CN107018408A - The Quality of experience appraisal procedure of mobile terminal HTTP video flowings - Google Patents
The Quality of experience appraisal procedure of mobile terminal HTTP video flowings Download PDFInfo
- Publication number
- CN107018408A CN107018408A CN201710037562.7A CN201710037562A CN107018408A CN 107018408 A CN107018408 A CN 107018408A CN 201710037562 A CN201710037562 A CN 201710037562A CN 107018408 A CN107018408 A CN 107018408A
- Authority
- CN
- China
- Prior art keywords
- video
- quality
- parameter
- experience
- index
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
- H04N17/004—Diagnosis, testing or measuring for television systems or their details for digital television systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/472—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content
- H04N21/47217—End-user interface for requesting content, additional data or services; End-user interface for interacting with content, e.g. for content reservation or setting reminders, for requesting event notification, for manipulating displayed content for controlling playback functions for recorded or on-demand content, e.g. using progress bars, mode or play-point indicators or bookmarks
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of Quality of experience appraisal procedure of mobile terminal HTTP video flowings, including video flowing Quality of experience is set to assess parameter value index；Picking video experience scoring participant and video；Video tastes scoring participant viewing video simultaneously scores；Data when watching user video carry out data filtering；Set up using linear regression analysis and predict the data model that the service quality of video scores by assessing parameter index；Score in predicting is carried out to the Quality of experience of mobile terminal HTTP video flowings with obtained data model.The inventive method sets the factor of influence quality of experience of video in advance, and obtain scoring and the video total score of influence factor, and the scientific and reasonable corresponding relation model set up between the two, so as to obtain video quality assessment experience model, therefore the inventive method can accurate feedback user for video flowing Quality of experience, and method is simple and reliable, calculating speed is fast.
Description
Technical field
Present invention relates particularly to a kind of Quality of experience appraisal procedure of mobile terminal HTTP video flowings.
Background technology
The popularization of smart mobile phone and tablet personal computer already leads to produce substantial amounts of flow consumption in wired and wireless network,
And cause the tide of a large amount of flows of user's generation.Mobile data flow will be with 61% compound annual growth rate increase, to 2018
The flow of every month in year is up to 15.9 Chinese mugwort bytes.These ratios increased are made up of mobile video.In movement in 2016
The flow that video is produced accounts for the 53% of mobile data, and such trend will be up to 69% in 2018.These flows include each
The Video service of various kinds is planted, for example：Progressive download, live video stream, the stream service based on http protocol and mutual mobile phone.
Current smart mobile phone, which has become most typical mobile device, to be used to access Internet.Nearest prediction shows
Show, untill 2016, the people of a quarter in the world will access most popular service using smart mobile phone, for example：
YouTube, Facebook, What'sApp etc..According to the global mobile data volume forecasting of Cisco (Cisco), to intelligence in 2019
The flow that mobile phone is produced will occupy 3/4ths of mobile data flow.According to this trend, crest network (the one of wireless network
Kind) how operator will determine network and how to manage the flow of user more so as to attract that they access to understanding
New user is more and more interested.In this case, the concept of Quality of experience (QoE) has as management crest network quality very much
Main instruct specification.
Currently, the user of mobile network can enjoy many multimedia services in Internet.Multimedia service needs
The network of high speed data transfer.In fact, there is provided a high speed number for 3G, 4G or other Internet Service Providers
It is not a difficult thing according to the service of rate.It is technical to have limited not based on present technology (as HSDPA, LTE-A)
With consideration.In fact, any Internet Service Provider can provide a high-speed service to any one user.Cause
This, in order to increase and maintain the quantity of user, service provider starts progressively to be concerned about the Quality of experience (QoE) of user：Experience matter
Amount (QoE) be user in certain environment to it is used service or business the overall degree of recognition.So far, it there is no
It is a kind of can the method for the Quality of experience of video flowing that is provided for service provider of accurate feedback user so that video
The service provider of stream can not obtain the satisfaction of user by obtaining the qos parameter of user, so as to can not improve
The satisfaction of user.
The content of the invention
It is an object of the invention to provide it is a kind of can accurate feedback user for the Quality of experience of video flowing mobile terminal
The Quality of experience appraisal procedure of HTTP video flowings.
The Quality of experience appraisal procedure for this mobile terminal HTTP video flowings that the present invention is provided, comprises the following steps：
S1., the assessment parameter value index of video flowing Quality of experience is set；
S2. select for the video tastes scoring participant of modeling and video；
S3. the video tastes scoring participant for allowing step S2 to select watches corresponding video, and obtains video tastes scoring
Assessment parameter value index of the participant when watching video, and finish in user's viewing video or allow user couple when closing video
The service quality of the video is scored；
S4. the scoring of parameter value index and user to Video service quality of assessing obtained to step S3 carries out data filter
Ripple；
S5. the data obtained using step S4, parameter index and Video service matter will be assessed using linear regression analysis
Amount scoring is associated, so as to obtain predicting the data model that the service quality of video scores by assessing parameter index；
S6. the data model obtained with step S5 carries out score in predicting to the Quality of experience of mobile terminal HTTP video flowings.
The assessment parameter value index of video flowing Quality of experience described in step S1, specifically includes total duration, the video of video
Currently playing duration, video buffer number of times, the total duration of video buffer, the initial buffering time of video, video pause time
Number, the total duration of video pause, the number of times for dragging video and the aggregate scheduling for dragging video.
Selecting described in step S2 is scored participant for the video tastes of modeling, and the principle selected is：From all ages and classes,
It is random in different sexes, the user group of different educational background to select 4~40 people as video tastes scoring participant.
Assessment parameter value index of the acquisition video tastes scoring participant when watching video described in step S3, be specially
Parameter index is assessed using following Rule：
A. the total duration of video and the currently playing duration of video are directly obtained；
B. in video playback, addition buffering monitors variable, occurs buffering situation when video playback and then buffers monitoring variable
Automatically accumulate once, so as to obtain the buffering number of times of video；(for example, in MediaPlayer classes in android development languages
It can be added by setOnBufferingUpdateListener (new OnBufferingUpdateListener ()) method
Plus buffering is monitored, when occurring buffered event every time, system is called in OnBufferingUpdateListener interfaces
OnBufferingUpdate () method carries out processing buffered event.Buffering monitors variable can be in onBufferingUpdate
From increasing 1 in () method.Also there is corresponding method in ios development languages)
C. in video playback, addition buffering monitors variable, automatic record buffer when there is buffering situation when video playback
The system time T1 of beginning, while the system time T2 at the end of recording video buffer, then buffer duration Th=T2-T1 every time；
Each buffer time T is added up simultaneously, so as to obtain the buffering total duration of screen；
D. system time T3 when acquisition video is ready, while obtaining system time when user's click video starts
T4, then video initial buffering time Tch=T3-T4；
E. the number of times that user clicks on " pause " function key in video playback is obtained, so as to obtain the number of times of video pause；
F. system time T5 when user clicks on " pause " function key in video playback is obtained, while obtaining user temporary
Stop replaying the system time T6 of video after video, so as to obtain user as time duration T z=T6-T5 of pause video；
And to user every time fix tentatively video duration T z added up so that obtain user suspend total duration；
G. obtain for clicking on the number of times that progress bar drags video in video playback, so as to obtain the dragging video of user
Number of times；
H. the initial video play position information (time of such as video playback) during user's dragging video is obtained, and is obtained
Video playback position (time of such as video playback) when user stops dragging and normally watches video, is dragged so as to obtain user
The aggregate scheduling of dynamic video.
Allow user to score the service quality of the video described in step S3, specially built using International Telecommunication Union
The Mean Opinion Score value of view, is divided into 5 levels, the i.e. span of Mean Opinion Score value for integer value by the Quality of experience of video
1~5, and each integer value represent respectively the value of Quality of experience as：Bad、Poor、Fair、Good、Excellent.
Being filtered to data described in step S4, specially carries out data filtering using following steps：
(1) parameter for assessing a certain item parameter index carries out data cleansing；
Specifically include：The replacement of default parameters and the smooth treatment of noise data.
If the parameter that a certain item assesses parameter index has missing, using being averaged for this all non-default parameters of parameter kind
Number or median are replaced to the parameter of the missing；
The noise data assessed simultaneously using branch mailbox method a certain item in parameter index carries out smooth treatment：
A. a certain item is assessed into N number of measured value x in parameter index1,x2,...,xNSort from small to large, then after sorting
Sequence is x1',x'2,...,x'N；
B. N number of observation is dispensed using m chest；
C. using etc. by the way of frequency by N number of observation x after sequence1',x'2,...,x'NIt is transferred in m chest, then each
Have in chestIndividual observation；
D. in calculating in each case all observations average, then calculate respectively each observation in the case with it is flat
The Euclidean distance of average, being averaged for the case is replaced with by the maximum observation of Euclidean distance values (being noise data)
Value.(method of i.e. smooth noise data)
(2) system is arrived into all assessment parameter index standardization after data cleansing according to min-max normalization method
One interval：
If the n observation that a certain item assesses parameter index is v1,v2,...,vn, and minAAnd maxAAssess and join for this
The minimum value and maximum of number index, then calculate the assessment parameter index value v after standardization using equation belowi'：
V in formulaiFor the assessment parameter index value before standardization, new_maxAFor the assessment parameter index value area after standardization
Between maximum, new_minAFor the minimum value for the assessment parameter index value interval after standardization；
(3) greedy algorithm selected using attribute set carries out data drop to all assessment parameter indexs after standardization
Dimension.
Use linear regression analysis described in step S5 obtains predicting the service quality of video by assessing parameter index
The data model of scoring, is specially calculated using following steps：
Each QoS parameter value is passed through unitary least square polynomial regression by I, with service quality scoring respectively
Method Modeling, so as to be utilized the data model of single QoS parameter value prediction service quality scoring；
II, repeat steps S2~step S4, the new assessment parameter value index of acquisition and Video service quality scoring data, it is right
The data model that step I is obtained is modified；
All QoS parameters and service quality scoring are passed through multiple linear regression analysis Method Modeling by III, from
And it is utilized the data model of all QoS parameter value prediction service quality scorings；
IV, repeat steps S2~step S4, the new assessment parameter value index of acquisition and Video service quality scoring data, it is right
The data model that step III is obtained is modified.
Pass through unitary least square polynomial regression Method Modeling, the specially progress of use following steps described in step I
Modeling：
I represents a certain assessment parameter value index determined with x, uses xkRepresent the kth time that this kind assesses parameter value index
Measured value；Corresponding Video service quality score is represented with y, y is usedkRepresent the measured value of the kth time of Video service quality score；
Obtain common m data sample (x simultaneouslyk, yk)；
II parameters a0,a1,..,an, wherein n<M so that multinomial p (x)=a0+a1x+...+anxnMeetValue it is minimum；Then n unitary least square fitting multinomial is called p (x)
=a0+a1x+...+anxn；Wherein, n>=0, n<M, parameter a0,a1,...,anFor unknown parameter；
III is drawn on unknown parameter a0,a1,...,anMatrix form normal equation group be Y=XA, whereinThe matrix is n+1 rows, 1 row；
A=[a0 a1 ... an]T, the matrix is n+1 rows, 1 row；
For n+1 rank matrixes；
The equation group of IV solution procedure III, obtains a0,a1,...,an, that is, obtain forecast model.
Described in step III by multiple linear regression analysis Method Modeling, be specially modeled using following steps：
(i) x1,x2,...,xmCorresponding to the assessment parameter value index of all kinds, corresponding Video service is represented with y
Quality score, (xi1,xi2,...,xim,Yi)T(i=1,2 ..., n) it is (x1,x2,...,xm,Y)TN observation；
(ii) assumes Y and x1,x2,...,xmBetween wired sexual intercourse, then multiple linear regression model be：Y=β0+β1x1
+...+βmxm+ ε, wherein, β0,β1,...,βmIt is unknown parameter, m>1, m is the number of independent variable；Claim x1,x2,...,xmTo return
Return variable, β0,β1,...,βmFor regression coefficient, ε is residual error, ε~N (0, σ2) and σ2It is unknown；
(iii) is due to (xi1,xi2,...,xim,Yi)T(i=1,2 ..., n) it is (x1,x2,...,xm,Y)TN observation,
And meet relation Yi=β0+β1xi1+β2xi2+...+βmxim+εi, (i=1 ... n)
(iv) solves formula with the differential methodWherein
xi0=1, i=1,2 ..., n, so that the matrix form for obtaining normal equation group is：Wherein, Y=
(Y1,Y2,...,Yn)T, β=(β0,β1,...,βm)T, ε=(ε0,ε1,...,εn)T,
(v) tries to achieve step (iv) the equation is obtainedAnd willBring into model can obtain it is final
Forecast model.
The Quality of experience appraisal procedure for this mobile terminal HTTP video flowings that the present invention is provided, by obtaining experiment number in advance
According to being modeled, and the Quality of experience assessment models of mobile terminal HTTP video flowings are obtained to the mode that model is modified, so that
Quality of experience assessment can be carried out to mobile terminal HTTP video flowings；The inventive method set in advance influence quality of experience of video because
Element, and scoring and the video total score of influence factor, and the scientific and reasonable corresponding relation model set up between the two are obtained, from
And obtain video quality assessment experience model, therefore the inventive method can accurate feedback user for video flowing experience matter
Amount, and method is simple and reliable, and calculating speed is fast.
Brief description of the drawings
Fig. 1 is the step schematic diagram of the inventive method.
Fig. 2 is the schematic flow sheet of the inventive method.
Embodiment
It is the step schematic diagram of the inventive method as shown in Figure 1：User watches video by mobile device, during viewing video
Data by network transmission to multimedia server, server storage data are simultaneously saved in database, and database then stores this
Assessment parameter value and experience score data required for inventive method.
It is illustrated in figure 2 the schematic flow sheet of the inventive method：This mobile terminal HTTP video flowings that the present invention is provided
Quality of experience appraisal procedure, comprises the following steps：
S1., the assessment parameter value index of video flowing Quality of experience is set, specifically include the total duration of video, video it is current
Playing duration, video buffer number of times, the total duration of video buffer, the initial buffering time of video, the number of times of video pause, video
The total duration of pause, the number of times for dragging video and the aggregate scheduling for dragging video；
S2. select for the video tastes scoring participant of modeling and video；When selecting, from all ages and classes, dissimilarity
, it is not random in the user group of different educational background to select 4~40 people as video tastes scoring participant；
S3. the video tastes scoring participant for allowing step S2 to select watches corresponding video, and obtains video tastes scoring
Assessment parameter value index of the participant when watching video, and finish in user's viewing video or allow user couple when closing video
The service quality of the video is scored, and specially assesses parameter index using following Rule：
A. the total duration of video and the currently playing duration of video are directly obtained；
B. in video playback, addition buffering monitors variable, occurs buffering situation when video playback and then buffers monitoring variable
Automatically accumulate once, so as to obtain the buffering number of times of video；(for example, in MediaPlayer classes in android development languages
It can be added by setOnBufferingUpdateListener (new OnBufferingUpdateListener ()) method
Plus buffering is monitored, when occurring buffered event every time, system is called in OnBufferingUpdateListener interfaces
OnBufferingUpdate () method carries out processing buffered event.Buffering monitors variable can be in onBufferingUpdate
From increasing 1 in () method.Also there is corresponding method in ios development languages)
C. in video playback, addition buffering monitors variable, automatic record buffer when there is buffering situation when video playback
The system time T1 of beginning, while the system time T2 at the end of recording video buffer, then buffer duration Th=T2-T1 every time；
Each buffer time T is added up simultaneously, so as to obtain the buffering total duration of screen；
D. system time T3 when acquisition video is ready, while obtaining system time when user's click video starts
T4, then video initial buffering time Tch=T3-T4；
E. the number of times that user clicks on " pause " function key in video playback is obtained, so as to obtain the number of times of video pause；
F. system time T5 when user clicks on " pause " function key in video playback is obtained, while obtaining user temporary
Stop replaying the system time T6 of video after video, so as to obtain user as time duration T z=T6-T5 of pause video；
And to user every time fix tentatively video duration T z added up so that obtain user suspend total duration；
G. obtain for clicking on the number of times that progress bar drags video in video playback, so as to obtain the dragging video of user
Number of times；
H. the initial video play position information (time of such as video playback) during user's dragging video is obtained, and is obtained
Video playback position (time of such as video playback) when user stops dragging and normally watches video, is dragged so as to obtain user
The aggregate scheduling of dynamic video；
Meanwhile, according to the Mean Opinion Score value of international telecommunication union recommendation, the Quality of experience of video is divided into 5 levels,
I.e. the span of Mean Opinion Score value be integer value 1~5, and each integer value represent respectively the value of Quality of experience as：
Bad、Poor、Fair、Good、Excellent；
S4. the scoring of parameter value index and user to Video service quality of assessing obtained to step S3 carries out data filter
Ripple；Specially data filtering is carried out using following steps：
(1) parameter for assessing a certain item parameter index carries out data cleansing；
Specifically include：The replacement of default parameters and the smooth treatment of noise data.
If the parameter that a certain item assesses parameter index has missing, using being averaged for this all non-default parameters of parameter kind
Number or median are replaced to the parameter of the missing；
The noise data assessed simultaneously using branch mailbox method a certain item in parameter index carries out smooth treatment：
A. a certain item is assessed into N number of measured value x in parameter index1,x2,...,xNSort from small to large, then after sorting
Sequence is x1',x'2,...,x'N；
B. N number of observation is dispensed using m chest；
C. using etc. by the way of frequency by N number of observation x after sequence1',x'2,...,x'NIt is transferred in m chest, then each
Have in chestIndividual observation；
D. in calculating in each case all observations average, then calculate respectively each observation in the case with it is flat
The Euclidean distance of average, being averaged for the case is replaced with by the maximum observation of Euclidean distance values (being noise data)
Value.
(2) system is arrived into all assessment parameter index standardization after data cleansing according to min-max normalization method
One interval：
If the n observation that a certain item assesses parameter index is v1,v2,...,vn, and minAAnd maxAAssess and join for this
The minimum value and maximum of number index, then calculate the assessment parameter index value v after standardization using equation belowi'：
V in formulaiFor the assessment parameter index value before standardization, new_maxAFor the assessment parameter index value area after standardization
Between maximum, new_minAFor the minimum value for the assessment parameter index value interval after standardization；
(3) greedy algorithm selected using attribute set carries out data drop to all assessment parameter indexs after standardization
Dimension；
S5. the data obtained using step S4, parameter index and Video service matter will be assessed using linear regression analysis
Amount scoring is associated, so as to obtain predicting the data model that the service quality of video scores by assessing parameter index；In tool
When body is implemented, it can be calculated using following steps：
Each QoS parameter value is passed through unitary least square polynomial regression by I, with service quality scoring respectively
Method Modeling, so as to be utilized the data model of single QoS parameter value prediction service quality scoring；
II, repeat steps S2~step S4, the new assessment parameter value index of acquisition and Video service quality scoring data, it is right
The data model that step I is obtained is modified；
All QoS parameters and service quality scoring are passed through multiple linear regression analysis Method Modeling by III, from
And it is utilized the data model of all QoS parameter value prediction service quality scorings；
IV, repeat steps S2~step S4, the new assessment parameter value index of acquisition and Video service quality scoring data, it is right
The data model that step III is obtained is modified.
Pass through unitary least square polynomial regression Method Modeling, the specially progress of use following steps described in step I
Modeling：
I represents a certain assessment parameter value index determined with x, uses xkRepresent the kth time that this kind assesses parameter value index
Measured value；Corresponding Video service quality score is represented with y, y is usedkRepresent the measured value of the kth time of Video service quality score；
Obtain common m data sample (x simultaneouslyk, yk)；
II parameters a0,a1,..,an, wherein n<M so that multinomial p (x)=a0+a1x+...+anxnMeetValue it is minimum；Then n unitary least square fitting multinomial is called p (x)
=a0+a1x+...+anxn；Wherein, n>=0, n<M, parameter a0,a1,...,anFor unknown parameter；
III is drawn on unknown parameter a0,a1,...,anMatrix form normal equation group be Y=XA, whereinThe matrix is n+1 rows, 1 row；
A=[a0 a1 ... an]T, the matrix is n+1 rows, 1 row；
For n+1 rank matrixes；
The equation group of IV solution procedure III, obtains a0,a1,...,an, that is, obtain forecast model.
Described in step III by multiple linear regression analysis Method Modeling, be specially modeled using following steps：
(i) x1,x2,...,xmCorresponding to the assessment parameter value index of all kinds, corresponding Video service is represented with y
Quality score, (xi1,xi2,...,xim,Yi)T(i=1,2 ..., n) it is (x1,x2,...,xm,Y)TN observation；
(ii) assumes Y and x1,x2,...,xmBetween wired sexual intercourse, then multiple linear regression model be：Y=β0+β1x1
+...+βmxm+ ε, wherein, β0,β1,...,βmIt is unknown parameter, m>1, m is the number of independent variable；Claim x1,x2,...,xmTo return
Return variable, β0,β1,...,βmFor regression coefficient, ε is residual error, ε~N (0, σ2) and σ2It is unknown；
(iii) is due to (xi1,xi2,...,xim,Yi)T(i=1,2 ..., n) it is (x1,x2,...,xm,Y)TN observation,
And meet relation Yi=β0+β1xi1+β2xi2+...+βmxim+εi, (i=1 ... n)
(iv) solves formula with the differential methodWherein
xi0=1, i=1,2 ..., n, so that the matrix form for obtaining normal equation group is：Wherein, Y=
(Y1,Y2,...,Yn)T, β=(β0,β1,...,βm)T, ε=(ε0,ε1,...,εn)T,
(v) tries to achieve step (iv) the equation is obtainedAnd willBring into model can obtain it is final
Forecast model.
S6. the data model obtained with step S5 carries out score in predicting to the Quality of experience of mobile terminal HTTP video flowings.
Patent of the present invention obtains the support of state natural sciences fund (bullets 61672221,61273232).
Claims (9)
1. a kind of Quality of experience appraisal procedure of mobile terminal HTTP video flowings, comprises the following steps：
S1., the assessment parameter value index of video flowing Quality of experience is set；
S2. select for the video tastes scoring participant of modeling and video；
S3. the video tastes scoring participant for allowing step S2 to select watches corresponding video, and obtains video tastes scoring participation
Assessment parameter value index of the person when watching video, and allow user to regard this when user's viewing video is finished or closes video
The service quality of frequency is scored；
S4. the scoring of parameter value index and user to Video service quality of assessing obtained to step S3 carries out data filtering；
S5. the data obtained using step S4, are commented using linear regression analysis by parameter index is assessed with Video service quality
Divide and be associated, so as to obtain predicting the data model that the service quality of video scores by assessing parameter index；
S6. the data model obtained with step S5 carries out score in predicting to the Quality of experience of mobile terminal HTTP video flowings.
2. the Quality of experience appraisal procedure of HTTP video flowings in mobile terminal according to claim 1, it is characterised in that step S1
The assessment parameter value index of described video flowing Quality of experience, specifically include the total duration of video, the currently playing duration of video,
Video buffer number of times, the total duration of video buffer, the initial buffering time of video, the number of times of video pause, video pause it is total
Duration, the number of times for dragging video and the aggregate scheduling for dragging video.
3. the Quality of experience appraisal procedure of HTTP video flowings in mobile terminal according to claim 1, it is characterised in that step S2
Described select is scored participant for the video tastes of modeling, and the principle selected is：From all ages and classes, different sexes, difference
It is random in the user group of education background to select 4~40 people as video tastes scoring participant.
4. the Quality of experience appraisal procedure of the mobile terminal HTTP video flowings according to one of claims 1 to 3, it is characterised in that
Assessment parameter value index of the acquisition video tastes scoring participant when watching video described in step S3, specially using as follows
Rule assesses parameter index：
A. the total duration of video and the currently playing duration of video are directly obtained；
B. in video playback addition buffering monitor variable, occur when video playback buffering situation then buffer monitoring variable it is automatic
Accumulate once, so as to obtain the buffering number of times of video；
C. in video playback, addition buffering monitors variable, and automatic record buffer starts when there is buffering situation when video playback
System time T1, while the system time T2 at the end of recording video buffer, then buffering duration Th=T2-T1 every time；Simultaneously
Each buffer time T is added up, so as to obtain the buffering total duration of screen；
D. system time T3 when acquisition video is ready, while system time T4 when user's click video starts is obtained,
Then video initial buffering time Tch=T3-T4；
E. the number of times that user clicks on " pause " function key in video playback is obtained, so as to obtain the number of times of video pause；
F. system time T5 when user clicks on " pause " function key in video playback is obtained, is regarded while obtaining user in pause
The system time T6 of video is replayed after frequency, so as to obtain user as time duration T z=T6-T5 of pause video；And it is right
The duration T z that user fixes tentatively video every time is added up, so as to obtain the total duration of user's pause；
G. obtain and drag the number of times of video for clicking on progress bar in video playback, so as to obtain time of the dragging video of user
Number；
H. the initial video play position information (time of such as video playback) during user's dragging video is obtained, and obtains user
Video playback position (time of such as video playback) when stopping dragging and normally watching video, so as to obtain user's dragging
The aggregate scheduling of video.
5. the Quality of experience appraisal procedure of HTTP video flowings in mobile terminal according to claim 4, it is characterised in that step S3
Described allows user to score the service quality of the video, specially using the Mean Opinion Score of international telecommunication union recommendation
Value, is divided into 5 levels, the i.e. span of Mean Opinion Score value for integer value 1~5 by the Quality of experience of video, and each
Integer value represent respectively the value of Quality of experience as：Bad、Poor、Fair、Good、Excellent.
6. the Quality of experience appraisal procedure of HTTP video flowings in mobile terminal according to claim 5, it is characterised in that step S4
Described is filtered to data, specially carries out data filtering using following steps：
(1) parameter for assessing a certain item parameter index carries out data cleansing；
Specifically include：The replacement of default parameters and the smooth treatment of noise data；
If the parameter that a certain item assesses parameter index has missing, using all non-default parameters of this parameter kind average or
Median is replaced to the parameter of the missing；
The noise data assessed simultaneously using branch mailbox method a certain item in parameter index carries out smooth treatment：
A. a certain item is assessed into N number of measured value x in parameter index1,x2,...,xNSort from small to large, then the sequence after sorting
For x '1,x'2,...,x'N；
B. N number of observation is dispensed using m chest；
C. using etc. by the way of frequency by N number of observation x ' after sequence1,x'2,...,x'NIt is transferred in m chest, then each chest
In haveIndividual observation；
D. in calculating in each case all observations average, each observation and the average value in the case are then calculated respectively
Euclidean distance, the maximum observation of Euclidean distance values is replaced with to the average value of the case；
(2) unification is arrived into all parameter index standardization of assessing after data cleansing according to min-max normalization method
It is interval：
If the n observation that a certain item assesses parameter index is v1,v2,...,vn, and minAAnd maxAParameter is assessed for this to refer to
Target minimum value and maximum, then calculate the assessment parameter index value v after standardization using equation belowi'：
V in formulaiFor the assessment parameter index value before standardization, new_maxAInterval for the assessment parameter index value after standardization
Maximum, new_minAFor the minimum value for the assessment parameter index value interval after standardization；
(3) greedy algorithm selected using attribute set carries out Data Dimensionality Reduction to all assessment parameter indexs after standardization.
7. the Quality of experience appraisal procedure of HTTP video flowings in mobile terminal according to claim 6, it is characterised in that step S5
Described use linear regression analysis obtains predicting the data model that the service quality of video scores by assessing parameter index,
Specially calculated using following steps：
Each QoS parameter value is passed through unitary least square polynomial regression method by I, with service quality scoring respectively
Modeling, so as to be utilized the data model of single QoS parameter value prediction service quality scoring；
II, repeat steps S2~step S4, the new assessment parameter value index of acquisition and Video service quality scoring data, to step
I obtained data model is modified；
All QoS parameters and service quality scoring are passed through multiple linear regression analysis Method Modeling by III, so that
To the data model that service quality scoring is predicted using all QoS parameter values；
IV, repeat steps S2~step S4, the new assessment parameter value index of acquisition and Video service quality scoring data, to step
III obtained data model is modified.
8. the Quality of experience appraisal procedure of HTTP video flowings in mobile terminal according to claim 7, it is characterised in that step I
It is described by unitary least square polynomial regression Method Modeling, be specially modeled using following steps：
I represents a certain assessment parameter value index determined with x, uses xkRepresent the kth time measurement that this kind assesses parameter value index
Value；Corresponding Video service quality score is represented with y, y is usedkRepresent the measured value of the kth time of Video service quality score；Simultaneously
Obtain common m data sample (xk, yk)；
II parameters a0,a1,..,an, wherein n<M so that multinomial p (x)=a0+a1x+...+anxnMeetValue it is minimum；Then n unitary least square fitting multinomial is called p (x)=a0+
a1x+...+anxn；Wherein, n>=0, n<M, parameter a0,a1,...,anFor unknown parameter；III is drawn on unknown parameter a0,a1,...,
anMatrix form normal equation group be Y=XA, whereinShould
Matrix is n+1 rows, 1 row；
A=[a0 a1 ... an]T, the matrix is n+1 rows, 1 row；
For n+1 rank matrixes；
The equation group of IV solution procedure III, obtains a0,a1,...,an, that is, obtain forecast model.
9. the Quality of experience appraisal procedure of HTTP video flowings in mobile terminal according to claim 8, it is characterised in that step III
It is described by multiple linear regression analysis Method Modeling, be specially modeled using following steps：
(i) x1,x2,...,xmCorresponding to the assessment parameter value index of all kinds, corresponding Video service quality is represented with y
Scoring, (xi1,xi2,...,xim,Yi)T(i=1,2 ..., n) it is (x1,x2,...,xm,Y)TN observation；
(ii) assumes Y and x1,x2,...,xmBetween wired sexual intercourse, then multiple linear regression model be：Y=β0+β1x1+...+βmxm+ ε, wherein, β0,β1,...,βmIt is unknown parameter, m>1, m is the number of independent variable；Claim x1,x2,...,xmBecome to return
Amount, β0,β1,...,βmFor regression coefficient, ε is residual error, ε~N (0, σ2) and σ2It is unknown；
(iii) is due to (xi1,xi2,...,xim,Yi)T(i=1,2 ..., n) it is (x1,x2,...,xm,Y)TN observation, it is and full
Sufficient relation Yi=β0+β1xi1+β2xi2+...+βmxim+εi, (i=1 ... n)；
(iv) solves formula with the differential methodWherein xi0=1,
I=1,2 ..., n, so that the matrix form for obtaining normal equation group is：Wherein, Y=(Y1,
Y2,...,Yn)T, β=(β0,β1,...,βm)T, ε=(ε0,ε1,...,εn)T,
(v) tries to achieve step (iv) the equation is obtainedAnd willBring model into and can obtain final prediction mould
Type.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710037562.7A CN107018408A (en) | 2017-01-19 | 2017-01-19 | The Quality of experience appraisal procedure of mobile terminal HTTP video flowings |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710037562.7A CN107018408A (en) | 2017-01-19 | 2017-01-19 | The Quality of experience appraisal procedure of mobile terminal HTTP video flowings |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107018408A true CN107018408A (en) | 2017-08-04 |
Family
ID=59440087
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710037562.7A Pending CN107018408A (en) | 2017-01-19 | 2017-01-19 | The Quality of experience appraisal procedure of mobile terminal HTTP video flowings |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107018408A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109451300A (en) * | 2018-11-12 | 2019-03-08 | 中国联合网络通信集团有限公司 | The determination method and apparatus of video quality score |
| CN109451303A (en) * | 2018-12-24 | 2019-03-08 | 合肥工业大学 | A kind of modeling method for user experience quality QoE in VR video |
| CN110139160A (en) * | 2019-05-10 | 2019-08-16 | 北京奇艺世纪科技有限公司 | A kind of forecasting system and method |
| CN111401637A (en) * | 2020-03-16 | 2020-07-10 | 湖南大学 | User experience quality prediction method fusing user behavior and expression data |
| CN112822482A (en) * | 2020-12-31 | 2021-05-18 | 上海掌门科技有限公司 | Method and equipment for determining evaluation score of audio and video call |
| CN112862250A (en) * | 2021-01-12 | 2021-05-28 | 浙江知行教育科技有限公司 | College learning evaluation system and method based on big data |
| CN114598934A (en) * | 2022-02-11 | 2022-06-07 | 山东悦知教育科技有限公司 | Education software data processing method based on big data |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103179592A (en) * | 2013-03-20 | 2013-06-26 | 南京邮电大学 | QoE (Quality of Experience) comprehensive evaluation method based on hierarchical tree structure |
| CN104023232A (en) * | 2014-06-27 | 2014-09-03 | 北京邮电大学 | Mobile video quality assessment method based on hierarchy analysis and multiple linear regressions |
| CN104281770A (en) * | 2014-06-30 | 2015-01-14 | 许蔚蔚 | Unary linear regression method |
| CN105049930A (en) * | 2015-08-14 | 2015-11-11 | 浙江大学 | Wireless video streaming service QoE estimation method based on support vector machine |
| JP2017005476A (en) * | 2015-06-09 | 2017-01-05 | 日本電信電話株式会社 | Quality management device, quality management method and program |
-
2017
- 2017-01-19 CN CN201710037562.7A patent/CN107018408A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103179592A (en) * | 2013-03-20 | 2013-06-26 | 南京邮电大学 | QoE (Quality of Experience) comprehensive evaluation method based on hierarchical tree structure |
| CN104023232A (en) * | 2014-06-27 | 2014-09-03 | 北京邮电大学 | Mobile video quality assessment method based on hierarchy analysis and multiple linear regressions |
| CN104281770A (en) * | 2014-06-30 | 2015-01-14 | 许蔚蔚 | Unary linear regression method |
| JP2017005476A (en) * | 2015-06-09 | 2017-01-05 | 日本電信電話株式会社 | Quality management device, quality management method and program |
| CN105049930A (en) * | 2015-08-14 | 2015-11-11 | 浙江大学 | Wireless video streaming service QoE estimation method based on support vector machine |
Non-Patent Citations (1)
| Title |
|---|
| 陈伟宏: "有小凹的视频质量评价模型", 《湖南城市学院学报（自然科学版）》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109451300A (en) * | 2018-11-12 | 2019-03-08 | 中国联合网络通信集团有限公司 | The determination method and apparatus of video quality score |
| CN109451303A (en) * | 2018-12-24 | 2019-03-08 | 合肥工业大学 | A kind of modeling method for user experience quality QoE in VR video |
| CN110139160A (en) * | 2019-05-10 | 2019-08-16 | 北京奇艺世纪科技有限公司 | A kind of forecasting system and method |
| CN110139160B (en) * | 2019-05-10 | 2022-07-22 | 北京奇艺世纪科技有限公司 | Prediction system and method |
| CN111401637A (en) * | 2020-03-16 | 2020-07-10 | 湖南大学 | User experience quality prediction method fusing user behavior and expression data |
| CN111401637B (en) * | 2020-03-16 | 2023-06-16 | 湖南大学 | User experience quality prediction method integrating user behavior and expression data |
| CN112822482A (en) * | 2020-12-31 | 2021-05-18 | 上海掌门科技有限公司 | Method and equipment for determining evaluation score of audio and video call |
| CN112862250A (en) * | 2021-01-12 | 2021-05-28 | 浙江知行教育科技有限公司 | College learning evaluation system and method based on big data |
| CN112862250B (en) * | 2021-01-12 | 2023-12-26 | 北京漂洋过海科技有限责任公司 | College learning evaluation system and method based on big data |
| CN114598934A (en) * | 2022-02-11 | 2022-06-07 | 山东悦知教育科技有限公司 | Education software data processing method based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107018408A (en) | The Quality of experience appraisal procedure of mobile terminal HTTP video flowings | |
| CN107038213B (en) | Video recommendation method and device | |
| CN107197368A (en) | Determine method and system of the user to multimedia content degree of concern | |
| CN103338223B (en) | A kind of recommendation method of Mobile solution and server | |
| US9849381B2 (en) | Methods and systems of automatic management online fantasy sports rosters | |
| US11770569B2 (en) | Providing risk based subscriber enhancements | |
| CN110324662B (en) | Video cover generation method and device | |
| CN103533393B (en) | The family's analysis noted down based on home audience and program commending method | |
| CN109672939A (en) | A kind of method and device of marking video content temperature | |
| CN106339507B (en) | Streaming Media information push method and device | |
| CN107205178A (en) | Direct broadcasting room recommends method and device | |
| CN109636481A (en) | User's portrait construction method and device towards domestic consumer | |
| CN107172452A (en) | Direct broadcasting room recommends method and device | |
| CN104216883A (en) | Video recommendation reason generating system and method | |
| CN101764661A (en) | Data fusion based video program recommendation system | |
| CN105843876B (en) | Quality evaluation method and device for multimedia resources | |
| CN105721899B (en) | A kind of method and system of video quality score | |
| CN109151488B (en) | Method and system for recommending live broadcast room in real time according to user behaviors | |
| KR20110073282A (en) | Consolidating input messages for social activity summarization | |
| CN109543132A (en) | Content recommendation method, device, electronic equipment and storage medium | |
| CN109729433A (en) | A kind of video playing appraisal procedure and device | |
| CN110933473A (en) | Video playing heat determining method and device | |
| US20230091980A1 (en) | Analytics in video/audio content distribution networks | |
| CN109635192A (en) | Magnanimity information temperature seniority among brothers and sisters update method and platform towards micro services | |
| CN105550334B (en) | A kind of video recommendation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170804 |