Spatial Graph Coarsening: Weather and Weekday Prediction with London's Bike-Sharing Service using GNN

TL;DR

This paper applies Graph Neural Networks with novel spatial coarsening and feature concatenation techniques to predict weather and weekdays in London using bike-sharing data, outperforming baseline models.

Contribution

Introduces Spatial Graph Coarsening and feature concatenation operators in GNNs for improved city-scale prediction tasks.

Findings

Outperformed baseline models in accuracy and loss

Effective use of geographical and land-use features

Validated on London's bike-sharing dataset

Abstract

This study introduced the use of Graph Neural Network (GNN) for predicting the weather and weekday of a day in London, from the dataset of Santander Cycles bike-sharing system as a graph classification task. The proposed GNN models newly introduced (i) a concatenation operator of graph features with trained node embeddings and (ii) a graph coarsening operator based on geographical contiguity, namely "Spatial Graph Coarsening". With the node features of land-use characteristics and number of households around the bike stations and graph features of temperatures in the city, our proposed models outperformed the baseline model in cross-entropy loss and accuracy of the validation dataset.