Video QoE Prediction Based on User Profile

Raffael Shalala, Ran Dubin, Ofer Hadar, Amit Dvir

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

5 Scopus citations

Abstract

The increasing popularity of online video content and adaptive video streaming services, especially ones based on HTTP Adaptive Streaming, highlights the need for streaming optimization solutions. Predicting end users Quality of Experience (QoE) by using machine learning algorithms, may allow content servers to allocate bandwidth smartly and more efficiently. In this work, we present a new user quality of experience prediction algorithm which extracts features based on user traffic pattern parameters such as bit-rate, resolution, frame rate, etc. In order to optimize the features set and the corresponding machine learning algorithms, we have used three different feature selection algorithms and six different classifiers. We show that the Decision Tree algorithm achieved 86% accuracy in predicting the user quality of experience.

Original languageEnglish
Title of host publication2018 International Conference on Computing, Networking and Communications, ICNC 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages588-592
Number of pages5
ISBN (Electronic)9781538636527
DOIs
StatePublished - 19 Jun 2018
Event2018 International Conference on Computing, Networking and Communications, ICNC 2018 - Maui, United States
Duration: 5 Mar 20188 Mar 2018

Publication series

Name2018 International Conference on Computing, Networking and Communications, ICNC 2018

Conference

Conference2018 International Conference on Computing, Networking and Communications, ICNC 2018
Country/TerritoryUnited States
CityMaui
Period5/03/188/03/18

ASJC Scopus subject areas

  • Computer Science Applications
  • Hardware and Architecture
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Video QoE Prediction Based on User Profile'. Together they form a unique fingerprint.

Cite this