Jon Stewart often mocks the contorted phrases and talking points repeated by politicians. Can we detect this strategic coordination of language?
Tracing Coordination and Cooperation Structures via Semantic Burst Detection
Yu-Ru Lin, Drew Margolin & David Lazer
Developing technologies that support collaboration requires understanding how knowledge and expertise
are shared and distributed among community members. In this article, we explore two forms of knowledge
distribution structures, coordination and cooperation, that are central to successful collaboration, among
members in political communities. We propose a novel method for detecting the coordination of strategic
communication among these members. Our method is grounded on diverse research in communications,
centering on the idea that a “semantic burst” (a sudden burst in the usage of a particular set of concepts
by multiple individuals within a short time) is a signature of communication coordination. We detect
such semantic bursts as being the rapid emergence of unusual three-word sequences (trigrams). We apply
our method to public statements by U.S. Senators in the 112th U.S. Congress. Several cases of bursts
detected by this method are examined, each corresponding to a theoretical explanation for semantic
convergence. Coordination and cooperation networks amongst Senators are then constructed from the
semantic convergences that are detected. The structures of these networks are compared to one another and
to networks and attributes gathered for the Senate. Results indicate that the detected networks reflect certain
underlying tendencies in the social relationships amongst Senators. Analysis of the networks also reveals
interesting differences in how members of the different parties select whom to converge with.