Exploring Context Generalizability in Citywide Crowd Mobility Prediction: An Analytic Framework and Benchmark

Abstract

Contextual features are important data sources for building citywide crowd mobility prediction models. However, the difficulty of applying context lies in the unknown generalizability of contextual features (e.g., weather, holiday, and points of interests) and context modeling techniques across different scenarios. In this paper, we present a unified analytic framework and a large-scale benchmark for evaluating context generalizability. The benchmark includes crowd mobility data, contextual data, and advanced prediction models. We conduct comprehensive experiments in several crowd mobility prediction tasks such as bike flow, metro passenger flow, and electric vehicle charging demand. Our results reveal several important observations: (1) Using more contextual features may not always result in better prediction with existing context modeling techniques; in particular, the combination of holiday and temporal position can provide more generalizable beneficial information than other contextual feature combinations. (2) In context modeling techniques, using a gated unit to incorporate raw contextual features into the deep prediction model has good generalizability. Besides, we offer several suggestions about incorporating contextual factors for building crowd mobility prediction applications. From our findings, we call for future research efforts devoted to developing new context modeling solutions.

References in corpus (14)

• XGBoost: A Scalable Tree Boosting System
• Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting
• WaveNet: A Generative Model for Raw Audio
• Short-Term Forecasting of Passenger Demand under On-Demand Ride Services: A Spatio-Temporal Deep Learning Approach
• Spatial-Temporal Graph ODE Networks for Traffic Flow Forecasting
• Deep and Confident Prediction for Time Series at Uber
• Predicting Citywide Crowd Flows in Irregular Regions Using Multi-View Graph Convolutional Networks
• BuildSenSys: Reusing Building Sensing Data for Traffic Prediction with Cross-domain Learning
• Preserving Dynamic Attention for Long-Term Spatial-Temporal Prediction
• Exploring the Generalizability of Spatio-Temporal Traffic Prediction: Meta-Modeling and an Analytic Framework
• BusTr: Predicting Bus Travel Times from Real-Time Traffic
• Prediction of Traffic Flow via Connected Vehicles
• Modeling Network-level Traffic Flow Transitions on Sparse Data
• Enhancing Spatiotemporal Traffic Prediction through Urban Human Activity Analysis