Graph Neural Networks and Time Series as Directed Graphs for Quality Recognition

TL;DR

This paper explores representing time series as directed graphs to leverage Graph Neural Networks for quality recognition, introducing two models: a classifier and an autoencoder, demonstrating their effectiveness on a specific task.

Contribution

It proposes a novel approach of modeling time series as directed graphs and develops two GNN-based models for classification and reconstruction tasks.

Findings

GNNs can effectively model time series as directed graphs.

The models achieve promising results on quality recognition.

The approach offers a new perspective for time series analysis.

Abstract

Graph Neural Networks (GNNs) are becoming central in the study of time series, coupled with existing algorithms as Temporal Convolutional Networks and Recurrent Neural Networks. In this paper, we see time series themselves as directed graphs, so that their topology encodes time dependencies and we start to explore the effectiveness of GNNs architectures on them. We develop two distinct Geometric Deep Learning models, a supervised classifier and an autoencoder-like model for signal reconstruction. We apply these models on a quality recognition problem.