Site Search
Computer Science


Dr. Ziad Kobti lecturingDr. Ziad Kobti
Dr. Ziad Kobti
Dr. Scott GoodwinDr. Scott Goodwin
Dr. Scott Goodwin
Robin Gras, Ph.D.Dr. Robin Gras
Dr. Robin Gras
Arunita Jaekel, Ph.D.Dr. Arunita Jaekel
Dr. Arunita Jaekel
Dr. Luis RuedaDr. Luis Rueda
Dr. Luis Rueda
Jessica Chen, Ph.D.Dr. Jessica Chen
Dr. Jessica Chen
Lambton TowerLambton Tower
Lambton Tower
Alioune Ngom, Ph.D.Dr. Alioune Ngom
Dr. Alioune Ngom
Windsor WaterfrontWindsor Waterfront Park
Windsor Waterfront Park
Imran Ahmad, Ph.D.Dr. Imran Ahmad
Dr. Imran Ahmad
Dr. Robert KentDr. Robert Kent
Dr. Robert Kent
Xiaobu Yuan, Ph.D.Dr. Xiaobu Yuan
Dr. Xiaobu Yuan
Christie Ezeife, Ph.D.Dr. Christie Ezeife
Dr. Christie Ezeife

Improve Adaptive Video Streaming through Machine Learning

Add this event into your calendar using the iCAL format
  • Fri, 01/19/2018 - 10:00am - 12:00pm


Improve Adaptive Video Streaming through Machine Learning

MSc Thesis Defense by:

Anh Minh Le

Date:  Friday, January 19, 2018
Time:  10:00 am – 12:00pm
Location: 3105, Lambton Tower

Abstract: Many Adaptive Bitrate (ABR) algorithms have emerged recently in order to improve the Quality of Experience (QoE) in video streaming. These ABRs all function by predicting the future throughput rate of the current video session. Such prediction could be formulated into the parameters to the algorithms. We follow the data-driven approach to learning the best parameter values from the logged sessions of throughput values. To improve the quality of the prediction, we propose to properly classify the logged sessions according to the critical features that affect the network conditions such as Internet Service Provider(ISP), geographical location etc. In this thesis, we present our work on modifying the existing Decision Tree algorithm for our feature-based session partitioning to help learning the best parameter values and improve the performance of existing ABR algorithms. The experiment shows that this approach can improve the performance of the ABR algorithm by up to 8.59%, with 98.38% of the testing sessions performing better.

Thesis Committee:
Internal Reader:          Dr. Stephanos Mavromoustakos
External Reader:         Dr. Sévérien Nkurunziza
Advisor:                      Dr. Jessica Chen
Chair:                          Dr. Yung H. Tsin

See More: