On the development of the web service quality modelling space

Abstract

Most web service (WS) based systems consider the quality of service (QoS) to select an appropriate WS to run an application. Some authors have proposed QoS attributes that characterize WS at the technological layer to determine QoS. However, this is not enough to meet user needs and ensure user satisfaction at the domain layer. Consequently, there is a need to develop a deeper understanding of the user Quality of Experience (QoE) for WS. This paper proposes a QoE modelling space, which views and allows predicting QoE through QoS attributes, WS architecture layers, and stakeholder viewpoints from a particular perspective. The proposed WS quality modelling space was validated using the objective (the WS-DREAM dataset), and subjective (expert’s judgments) attributes to estimate QoS performance by ARAS and to predict QoE performance by a Fuzzy Control System (FCS). The obtained results show the strong positive linear and strong positive monotonic relationships between WS QoS performance and WS QoE performance. Therefore, the proposed WS quality modelling space is suitable for QoE performance prediction based on measurable QoS attributes.