Predicting citywide crowd flows using deep spatio-temporal residual networks

DOI10.1016/J.ARTINT.2018.03.002zbMATH Open1445.68196arXiv1701.02543OpenAlexW2964319113WikidataQ114388142 ScholiaQ114388142MaRDI QIDQ1647516FDOQ1647516

Authors: Junbo Zhang, Yu Zheng, Dekang Qi, Ruiyuan Li, Xiuwen Yi, Tianrui Li 

Publication date: 26 June 2018

Published in: Artificial Intelligence (Search for Journal in Brave)

Abstract: Forecasting the flow of crowds is of great importance to traffic management and public safety, and very challenging as it is affected by many complex factors, including spatial dependencies (nearby and distant), temporal dependencies (closeness, period, trend), and external conditions (e.g., weather and events). We propose a deep-learning-based approach, called ST-ResNet, to collectively forecast two types of crowd flows (i.e. inflow and outflow) in each and every region of a city. We design an end-to-end structure of ST-ResNet based on unique properties of spatio-temporal data. More specifically, we employ the residual neural network framework to model the temporal closeness, period, and trend properties of crowd traffic. For each property, we design a branch of residual convolutional units, each of which models the spatial properties of crowd traffic. ST-ResNet learns to dynamically aggregate the output of the three residual neural networks based on data, assigning different weights to different branches and regions. The aggregation is further combined with external factors, such as weather and day of the week, to predict the final traffic of crowds in each and every region. We have developed a real-time system based on Microsoft Azure Cloud, called UrbanFlow, providing the crowd flow monitoring and forecasting in Guiyang City of China. In addition, we present an extensive experimental evaluation using two types of crowd flows in Beijing and New York City (NYC), where ST-ResNet outperforms nine well-known baselines.

Full work available at URL: https://arxiv.org/abs/1701.02543

Recommendations

• A spatiotemporal hierarchical attention mechanism-based model for multi-step station-level crowd flow prediction
• Metro short-term traffic flow prediction with deep learning
• An interpretable model for short term traffic flow prediction
• Short-Term Traffic Flow Prediction Based on Road Network Topology
• Short-term urban traffic flow prediction based on MLS-SVM and spatiotemporal features

zbMATH Keywords

cloudconvolutional neural networksspatio-temporal datacrowd flowsresidual learning

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Artificial neural networks and deep learning (68T07)

Cites Work

• Time series forecasting using a hybrid ARIMA and neural network model
• Neural network forecasting for seasonal and trend time series
• Predicting traffic volumes and estimating the effects of shocks in massive transportation systems

Cited In (7)

• Context-aware spatio-temporal event prediction via convolutional Hawkes processes
• Multi-view graph convolutional networks with attention mechanism
• A spatiotemporal hierarchical attention mechanism-based model for multi-step station-level crowd flow prediction
• Deep learning for real-time crime forecasting and its ternarization
• AdaLo: adaptive learning rate optimizer with loss for classification
• Dual-track spatio-temporal learning for urban flow prediction with adaptive normalization
• Robust learning with imperfect privileged information

Uses Software

• TensorFlow
• Keras
• ImageNet
• Adam
• AlexNet
• Faster R-CNN

This page was built for publication: Predicting citywide crowd flows using deep spatio-temporal residual networks

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1647516)