As the amount of data created by today's pervasive environments increases exponentially, there is a stronger need to decipher the important information that is hidden among it. By using complex event processing, we can obtain the information that really matters to our organization and use it to improve our processes. However, even when this information is discovered, the processes remain static and cannot be changed to adapt to the actual scenario, diminishing the advantages that can be achieved. In this paper we present CEVICHE, a framework that combines the strengths of complex event processing and process adaption. CEVICHE deals with the different languages used to detect complex events while presenting a single interface to the user.