CEP Rule Extraction From Unlabeled Data in IoT

With the recent development of the Internet of Things, produced data are increasing day by day. These data have to be analyzed in real time. To provide real time analysis, Complex Event Processing is proposed to analyze the continuous and timely annotated data. Complex event processing detects complex events from atomic events via predefined rules which are mostly determined by domain experts. Determining complex event processing rules requires thorough knowledge of the data and data relations among data sources. It will be difficult to define a rule when it is considered that the scope and quantity of data is increased. Therefore, there is a need for extracting rules automatically. In this paper, we propose a novel model that extracts rules from unlabeled data by using clustering and rule mining algorithms. The model is evaluated in terms of classification performance and the results show that the proposed model is a promising solution for extracting complex event processing rules.