This thesis discusses the dynamic web service selection in the semantic context with QoS constraints. The goal of this work is to investigate the mechanism of automated QoS-based semantic web service selection. Semantic Web Service (SWS) aims to achieve the automation of web service tasks, such as service discover, selection, composition and invocation. The task of semantic web service selection is further investigated through this thesis. An architecture is proposed to achieve this task by considering QoS parameters. The QoS parameters are classified into dynamic and static attributes in the architecture. The dynamic attributes are evaluated and measured as an overall value by applying utility functions. This overall value can be modeled in the semantic context for the purpose of service selection. Furthermore, the architecture directly models the static QoS attributes in the semantic context for service selection. Finally, an open SWS challenge scenario named hardware purchasing is used in several experiments in order to evaluate the proposed architecture.