Unveiling the limitations of synthetic mobility data generation algorithms in real-life scenarios

What is it about?

Synthetic generation models for trajectory data are currently an active field of research. They promise high flexibility and thus utility while simultaneously offering high privacy. We define new utility measures closer to real-life requirements and find that current state-of-the-art models do not meet these requirements yet.

Featured Image

Why is it important?

Synthetic data will only be used in practice if it fulfills real-life requirements, thus, respective utility evaluations are vital to understand the potential and limits of synthetic data.