interests

I study social systems using a mix of computational and traditional methods. Substantively, my interests tend towards political communication, political participation, public opinion, and campaigns and elections. Methodologically, I use tools from survey research, content analysis, machine learning (especially natural language processing), and game theory.



Dissertation
My dissertation taps one of the richest sources of political data in all history---the political blogosphere---in order to understand timeless aspects of human interaction: attention, opinion, disagreement, persuasion, and participation.

One-page abstract


Keywords:
political communication, political participation, public opinion, campaigns and elections, media and information technology, survey methodology, content analysis, game theory, network theory, incentive-centered design, machine learning, natural language processing, econometrics.

vita

my CV
Current as of May 2012

contact info

Mailing address:
Abe Gong
CO: University of Michigan Political Science
5700 Haven Hall
Ann Arbor, MI 48109

Phone:
c: (801) 830-7029

Email:
put "umich.edu" after "agong@"

publications

An automated snowball census of the political web

working papers

SnowCrawl: Software for directed webcrawling
Genres of journalism and blogging: A large-scale content analysis
With Aleksandra Leyzerovskaya and Sean Walser.
Predicting individual survey responses from blog content
With Michael Bommarito
Computational political science
Review article.
Targeting, not position-taking: A formal model of targeted communication in electoral campaigns.

conference submissions

Networks and Language in the 2010 Election
Political Networks 2011, with Avishay Livne, Matthew Simmons, Eytan Adar, and Lada Adamic.
An automated snowball census of the political web
Voted best paper at JITP 2011
To what degree does the political blogosphere represent -- or distort -- the voice of the electorate?
MPSA 2011.
Known Certainty, Measurement Error, and the Validity of Automated Text Classification in Regression Analysis
APSA 2010
Targeting, not position-taking
   A formal model of targeted communication in American elections

Poster presented at MPSA 2010
Applying Voice Recognition to Vox Populi:
   State Transition Models in the Study of Public Opinion and Political Communication

APSA 2009
Chunky or smooth?
   New methods for analyzing the dynamics of aggregate opinion during presidential campaigns.

MPSA 2009
A model of In-group Pressure and Inter-group Conflict
With Joshua Gubler. MPSA 2009.
Network analysis of toxic chemicals and symptoms:
   Implications for designing first-responder systems

Proceedings of AMIA '07
Popping the undergraduate news bubble:
   Factors influencing current events knowledge among college students

Proceedings of the Central States Communication Association 2007 meeting in Minneapolis, Minnesota