Time Series Cluster Kernel for Learning Similarities between Multivariate Time Series with Missing Data

TL;DR

This paper introduces the Time Series Cluster Kernel (TCK), a robust method for measuring similarities between multivariate time series with missing data, leveraging Gaussian mixture models and ensemble learning to outperform existing techniques.

Contribution

The paper proposes a novel kernel method that effectively handles missing data and attribute dependencies in multivariate time series using GMMs and ensemble learning.

Findings

TCK is robust to parameter choices.

TCK achieves competitive results on complete data.

TCK outperforms state-of-the-art methods with missing data.

Abstract

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.