Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction
By Junbo Zhang, Yu Zheng, Dekang Qi. AAAI Conference on Artificial Intelligence, 2016
2,444 citations recorded by the source corpus.
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, such as inter-region traffic, events, and weather. We propose a deep-learning-based approach, called ST-ResNet, to collectively forecast the inflow and outflow of crowds 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. Experiments on two types of crowd flows in Beijing and New York City (NYC) demonstrate that the proposed ST-ResNet outperforms six well-known methods.
Subfields: Artificial Intelligence, Building and Construction, Ocean Engineering
View Paper: https://www.semanticscholar.org/paper/0f7ef24660d2524fbda2a3822d070f38dcc239ef
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
In Building and Construction
- Yinhai Wang
- Philip S. Yu
- Ruimin Ke
- Zhiyong Cui
- Senzhang Wang
- Nizar Bouguila
- Saeed Rahmani
- Asiye Baghbani
In Ocean Engineering
Citations in the corpus
- BRITS: Bidirectional Recurrent Imputation for Time Series (2018, background): “Multivariate time series data are abundant in many application areas, such as financial marketing [5, 4], health-care [10, 22], meteorology [31, 26], and traffic engineering [29, 35].”
- Flow Prediction in Spatio-Temporal Networks Based on Multitask Deep Learning (2020, methodology): “For fusing external factors, one can choose one of the following ways: the G ATED fusion introduced in Section 3.3, S IMPLE fusion (the sum fusion in [33]), or not use (i.e., w/o).”
- Self-Attention ConvLSTM for Spatiotemporal Prediction (2020, background): “It deserves to be studied in depth to serve the practical applications, such as traffic flows prediction (Zhang et al. 2017; Xu et al. 2018), precipitation fore-”
- Spatio-Temporal Self-Supervised Learning for Traffic Flow Prediction (2022, background): “• STFGNN (Li and Zhu 2021): it integrates with STFGN module and a novel gated CNN module, and captures hidden spatial dependencies by a data-driven graph and its further fusion with given spatial graphs.”
- GMAN: A Graph Multi-Attention Network for Traffic Prediction (2019, methodology): “To model spatial correlations, researchers apply convolutional neural networks (CNN) to capture the dependencies in Euclidean space (Zhang, Zheng, and Qi 2017; Yao et al. 2018; 2019).”
- Spatio-Temporal Graph Structure Learning for Traffic Forecasting (2020, methodology): “In the paset few years, scholars proposed many methods to capture the temporal dependencies, such as directly stacking the data of different times (Zhang, Zheng, and Qi 2017), classical CNN-LSTM framework (Yao et al. 2018), Conv-LSTM or GCGRU (Li et al. 2018), ST-Conv Block (Yu, Yin, and Zhu 2018)…”
- Deep Learning in Mobile and Wireless Networking: A Survey (2018, background): “, [204], [258]–[261]), as they can effectively exploit spatial and temporal correlations.”
- Spatial-Temporal Transformer Networks for Traffic Flow Forecasting (2020, methodology): “[15] [16] adopt CNNs to extract the spatial features in which traffic networks are converted to regular grids.”
- A Survey on Modern Deep Neural Network for Traffic Prediction: Trends, Methods and Challenges (2020, methodology): “data is used in:, [4], [54], [62], [65] and [67].”
- Dynamic Graph Convolutional Recurrent Network for Traffic Prediction: Benchmark and Solution (2021, methodology): “It is obvious that traffic condition can be significantly affected or inferred by external factors such as weather [33, 35] and POI [7, 17], which need extra collection and preprocessing.”
- Deep Air Quality Forecasting Using Hybrid Deep Learning Framework (2018, background): “proposed a novel deep residuals network to analyze how the congestions are evolving [25].”
- When Will You Arrive? Estimating Travel Time Based on Deep Neural Networks (2018, background): “(Zhang, Zheng, and Qi 2017) proposed a deep spatio-temporal residual network for predicting the crowd flows.”