[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

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.

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

Generating Neural Spatial Interaction Tables