Shapelets and Parallel Coordinates Based Automated Query Generation for Complex Event Processing

Automating the query generation for Complex Event Processing (CEP) has marked its own importance in allowing users to obtain useful insights from data. Existing techniques are both computationally expensive and require extensive domain-specific human interaction. In addressing these issues, we propose a technique that combines both parallel coordinates and shapelets. First, each instance of the multivariate data is represented as a line on a set of parallel coordinates. Then a shapelet-learner algorithm is applied to those lines to extract the relevant shapelets. Afterwards, the identified shapelets are ranked based on their information gain. Next, the shapelets with similar information gain are divided into groups by a shapelet-merger algorithm. The best group for each event is then identified based on the event distribution of the dataset. Then the best group is used to generate the query to detect the complex events. The proposed technique can be applied to both multivariate and multivariate time-series data, and it is computationally and memory efficient. It enables users to focus only on the shapelets with relevant information gains. We demonstrate the utility of the proposed technique using a set of real-world datasets.