[ICDE'20] Predicting Origin-Destination Flow via Multi-Perspective Graph Convolutional Network (Pytorch Replication)

Probabilistic graphical models to learn Origin-Destination matrices in transportation networks using TensorFlow

Generating Neural Spatial Interaction Tables

Implementation of the spatialGAT in the paper: Spatial Attention Based Grid Representation Learning for Predicting Origin–Destination Flow (IEEE Big Data 2022)

This is an experiment version of calibrating origin-destination matrix estimation using link traffic counts

An optimal planning for traffic measurement is a balance between deploying as few counters as possible while not significantly decreasing the coverage/accuracy of travel movements information.

Implementation of the HiUrNet in the paper: Explainable Hierarchical Urban Representation Learning for Commuting Flow Prediction (ACM SIGSPATIAL 2024)

Estimate origin and destination positions based on smartcard usage time and transit position.

Stochastic modelling of urban travel demand

The existing ODM valuation methods define the value of a new ODM by comparing it to an existing ODM. There is a need to define and quantify the quality of ODM through objective parameters. This thesis presents an alternative approach to ODM evaluation, defines objective quality parameters, comparison conditions and decision process