Mobility pattern prediction

Unlike purely data-driven methods, a simulation-based approach enables prediction of traffic states under recurrent conditions but also to predict the impact of incidents or changes to traffic control plans.
Understanding mobility using unsupervised learning

Unsupervised learning is a great tool for data analysis, particularly for understanding how mobility has changed due to COVID-19.
Metric selection for incident detection systems

Classification problems with highly unbalanced datasets, such as incident detection, pose the trade-off between true positive rate and false negative rate.
How to extract patterns from traffic data for better insights into mobility

Mobility patterns are a cornerstone of mobility demand modelling; clustering helps us extract usable daily patterns from huge sets of traffic data.