Graph Traffic Dataset Creator
A SUMO GUI that builds route-aware graph datasets from simulated traffic and real trajectories — the dataset-generation side of the NiDS 2026 paper, with two public Kaggle releases.
This is Graph Traffic Dataset Creator, the desktop tool that produces the route-aware dynamic graphs used in the NiDS 2026 paper Route-Aware Dynamic Traffic Graphs for ETA: Dataset Generation and Interactive Model Testing. You load a SUMO network, run the city simulation, and export mapping files, snapshots, and labels — or import real traces, map-match them, and write the same graph form.
Two complementary datasets from that pipeline are on Kaggle so others can train without rerunning the generator. One is the three-zone city SUMO simulation; the other is Porto taxi trajectories converted to graph snapshots.
Code: https://github.com/Turgibot/Multi-Variant-Simulated-Traffic-Dataset-Creator-and-Model-Tester
Kaggle (SUMO, three-zone city): https://www.kaggle.com/datasets/guytordjman/3-zones-city-traffic-simulation
Kaggle (Porto taxi graph snapshots): https://www.kaggle.com/datasets/guytordjman/porto-taxi-converted-to-graph-snapshots
Related publication: /publications/nids-2026-route-aware-traffic-graphs