Enhanced social distancing in multi-layer traffic networks through graph-based machine learning

• Applicants: Professor Dr. Stephan Günnemann; Professor Dr. Stephan Jonas
• Subject Area: Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing ; Data Management, Data-Intensive Systems, Computer Science Methods in Business Informatics
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 458547424

Pandemics spread directly (personal contact) or indirectly (touched surfaces). One potential driver of pandemics are bottlenecks in public transportation, through which thousands of commuters travel daily. Additionally, subway or train stations with interchange between lines or modes of transportation are often closed and narrow spaces, with limited possibilities for social distancing. These stations are therefore potential disease hotspots. We aim to investigate how infections can be controlled by proactively guiding travelers around potential hot spots during their commute. From a technical perspective, our approach requires novel models that can accurately forecast (short-term) traffic flow and predict hot spots which are likely to form in the near future. In this regard, we aim to investigate the potential of Machine Learning approaches, specifically studying Graph Neural Network principles. As our main research goals, we aim to understand how individuals adhere to crowd avoidance systems, how graph-based machine learning can support crowd avoidance, and how privacy of the underlying user data can be preserved while still enabling effective analysis. We plan to evaluate our approach with data of the Munich transportation authority.

• DFG Programme: Research Grants