QoE-Affective-computing-Dataset
This work is proposed to describe and to share a subjective QoE dataset that assess emotional factors of YouTube platform users. This dataset is collected in a controlled laboratory environment using YouTube JavaScript player and Video camera as a sensor. The subjective Absolute Category Rate (ACR) method is used to evaluate the user's QoE based on Mean Opinion Score (MOS) ratting score.
To build this dataset, a testbed is achieved in the LiSSi laboratory. 21 testers participated in the test campaign. All of them were researchers and students from different disciplines aged 18 to 34 years with few or no experience with video assessment experimentation. The collected parameters concern many video parameters, network parameters and emotionals factors.
The dataset was built from a controlled laboratory testbed where 120 samples covering 22 Quality of Experience Impact Factors (QoE IFs). Three videos are used. Each on had a different type/complexity.
You can download the dataset and find more details here:
https://github.com/Lamyne/QoE-Affective-computing-Datatset
References:
Lamine Amour, Sami Souihi, Said Hoceini, Abdelhamid Mellouk: A Hierarchical Classification Model of QoE Influence Factors. WWIC 2015: pages 225-238, May 2015.
Lamine Amour, Sami Souihi, Mohamed Ikbel Boulabiar and Mellouk Abdelhamid. An Improved QoE Estimation Method based on QoS and Affective Computing, The 13th International Symposium on Programming and Systems (ISPS'2018), Algeries, Algeria. April 24-26, 2018.