The Web of Things (WoT) aims at connecting things to applications using Web technologies and standards, on top of the Internet of Things. WoT applications allow gathering, processing and sharing various types of high-level information. They must be flexible and adapt to dynamic changes in their environment. The question explored in the scope of this thesis is: how can we deal with context adaptation in WoT applications? The objective is to enable scalable multi-purpose context adaptation for WoT applications, through reusable context models to describe context and reason about it.