MTS2Graph: Interpretable multivariate time series classification with temporal evolving graphs

Raneen Younis, Abdul Hakmeh, Zahra Ahmadi

Abstract: Highlights•We study the problem of multivariate time series classification using CNNs.•MTS2Graph is an interpretable framework focusing on highly activated signal segments.•We model the temporal dependencies of the segments via an evolution graph.•The proposed framework achieves state-of-the-art results over a wide set of methods.