QoE beyond the MOS: Added value using quantiles and distributions

Tobias Hoßfeld, Poul E. Heegaard, Martin Varela

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

17 Scopus citations

Abstract

Traditionally, Quality of Experience (QoE) assessment results (or objective estimations thereof) are presented as a single scalar value, typically a Mean Opinion Score (MOS). While useful, the limitations of MOS are evident even in its name; for many applications, just having a mean value is simply not enough. For service providers in particular, it would be more interesting to have an idea of how the scores are distributed, so as to ensure that a certain portion of the user population is experiencing satisfactory levels of quality, thus reducing churn. In this paper we propose different statistical measures to express important aspects of QoE beyond MOS like user diversity, uncertainty of user rating distributions, ratio of dissatisfied users. Further, we propose a way to use MOS values and the standard deviation of opinion scores (SOS) hypothesis, which postulates a quadratic relation between subjective scores and their standard deviation, in order to derive quantiles for subjective ratings.

Original languageEnglish
Title of host publication2015 7th International Workshop on Quality of Multimedia Experience, QoMEX 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781479989584
DOIs
StatePublished - 2 Jul 2015
Externally publishedYes
Event2015 7th International Workshop on Quality of Multimedia Experience, QoMEX 2015 - Costa Navarino, Messinia, Greece
Duration: 26 May 201529 May 2015

Publication series

Name2015 7th International Workshop on Quality of Multimedia Experience, QoMEX 2015

Conference

Conference2015 7th International Workshop on Quality of Multimedia Experience, QoMEX 2015
Country/TerritoryGreece
CityCosta Navarino, Messinia
Period26/05/1529/05/15

Fingerprint

Dive into the research topics of 'QoE beyond the MOS: Added value using quantiles and distributions'. Together they form a unique fingerprint.

Cite this