Objective Quality Metrics Assessment for Cloud Gaming

Abstract

This paper aims to provide objective quality metrics assessment for cloud gaming using machine learning algorithms. Three classification algorithms (i.e., Random Forest, Random Three and J-48) have been used for the development of models for objective quality assessment of two metrics: blurriness and blockiness. The results indicate that Random Forest has the best performance in this experimental case of objective quality metrics assessment for cloud gaming. Future research activities will cover comparison of a broad range of objective quality metrics and machine learning algorithms while using larger dataset to enhance the results significance.