Research on Generation Algorithm of Complex Event Processing Rules Based on Time Series

Abstract

Complex event processing (CEP) technology filters and aggregates events according to user-defined rules to extract the information needed by users. It is widely used in data stream analysis and processing. Traditionally, the rule of CEP engines are often manually deployed. Manual deployment put great limitation to the application of CEP. It is difficult for domain experts to accurately adapt to changing environments and different applications. The combination of data mining, machine learning algorithms and complex event processing to achieve automatic rule generation has been proposed by many scholars. Aiming at the research of the recently proposed time series shapelets in automatic rule generation, an improved automatic rule generation algorithm is presented. Compared with the original algorithm, the experimental results show that it has a good effect in improving the accuracy of data processing and the earliness of classification.