Field-theoretic modeling method for emotional context in social media: Theory and case study

Abstract

Just as masses and charges give rise to gravitational and electric fields, the online behaviors of individuals engaged in online social discourse give rise to an “emotional context” that conditions, and is conditioned by, these behaviors. Using Information Geometry and Unsupervised Learning, we have formulated a mathematical field theory for modeling online emotional context. This theory has been used to create a soft-ware application, Sirius15, that infers, characterizes, and visualizes the field structure (“emotional context”) arising from this discourse. A mathematical approach is presented to social media modeling that enables automated characterization and analysis of the emotional context associated with social media interactions. The results of a small, preliminary case study carried out by our team are presented.