Towards a Smart(er) Social Science Using High-Dimensional Continuous-Time Trajectories from the Open Dynamic Interaction Networks (ODIN) Platform

In this paper, we describe Open Dynamic Interaction Networks (ODIN), a software platform designed to move the social, behavioral, and public health sciences toward a new investigative paradigm. ODIN enables us to collect and analyze rich, contextual continuous-time data on both personal change and interpersonal interaction. It achieves this by supporting dynamic delivery of questions based on the currently sensed context of each participant. The ODIN system extends beyond static (or even stepwise dynamic) graph-theoretic renderings of social life and individual behavior by considering “social relationship” to be measured in terms of high-dimensional continuous-time trajectories. The system is designed to be extensible, allowing seamless incorporation of new sensors, and correspondingly sophisticated compound rules by which contexts of interest may be specified. As such, ODIN opens the door for a "smarter" social science based on continuous contextual data, and a "smarter" data science that is reflective and sociologically informed.