Application of Functional Kernel Hypothesis Testing for Channel Selection in Time Series Classification

IEEE Conference on Artificial Intelligence (CAI), vol. 118, pp. 245–247

Abstract

Multi-variate time series classification tasks are prevalent in various real-world engineering domains, including but not limited to activity recognition and anomaly detection. However, due to the abundance of sensors available, selecting the appropriate channels for successful classification can be a daunting task. In this study, we propose to use a two-sample hypothesis test, to determine the relevance of channels in time series classification tasks. We illustrate the industrial usecase that motivated this algorithm and validate our approach on open-source benchmarks. The proposed method has the potential to address the challenge of channel selection in multi-variate time series classification tasks and can significantly impact various real-world engineering applications.

Authors 3

1. Qi Huang Aachen

   Leiden University

   Affiliation as printed

   Leiden University,LIACS,Leiden,The Netherlands

   Leiden University Leiden, The Netherlands 0009-0007

2. Thomas Bäck Aachen

   Leiden University

   Affiliation as printed

   Leiden University,LIACS,Leiden,The Netherlands

   LIACS, Leiden University Leiden, The Netherlands 0000-0001

3. Bas van Stein Aachen

   Leiden University

   Affiliation as printed

   Leiden University,LIACS,Leiden,The Netherlands

   LIACS, Leiden University Leiden, The Netherlands 0000-0002-0013