Prediction, Expectation, and Surprise: Methods, Designs, and Study of a Deployed Traffic Forecasting Service

Abstract

We present research on developing models that forecast traffic flow and congestion in the Greater Seattle area. The research has led to the deployment of a service named JamBayes, that is being actively used by over 2,500 users via smartphones and desktop versions of the system. We review the modeling effort and describe experiments probing the predictive accuracy of the models. Finally, we present research on building models that can identify current and future surprises, via efforts on modeling and forecasting unexpected situations.

Appears in Proceedings of the Twenty-First Conference on Uncertainty in Artificial Intelligence (UAI2005)

References in corpus (4)

• A Bayesian Approach to Learning Bayesian Networks with Local Structure
• Display of Information for Time-Critical Decision Making
• Learning Bayesian Networks: A Unification for Discrete and Gaussian Domains
• Staged Mixture Modelling and Boosting

Cited by in corpus (7)

• Deep Learning for Short-Term Traffic Flow Prediction
• Distributed Classification of Urban Congestion Using VANET
• Location-Based Reasoning about Complex Multi-Agent Behavior
• Bayesian Particle Tracking of Traffic Flows
• Optimizing city-scale traffic through modeling observations of vehicle movements
• Variational Inference with Agent-Based Models
• What Does The Crowd Say About You? Evaluating Aggregation-based Location Privacy