Observing and Modeling User Behavior onSocio-Spatial Interaction Networks: Conformance, Exceptions, and Anomalies

For modeling user behavior in AI systems, we canmake use of diverse data sources, that are heterogeneous, andcover different user facets. This paper investigates socio-spatialinteraction networks for modeling user interactions from threeperspectives: We analyze preferences and perceptions of face-to-face human interactions in relation to the interactions observedusing wearable sensors. For that, we investigate the correspon-dence of according networks, in order to identify conformance,exceptions, and anomalies. The analysis is performed on a real-world dataset capturing networks of face-to-face proximity (asa proxy for actual face-to-face communication between partici-pants) coupled with self-report questionnaires about preferencesand perception of those interactions.