Time Series Cluster Kernel for Learning Similarities between Multivariate Time Series with Missing Data
This addresses the challenge of similarity-based analysis for multivariate time series with missing data, which is incremental as it builds on kernel methods and ensemble techniques.
The paper tackles the problem of learning similarities between multivariate time series with missing data by proposing a time series cluster kernel (TCK), which leverages Gaussian mixture models and ensemble learning to handle missing data and reduce parameter tuning. The results show that TCK is robust to parameters, competitive for multivariate time series without missing data, and outstanding for missing data.
Similarity-based approaches represent a promising direction for time series analysis. However, many such methods rely on parameter tuning, and some have shortcomings if the time series are multivariate (MTS), due to dependencies between attributes, or the time series contain missing data. In this paper, we address these challenges within the powerful context of kernel methods by proposing the robust \emph{time series cluster kernel} (TCK). The approach taken leverages the missing data handling properties of Gaussian mixture models (GMM) augmented with informative prior distributions. An ensemble learning approach is exploited to ensure robustness to parameters by combining the clustering results of many GMM to form the final kernel. We evaluate the TCK on synthetic and real data and compare to other state-of-the-art techniques. The experimental results demonstrate that the TCK is robust to parameter choices, provides competitive results for MTS without missing data and outstanding results for missing data.