Short-term Traffic Demand Prediction using Graph Convolutional Neural Networks

Abstract

Abstract. Short-term traffic demand prediction is one of the crucial issues in intelligent transport systems, which has attracted attention from the taxi industry and Mobility-on-Demand systems. Accurate predictions enable operators to dispatch their vehicles in advance, satisfying both drivers and passengers. This study aims to predict traffic demand over the entire city based on the Graph convolutional network (GCNN). Specially, we divide the study area into several non-overlap sub-regions. Each sub-region is treated as a node, and a traffic demand graph is constructed. Then, we build three graph convolution networks based on three different weighted adjacency matrices, which represent three graph structures. Furthermore, a data-driven graph convolutional network (DDGCNN) is developed, which can capture the correlation between pairs of sub-regions automatically. Finally, we compare our models with other prediction methods, including three GCNNs with a normal adjacency matrix, an existing data-driven graph convolutional neural network, historical average, and random forest. Results show that the weighted adjacency matrix can improve the prediction performance compared with a normal adjacency matrix. In addition, we proved that our DDGCNN outperforms other predictors in three aspects, i.e., performance over the test set, performance over the time aspect, and the performance over the spatial aspect.

Prediction of Bike‐sharing Trip Counts: Comparing Parametric Spatial Regression Models to a Geographically Weighted XGBoost Algorithm

“…The sMAPE can be calculated as follows for a data set including $$$ n $$$ observations $$$ {y}_i $$$ and predictions $$$ {\hat{y}}_i,\kern0.3em i=1,\dots, n $$$. $$$ c $$$ is a small positive constant (Li and Axhausen 2019), and we use $$$ c=1 $$$ here: $$$$ \mathrm{sMAPE}=\frac{1}{n}\sum \limits_{i=1}^n\frac{\left|{y}_i-{\hat{y}}_i\right|}{y_i+{\hat{y}}_i+c}. $$$$ This definition slightly differs from the original definition of this measure by Armstrong (1985).…”

“…First, the denominator is not divided by 2 to allow for an easier interpretation. Second, following Li and Axhausen (2019), the constant $$$ c $$$ is included to prevent division by zero that could otherwise occur if both observed and predicted value equal zero. The sMAPE is a dimensionless, percentage‐error metric (Botchkarev 2018).…”

“…Yang et al (2020b) apply a graph-based approach while also incorporating XGBoost in their study on bike-sharing. Li and Axhausen (2019) compare a whole range of different models including time-series models, machine learning approaches such as random forest or XGBoost, and deep learning approaches such as LSTM recurrent neural network to model taxi demand for two different data sets. The authors cannot determine one single best model.…”

“…The sMAPE can be calculated as follows for a data set including n observations y i and predictions ŷi , i = 1, ..., n. c is a small positive constant (Li and Axhausen 2019), and we use c = 1 here:…”

Hourly Long-Term Traffic Volume Prediction with Meteorological Information Using Graph Convolutional Networks

“…Traffic volume prediction plays a pivotal role in managing traffic conditions. Previous studies have explored short-term traffic volume prediction [1][2][3]. The high complexity of traffic volume prediction has hindered researchers from analyzing daily or hourly volume traffic volume predictions.…”