Online Program

Return to main conference page

All Times EDT

Thursday, October 7
Thu, Oct 7, 2:45 PM - 4:00 PM
Understanding Our World

Analysis of Spatio-Temporal Data Using Bayesian and Formal Methods (309761)

*Laura Vana,  

Keywords: Bayesian methods, spatio-temporal data, urban mobility

We propose an interdisciplinary framework combining Bayesian statistics and spatio-temporal logic for optimizing urban mobility and smart city scenarios, by illustrating a unifying approach to the modeling and statistical analysis of the spatio-temporal data (through Bayesian methods) and the translation of the model output into decision-making (through spatio-temporal logic and statistical model checking). We build a spatio-temporal Bayesian model of population density dynamics for the City of Milan by relying on mobile phone data collected by Telecom Italia and aggregated on a grid in 10 minute intervals. We then formulate relevant properties that the system shall satisfy and check their probability of satisfaction using statistical model checking.