The Quality of Experience (QoE) is an irreplaceable metric for evaluating the perceived quality of consumers of multimedia content. Due to the subjectiveness of QoE the most suitable way to measure it is by executing subjective studies. However, executing subjective studies is a complex and expensive process. Careful recreation of the viewing conditions is necessary, and a strict selection of the test subjects is required based on many criteria. This is why solutions are often found in various objective methodologies for measuring the QoE of multimedia. These solutions even though more practical are less accurate and cannot reflect the user expectations. In this work we present a method for building QoE prediction models using machine learning techniques from continuous real-time customer feedback, i.e., during the service execution. This online learning approach builds and adapts prediction models that estimate the QoE based on given Quality of Service metrics from real-time user feedback and does not require apriori execution of subjective studies.
|Title of host publication||Proceedings of the 2nd International Workshop on Quality of Multimedia Experience (QoMEX), June 21-23, Trondheim, Norway|
|Place of Publication||Piscataway|
|Publisher||Institute of Electrical and Electronics Engineers|
|Publication status||Published - 2010|