Learning Hidden Curved Exponential Family Models to Infer Face-to-Face Interaction Networks from Situated Speech Data

Abstract

In this paper, we present a novel probabilistic frame-work for recovering global, latent social network struc-ture from local, noisy observations. We extend curved exponential random graph models to include two types of variables: hidden variables that capture the structure of the network and observational variables that capture the behavior between actors in the network. We develop a novel combination of informative and intuitive conver-sational (local) and structural (global) features to spec-ify our model. The model learns, in an unsupervised manner, the relationship between observable behavior and hidden social structure while simultaneously learn-ing properties of the latent structure itself. We present empirical results on both synthetic data and a real world dataset of face-to-face conversations collected from 24 individuals using wearable sensors over the course of 6 months.