Causal and Local Correlations Based Network for Multivariate Time Series Classification

TL;DR

This paper introduces CaLoNet, a novel network that models causal and local correlations in multivariate time series to improve classification accuracy, addressing the neglect of spatial and local feature correlations in prior methods.

Contribution

The paper proposes CaLoNet, which integrates causality-based spatial correlation modeling and local feature fusion into a graph neural network for multivariate time series classification.

Findings

CaLoNet achieves competitive accuracy on UEA datasets.

The method effectively models spatial and local correlations.

It outperforms several state-of-the-art approaches.

Abstract

Recently, time series classification has attracted the attention of a large number of researchers, and hundreds of methods have been proposed. However, these methods often ignore the spatial correlations among dimensions and the local correlations among features. To address this issue, the causal and local correlations based network (CaLoNet) is proposed in this study for multivariate time series classification. First, pairwise spatial correlations between dimensions are modeled using causality modeling to obtain the graph structure. Then, a relationship extraction network is used to fuse local correlations to obtain long-term dependency features. Finally, the graph structure and long-term dependency features are integrated into the graph neural network. Experiments on the UEA datasets show that CaLoNet can obtain competitive performance compared with state-of-the-art methods.