Revisiting Spatial-Temporal Similarity: A Deep Learning Framework for Traffic Prediction

By Huaxiu Yao, Xianfeng Tang, Hua Wei, Guanjie Zheng, Zhenhui Jessie Li. AAAI Conference on Artificial Intelligence, 2018

833 citations recorded by the source corpus.

Traffic prediction has drawn increasing attention in AI research field due to the increasing availability of large-scale traffic data and its importance in the real world. For example, an accurate taxi demand prediction can assist taxi companies in pre-allocating taxis. The key challenge of traffic prediction lies in how to model the complex spatial dependencies and temporal dynamics. Although both factors have been considered in modeling, existing works make strong assumptions about spatial dependence and temporal dynamics, i.e., spatial dependence is stationary in time, and temporal dynamics is strictly periodical. However, in practice the spatial dependence could be dynamic (i.e., changing from time to time), and the temporal dynamics could have some perturbation from one period to another period. In this paper, we make two important observations: (1) the spatial dependencies between locations are dynamic; and (2) the temporal dependency follows daily and weekly pattern but it is not strictly periodic for its dynamic temporal shifting. To address these two issues, we propose a novel Spatial-Temporal Dynamic Network (STDN), in which a flow gating mechanism is introduced to learn the dynamic similarity between locations, and a periodically shifted attention mechanism is designed to handle long-term periodic temporal shifting. To the best of our knowledge, this is the first work that tackle both issues in a unified framework. Our experimental results on real-world traffic datasets verify the effectiveness of the proposed method.

Subfields: Building and Construction, Transportation

View Paper: https://www.semanticscholar.org/paper/f7cd55c8b39b2627233253f12c74933833d293ea

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Building and Construction

In Transportation

Citations in the corpus

Dynamic and Multi-faceted Spatio-temporal Deep Learning for Traffic Speed Forecasting (2021, methodology): “In the early stage, with the distributions of traffic status or demands divided into grids on a map, CNNs were used to learn the spatial interactions of different grids, and RNNs were used to learn the evolutionary traffic dynamics [28, 30, 33].”
A Survey on Deep Learning for Human Mobility (2020, neutral): “[3] 2018 ConvLSTM RMSE, MAE Yao et al.[215] STDN 2018 CNN, LSTM, Attention, FC RMSE, MAPE [14] STDN-2 Jin et al.”
STG2Seq: Spatial-temporal Graph to Sequence Model for Multi-step Passenger Demand Forecasting (2019, background): “When the target demand dimension is 2 (predict both start-demand and end-demand [ Yao et al. , 2019 ] ), we can simply conduct channel attention for each dimension and concatenate them together to form the predicted passenger demand D T .”
Deep Learning for Spatio-Temporal Data Mining: A Survey (2019, neutral): “Application domains Related works Transportation [189], [199], [125], [32], [30], [6], [33], [136], [60], [121], [27], [114], [177], [90], [23], [73], [117], [135], [89], [148], [85], [137], [155], [157], [79], [107], [201], [22], [24], [108], [105], [16], [106], [93], [15], [175], [176], [104], [14”
Traffic transformer: Capturing the continuity and periodicity of time series for traffic forecasting (2020, background): “Driven by big data machine learning models for traffic forecasting have attracted considerable attention from both academia and industry for several years (Dai et al., 2019; Liu, Zheng, Chawla, Yuan, & Xing, 2011; Lv, Duan, Kang, Li, & Wang, 2014; Yao et al., 2019; Zhao et al., 2019).”
Dynamic Graph Convolutional Recurrent Network for Traffic Prediction: Benchmark and Solution (2021, methodology): “CNN [1]–[4], GNN [5]–[8], [12]–[23] and attention mechanism [7]–[11], [17], [19], [23], [25] have been widely used to capture spatial dependency.”
Learning Dynamics and Heterogeneity of Spatial-Temporal Graph Data for Traffic Forecasting (2021, background): “In particular, the proposals [21], [22], [23], [26], [32], [33], [34], [35] partition the region into 2D or 3D cells,”
Graph Neural Networks for Intelligent Transportation Systems: A Survey (2023, methodology): “The performance of their framework is evaluated using two real datasets and baseline models including ARIMA, LSTM, Hetero-ConvLSTM [204], STGCN [27], STG2seq [51], STSGCN [205], STDN [206], DFN [207], MTPSO [208].”
Deep Learning on Traffic Prediction: Methods, Analysis, and Future Directions (2021, background): “Considering that trafﬁc data is highly periodic, but not strictly periodic, [80] designed a periodically shifted attention mechanism to deal with long-term periodic dependency and periodic temporal shifting.”