Multidisciplinary Institute in Artificial intelligence

Observation and modelling of the dynamics of proximity social networks

ABSTRACT

The emerging technologies of wearable wireless devices open entirely new ways to record various aspects of human social interactions in a broad range of settings. Such technologies allow to log the temporal dynamics of face-to-face interactions by detecting the physical proximity of participants. In this talk we build on a proximity dataset collected during a longitudinal social experiment aiming to understand the co-evolution of children’s language development and social network. Physical proximity and verbal communication of hundreds of pre-school children and their teachers are recorded over three years using autonomous wearable low power wireless devices. First, we rely on this dataset to explore several pipelines of dynamical event reconstruction methods based on Hidden Markov Model, or on Long Short-Term Memory models, some of them combined with supervised pre-classification of interaction packets. Second, we demonstrate how temporal scales can be detected in proximity data using the Fourier transformation of dynamical similarity signals. Finally, exploiting multiple proximity networks we study the temporal dynamics of pre-schoolers and university students at both individual and group levels, to characterise how individuals navigate groups and how groups form and disaggregate. We find robust patterns across contexts and propose a dynamical agent-based model that closely reproduces empirical observations.

Márton Karsai

is an Associate Professor at the Department of Network and Data Science at the Central European University in Vienna. He is a Research Professor at the Rényi Institute of Mathematics in Budapest, Fellow of the ISI Foundation in Torino, director of the Social Data Science MSc program at CEU, where he leads the Computational Human Dynamics team. He is a network scientist with research interest in human dynamics, computational social science, and data science, especially focusing on heterogeneous temporal dynamics, spatial and temporal networks, socioeconomic systems and social contagion phenomena. He is an expert in analysing large human behavioural datasets and in developing data-driven models of social phenomena.

Link website

Revoir le seminaire