[Colloq] Talk Today - Bruce Desmarais - Modeling interpersonal government communication networks: The integrated analysis of textual content and socio-organizational structure

[Jessica Biron]

Bruce Desmarais, University of Massachusetts – Amherst 

Modeling interpersonal government communication networks: The 
integrated analysis of textual content and socio-organizational 
structure 

Wednesday, January 30 
12:15-1:30 p.m. 
CCNR – 5th Floor of Dana Research Center (please take the elevator on the left) 

A light lunch will be served 


Abstract: In this talk I present a novel method for the statistical 
analysis of text-valued communication networks and introduce a larger 
project focusing on the study of relational government communication 
processes. A unique feature of interpersonal communication networks is 
that edges are annotated with content, commonly recorded in textual 
format. Few methods exist for the integrated analysis of textual edge 
content and network structure. We introduce a probabilistic Bayesian 
model for predictive and exploratory analysis in this context. The 
model is designed for the simultaneous discovery and visualization of 
topic-specific subnetworks in relational communication corpora. The 
method’s ability to discover, predict and visualize topic-specific 
communication patterns is showcased using a new email data set: the 
New Hanover County Government email network. The work presented 
constitutes pilot research for a broader project and research agenda 
that examines the interconnections across different domains in the 
governance process through the analysis of multifaceted electronic 
communications data available through the public record. 

Bio: Desmarais received his PhD in political science from the 
University of North Carolina at Chapel Hill in 2010. In Fall 2010, he 
started as an assistant professor in the Department of Political 
Science at the University of Massachusetts Amherst. He joined UMass as 
one of the core faculty members of the Computational Social Science 
Initiative, a collaborative organization of scholars in sociology, 
political science, statistics and computer science. 

Desmarais' primary areas of research include political network 
analysis, computational social science and political institutions 
(American emphasis). In his work, Desmarais focuses on rigorously and 
precisely identifying the complex ways in which political actors and 
institutions are interdependent, and developing quantitative methods 
capable of illuminating these dependencies. In his dissertation he 
developed methods for the identification and modeling of complex 
interdependence in collective decision-making processes, with 
extensive applications to the analysis of voting on the U.S. Supreme 
Court. 

His more recent work focuses on the development and application of 
methods for the analysis of political networks. Substantive 
applications of network analysis include international conflict, 
defense alliances, terrorist events, congressional campaign finance 
networks and legislative collaboration. His current work focuses on 
governmental communication networks, rooted in the analysis of 
large-scale electronic communications data available through the 
public record. 

Jessica Biron 
Administrative Assistant – Office of the Dean and CCIS Development 
College of Computer and Information Science 
Northeastern University 
202 West Village H 
617-373-5204 
bironje at ccis.neu.edu 
http://www.ccs.neu.edu/