Authors:
Fatima Laiche
1
;
Asma Ben Letaifa
2
and
Taoufik Aguili
1
Affiliations:
1
Communication Systems Laboratory, ENIT, University Tunis El Manar, Tunis, Tunisia
;
2
MEDIATRON LAB, SUPCOM, Carthage University, Tunis, Tunisia
Keyword(s):
Video Streaming, Influence Factors, QoE, Machine Learning, User Behavior, Context, User Engagement.
Abstract:
The widespread use of online video content in every area of the connected world increases the interest in Quality of Experience (QoE). QoE plays a crucial role in the success of video streaming services. However, QoE prediction is challenging as many compelling factors (i.e., human and context factors) impact the QoE and QoE management solution often neglect the impact of social context and user behavior factors on the end-user’s QoE. To address these challenges, we have developed a web application to conduct subjective study and collect data from application-layer, user-level, and service-level. The collected data is then used as training set for machine learning models including decision tree, K-nearest neighbor, and support vector machine for the purpose of QoE prediction.