Although several feature extraction algorithms have been developed for time series data, most of them cannot be directly applied to multivariate time series. In particular, only few studies attempted for extracting correlation or relationship between multivariate time series of different features. Here we develop an algorithm that extract features from multivariate time series. We examine the effectiveness of the extracted features under the unsupervised outlier detection scenario. We evaluate

more »

... he proposed algorithm, called feature extraction using kernel and stacking (FEKS), on artificial and real-world data sets.