Event-driven architecture and complex event processing have become important topics in achieving business reactivity and proactivity. Even though many technologies to support this have been developed, there is still a need for providing support for complex events based on different levels of required expressivity. On the other hand, semantic technologies and semantic-based systems have emerged and are becoming more and more used, whereas there is no recognized solution for event processing in information systems based on semantic technologies. In this paper, we address these issues. We present a framework for ontology-based support for complex events, which allows for semantically enriched event definitions and automatic recognition. In order to automatically recognize complex events, an appropriate event definition basis has to be defined. The paper represents a continuation of our previous work by enhancing this basis and developing a SPARQL-based framework to provide a richer mechanism for event definitions and more efficient event detection. As a proof-of-concept, we provide an implementation of this framework using the semantic integration platform - Information Workbench. The implementation is available as an app that runs on top of the Information Workbench platform.