Probabilistic Dependency Networks for Prediction and Diagnostics

TL;DR

This paper introduces probabilistic dependency networks to model, understand, and predict transportation events, demonstrating improved accuracy over existing methods by capturing causal relationships between events.

Contribution

It presents a novel implementation of probabilistic graphical models for learning dependencies and predicting attributes in transportation event networks.

Findings

Dependency graphs effectively model causal relationships.

The method achieves higher prediction accuracy than state-of-the-art approaches.

The approach provides insights into the causal effects among transportation events.

Abstract

Research in transportation frequently involve modelling and predicting attributes of events that occur at regular intervals. The event could be arrival of a bus at a bus stop, the volume of a traffic at a particular point, the demand at a particular bus stop etc. In this work, we propose a specific implementation of probabilistic graphical models to learn the probabilistic dependency between the events that occur in a network. A dependency graph is built from the past observed instances of the event and we use the graph to understand the causal effects of some events on others in the system. The dependency graph is also used to predict the attributes of future events and is shown to have a good prediction accuracy compared to the state of the art.