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108898415A (en) | A kind of the flow index of correlation prediction technique and device of video collection of drama | |
| CN103686237B (en) | Recommend the method and system of video resource | |
| CN105045831B (en) | A kind of information push method and device | |
| CN112364202B (en) | Video recommendation method and device and electronic equipment | |
| CN108363730B (en) | Content recommendation method, system and terminal equipment | |
| CN110222975A (en) | A kind of loss customer analysis method, apparatus, electronic equipment and storage medium | |
| CN111125429B (en) | Video pushing method, device and computer readable storage medium | |
| CN110324662A (en) | A kind of video cover generation method and device | |
| CN106294830A (en) | The recommendation method and device of multimedia resource | |
| CN105989074A (en) | Method and device for recommending cold start through mobile equipment information | |
| CN109508405B (en) | Method and device for determining recommended video, electronic equipment and storage medium | |
| CN111062527B (en) | Video traffic collection prediction method and device | |
| WO2021082603A1 (en) | Video pushing method and apparatus, computer device, and storage medium | |
| CN109587515B (en) | Video playing flow prediction method and device | |
| CN108197336B (en) | Video searching method and device | |
| CN108304399A (en) | The recommendation method and device of Web content | |
| CN107563804A (en) | A kind of method and apparatus for predicting the number of users covered under advertisement stereotactic conditions | |
| CN112765400B (en) | Weight updating method, content recommending method, device and equipment for interest labels | |
| CN111767430B (en) | Video resource pushing method, video resource pushing device and storage medium | |
| CN109511015A (en) | Multimedia resource recommended method, device, storage medium and equipment | |
| WO2019109592A1 (en) | Smart video recommendation method and system | |
| CN102486796A (en) | Method and device for acquiring video information | |
| CN113836348A (en) | Video-based classification label mark processing method, device, terminal and medium | |
| CN109740621B (en) | Video classification method, device and equipment | |
| CN110121088A (en) | A kind of customer attribute information determines method, apparatus and electronic equipment |
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: 20181127 |
|
| RJ01 | Rejection of invention patent application after publication |