STP-TrellisNets: Spatial-Temporal Parallel TrellisNets for Metro Station Passenger Flow Prediction

Abstract

Recent years have witnessed a drastic increase in the number of urban metro passengers, which inevitably causes the overcrowdedness in the metro systems of many cities. Clearly, an accurate prediction of passenger flows at metro stations is critical for a variety of metro system management operations, such as line scheduling and staff preallocation, that help alleviate such overcrowdedness. Thus, in this paper, we aim to address the problem of accurately predicting metro station passenger (MSP) flows. Similar to other traffic data, such as road traffic volume and highway speed, MSP flows are also spatial-temporal in nature. However, existing methods for other traffic prediction tasks are usually suboptimal to predict MSP flows due to MSP flows' unique spatial-temporal characteristics. As a result, we propose a novel deep learning framework STP-TrellisNets, which for the first time augments the newly-emerged temporal convolutional framework TrellisNet for spatial-temporal prediction. The temporal module of STP-TrellisNets (named CP-TrellisNets) employs two TrellisNets in serial to jointly capture the short-and long-term temporal correlation of MSP flows. In parallel to CP-TrellisNets, its spatial module (named GC-TrellisNet) adopts a novel transfer flow-based metric to characterize the spatial correlation among MSP flows, and implements multiple diffusion graph convolutional networks (DGCNs) in time-series order with their outputs connected to a TrellisNet to capture the dynamics of such spatial correlation. Clearly, GC-TrellisNet essentially integrates TrellisNet with graph convolution, and empowers TrellisNet with the ability to capture dynamic graph-structured correlation. We conduct extensive experiments with two large-scale real-world automated fare collection datasets, which contain respectively about 1.5 billion records in Shenzhen, China and 70 million records in Hangzhou, China. The experimental results demonstrate that STP-TrellisNets outperforms the state-of-the-art baselines. CCS CONCEPTS • Information systems → Spatial-temporal systems; Data mining; • Computing methodologies → Neural networks.

Sensing and Forecasting Crowd Distribution in Smart Cities: Potentials and Approaches

“…In particular, this study combines time-variant features such as weather conditions or historical traffic at subway stations, with data of station profile in order to deliver accurate predictions of crowd flow. On the same topic, that of metro station passengers flow prediction, the work in [109] proposes a novel deep learning framework for spatio-temporal correlation called STP-TrellisNet. The framework can capture the short-and long-term temporal correlations with graph convolution and has the ability to capture dynamic graph-structured correlations.…”

Completion and Augmentation based Spatiotemporal Deep Learning Approach for Short-Term Metro Origin-Destination Matrix Prediction under Limited Observable Data

“…Traffic flow mainly includes inflow/outflow, e.g. number of passengers departing or arriving at a specific set of regions [23,37] and OD flow, e.g. the number of passengers from one station to another [11,12].…”

Networked Time Series Prediction with Incomplete Data

“…An important task in such applications is to predict the future values of a NETS based on its historical values and the underlying graph. This has been studied extensively and many prediction methods have been proposed in various contexts; see [3,5,9,11,15,22,30,34,38,39,42,44] and references therein. Most of these methods use deep learning and require complete data for training [3,9,11,15,22,30,34,38,39,42].…”

“…This has been studied extensively and many prediction methods have been proposed in various contexts; see [3,5,9,11,15,22,30,34,38,39,42,44] and references therein. Most of these methods use deep learning and require complete data for training [3,9,11,15,22,30,34,38,39,42]. However, in real-world scenarios, it is not uncommon to have missing data due to sensor malfunction, incomplete sensing coverage, etc.…”

“…Existing methods typically use GCN [18] to capture inter-time series correlations, and variants of Convolutional Neural Network (CNN) [37] or Recurrent Neural Network (RNN) [32] to capture temporal correlations, e.g. [3,9,11,15,22,30,34,38,39,42,44]. However, these networks are not designed for incomplete data.…”