Establishing Patterns of the Urban Transport Flows on Clustering Analysis
YAISD-WS 2025 · CEUR Workshop Proceedings 3974, pp. 1-9
Editorial summary
Traffic states are grouped by similarities in flow levels, variability and change over time. Adaptive time windows and a combination of HDBSCAN and k-means help separate recurring patterns from noise. Evaluation uses a simulated Khmelnytskyi road network and identifies four main traffic scenarios.
Bibliographic reference
Use the publisher’s record for the citation format required by your journal or organisation.
Vitaliy Pavlyshyn, Oleksander Ryzhanskyi, Eduard Manziuk, Pavlo Radiuk, Oleksander Barmak, Iurii Krak. "Establishing Patterns of the Urban Transport Flows on Clustering Analysis". YAISD-WS 2025 · CEUR Workshop Proceedings 3974, pp. 1-9, 2025.
Test a method against your business task.
Describe the decision your system needs to support and the data you have. We assess whether a method fits, define a prototype and agree how to measure its quality, response time and operating cost.