![Page 1: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/1.jpg)
1
Developing a Predictive Model ofQuality of Experience for Internet Video
Athula Balachandran
Carnegie Mellon University
![Page 2: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/2.jpg)
2
Content
1 Why do quality of experience(QoE)?
2 Where are the challenges?
3 How to do ?
4 Implication and evaluation
![Page 3: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/3.jpg)
3
Content
11 Why do quality of experience(QoE)? Why do quality of experience(QoE)?
2 Where are the challenges?
3 How to do ?
4 Implication and evaluation
![Page 4: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/4.jpg)
4
Why?
Tow main revenue models:
1.Subscription
2.Advertisement
advertisement
subscription
![Page 5: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/5.jpg)
5
Why?
Tow main revenue models:
1.Subscription
2.Advertisement
advertisement
subscription
The more you watch, The more we profit.
![Page 6: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/6.jpg)
6
Why?
Tow main revenue models:
1.Subscription
2.Advertisement
advertisement
subscription
The more you watch, The more we profit.
Improving users’ quality of experience(QoE) is crucial
![Page 7: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/7.jpg)
7
Content
1 Why do quality of experience(QoE)?
22 Where are the challenges? Where are the challenges?
3 How to do ?
4 Implication and evaluation
![Page 8: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/8.jpg)
8
Era changed & Requirement improve
■Video quality: PSNR(Peak Signal-to-Noise Ration)
<average bitrate, join time, buffering ration, rate of buffering>
■ User experience: User Opinion Scores
User’s Engagement-centric( viewing time , number of visits)
![Page 9: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/9.jpg)
9
Era changed & Requirement improve
■Video quality: PSNR(Peak Signal-to-Noise Ration)
<average bitrate, join time, buffering ration, rate of buffering>
■ User experience: User Opinion Scores
User’s Engagement-centric( viewing time , number of visits)
Average bitrate: HD(High-Definition) SD(Standard-Definition) LD(Low-
Definition)
Join time: load time
Buffering ratio: buffer_time/(buffer_time+play_time)
Rate of buffering: frequency of buffering
![Page 10: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/10.jpg)
10
Challenge scope
Video quality User engagement
1.Video quality interdependence
2.Complex relationship
3.Confound factors influence
bitrate
Join time
bufratio
…
Visits num
Viewing time
…
Time of day
Type of video
…
![Page 11: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/11.jpg)
11
Challenge 1 quality interdependence
Among video quality are subtle interdependence
1.Video quality interdependence
bitrate
Join time
bufratio
…
![Page 12: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/12.jpg)
12
Challenge 2 complex relationship
Relationship between quality and engagement2.Complex relationship
bitrate
Join time
Visits num
Viewing time
…
![Page 13: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/13.jpg)
13
Challenge 3 confound factors
Confound factors affect quality -> engagement
3.Confound factors influence
Type of Device
Type of video
…
![Page 14: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/14.jpg)
14
Content
1 Why do quality of experience(QoE)?
2 Where are the challenges?
33 How to do ? How to do ?
4 Implication and evaluation
![Page 15: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/15.jpg)
15
Compare current work
![Page 16: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/16.jpg)
16
Compare current work
1.Model consider complex relationship and confound factors
2.Provide strategy for system design
![Page 17: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/17.jpg)
17
Requirements for predictive model
Tackling relationship (quality->engagement) and interdependency (among quality)
Tackling confounding factors 1.Identifying the import confounding factors2.Address the confounding factors
![Page 18: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/18.jpg)
18
Compare methods for tackling relation
Compare the accuracy of tackling relationship( quality -> engagement) and interdependency (among quality)
![Page 19: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/19.jpg)
19
Confounding factors-Identify
----3 round filter for all possible Con. Factors----
![Page 20: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/20.jpg)
20
Confounding factors-Identify
----3 round filter for all possible Con. Factors----
Round1: calculate Information Gain
Entropy:
Condition entropy:
Information gain:
Relative Information gain:
![Page 21: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/21.jpg)
21
Confounding factors-Identify
----3 round filter for all possible Con. Factors----
Round1: calculate Information Gain
Entropy:
Condition entropy:
Information gain:
Relative Information gain:
……
![Page 22: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/22.jpg)
22
Confounding factors-Identify
----3 round filter for all possible Con. Factors----
Round1: calculate Information Gain
Round2: compare Compacted Decision Tree
![Page 23: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/23.jpg)
23
Round2: compare Compacted Decision Tree
![Page 24: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/24.jpg)
24
Round2: compare Compacted Decision Tree
GE-1: A1<=4 ■ [sup=40%,con=100%]GE-2:A1>4 ○ [sup=60%,con=82%] Except: A1>7,A2<=2.5 ■ [sup=20%,con=86%] Except:A1<8.5,A2>1.8,A2<=2.5 ○ [sup=14%,con=100%]
![Page 25: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/25.jpg)
25
Round2: compare Compacted Decision Tree
GE-1: A1<=4 ■ [sup=40%,con=100%]GE-2:A1>4 ○ [sup=60%,con=82%] Except: A1>7,A2<=2.5 ■ [sup=20%,con=86%] Except:A1<8.5,A2>1.8,A2<=2.5 ○ [sup=14%,con=100%]
■ ○ Total
A1<=4 24 0 24
A1 > 4 6.45 29.55 36
Total 30.45 29.55 60
χ2(A1<=4 ■) = (24-30.45)^2/30.45 = 1.37χ2(A1>4 ○) = (29.55-29.55)^2/29.55 = 0
![Page 26: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/26.jpg)
26
Round2: compare Compacted Decision Tree
GE-1: A1<=4 ■ [sup=40%,con=100%]GE-2:A1>4 ○ [sup=60%,con=82%] Except: A1>7,A2<=2.5 ■ [sup=20%,con=86%] Except:A1<8.5,A2>1.8,A2<=2.5 ○ [sup=14%,con=100%]
■ ○ Total
A1>4,A2<=2.5,A1>7 6.45 1.05 7.5
A1>4,A2<=2.5,A1<=7 0 7.5 7.5
Total 6.45 8.55 15
χ2(..A1>7 ■) = (6.45-6.45)^2/6.45 = 0χ2(..A1<=7 ○) = (7.05-8.55)^2/8.55 = 0.26
1.Dif 2.sig 3. n(current) >= n(former) +1
![Page 27: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/27.jpg)
27
Round2: compare Compacted Decision Tree
GE-1: A1<=4 ■ [sup=40%,con=100%]GE-2:A1>4 ○ [sup=60%,con=82%] Except: A1>7,A2<=2.5 ■ [sup=20%,con=86%] Except:A1<8.5,A2>1.8,A2<=2.5 ○ [sup=14%,con=100%]
![Page 28: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/28.jpg)
28
Confounding factors-Identify
----3 round filter for all possible Con. Factors----
Round1: calculate Information Gain
Round2: compare Compacted Decision Tree
![Page 29: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/29.jpg)
29
Confounding factors-Identify
----3 round filter for all possible Con. Factors----
Round1: calculate Information Gain
Round2: compare Compacted Decision Tree
Round3: compare Tolerance
![Page 30: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/30.jpg)
30
Confounding factors-Identify
----3 round filter for all possible Con. Factors----
Round1: calculate Information Gain
Round2: compare Compacted Decision Tree
Round3: compare Tolerance
![Page 31: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/31.jpg)
31
Confounding factors-Identify
----3 round filter for all possible Con. Factors----
Round1: calculate Information Gain
Round2: compare Compacted Decision Tree
Round3: compare Tolerance
![Page 32: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/32.jpg)
32
Confounding factors-Address
Compare tow candidate way:
![Page 33: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/33.jpg)
33
Confounding factors-Address
Compare tow candidate way:
Add as a new feature
Split data by Con. factors
![Page 34: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/34.jpg)
34
Content
1 Why do quality of experience(QoE)?
2 Where are the challenges?
3 How to do ?
44 Implication and evaluation Implication and evaluation
![Page 35: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/35.jpg)
35
Implication for system design
For an example model:buffering ratio, rate of buffering,
join time
Estimate all possible combinations
![Page 36: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/36.jpg)
36
Implication for system design
![Page 37: 1 Developing a Predictive Model of Quality of Experience for Internet Video Athula Balachandran Carnegie Mellon University](https://reader030.vdocuments.net/reader030/viewer/2022032701/56649c7c5503460f9493060c/html5/thumbnails/37.jpg)
Thanks!