A framework for context-aware semantic complex event processing

Abstract

With the continuing increase in the heterogeneity and quantity of data source, such as data from sensors, social networks, business, usage of background knowledge and context information is a growing challenge in complex event processing. To remedy this problem, this paper proposes a framework for context-aware semantic complex event processing, in which domain ontology is used to annotate heterogeneous data and infer high-level context afterwards with declarative rules in the domain ontology and some application-specific rules, and event ontology is used to model events and analyze user query based on the definitions and relationships between events in it. In the process of complex event processing, context is taken into consideration to improve detection accuracy, meanwhile event correlation is made according to relations among events to forecast what will happen based on the detected events and current context information. User therefore can obtain timely and personalized results, i.e. user can get better understanding of current environment and make more correct decision. In this framework, semantic background knowledge is used to improve the expressiveness and flexibility of complex event processing systems. As a result, this framework can enable more intelligent event processing which can understand the semantics of events in a certain context.