Ideology Detection for Twitter Users via Link Analysis

The problem of ideology detection is to study the latent (political) placement for people, which is traditionally studied on politicians according to their voting behaviors. Recently, more and more studies begin to address the ideology detection problem for ordinary users based on their online behaviors that can be captured by social media, e.g., Twitter. As far as we are concerned, the vast majority of the existing methods on ideology detection on social media have oversimplified the problem as a binary classification problem (i.e., liberal vs. conservative). Moreover, though social links can play a critical role in deciding one’s ideology, most of the existing work ignores the heterogeneous types of links in social media. In this paper we propose to detect numerical ideology positions for Twitter users, according to their follow, mention, and retweet links to a selected set of politicians. A unified probabilistic model is proposed that can (1) integrate heterogeneous types of links together in determining people’s ideology, and (2) automatically learn the quality of each type of links in deciding one’s ideology. Experiments have demonstrated the advantages of our model in terms of both ranking and political leaning classification accuracy.

[1]  R. M. Alvarez.  Birds of the Same Feather Tweet Together: Bayesian Ideal Point Estimation Using Twitter Data , 2014 .

[2]  M. McPherson,et al.  Birds of a Feather: Homophily in Social Networks , 2001 .

[3]  Tim Groseclose,et al.  A Measure of Media Bias , 2005 .

[4]  K. T. Poole,et al.  Congress: A Political-Economic History of Roll Call Voting , 1997 .

[5]  Derek L. Hansen,et al.  Computing political preference among twitter followers , 2011, CHI.

[6]  Sean Gerrish,et al.  How They Vote: Issue-Adjusted Models of Legislative Behavior , 2012, NIPS.

[7]  Sean Gerrish,et al.  Predicting Legislative Roll Calls from Text , 2011, ICML.

[8]  Yupeng Gu,et al.  Topic-factorized ideal point estimation model for legislative voting network , 2014, KDD.

[9]  K. T. Poole,et al.  Patterns of congressional voting , 1991 .

[10]  Qiaozhu Mei,et al.  Classifying the Political Leaning of News Articles and Users from User Votes , 2011, ICWSM.

[11]  K. T. Poole,et al.  A Spatial Model for Legislative Roll Call Analysis , 1985 .

[12]  James J. Heckman,et al.  Linear Probability Models of the Demand for Attributes with an Empirical Application to Estimating the Preferences of Legislators , 1996 .

[13]  Mung Chiang,et al.  Quantifying Political Leaning from Tweets and Retweets , 2013, ICWSM.

[14]  Joshua D. Clinton,et al.  The Statistical Analysis of Roll Call Data , 2004, American Political Science Review.

[15]  Simon Jackman,et al.  Multidimensional Analysis of Roll Call Data via Bayesian Simulation: Identification, Estimation, Inference, and Model Checking , 2001, Political Analysis.

[16]  Kevin A. Hassett,et al.  Is newspaper coverage of economic events politically biased? , 2014 .

[17]  Felix Ming,et al.  Media, Pundits and the U.S. Presidential Election: Quantifying Political Leanings from Tweets , 2012 .

[18]  John Londregan,et al.  Estimating Legislators' Preferred Points , 1999, Political Analysis.

[19]  Steven Skiena,et al.  DeepWalk: online learning of social representations , 2014, KDD.

[20]  Philip Resnik,et al.  Political Ideology Detection Using Recursive Neural Networks , 2014, ACL.

[21]  Charles Elkan,et al.  Link Prediction via Matrix Factorization , 2011, ECML/PKDD.

[22]  Philip Resnik,et al.  Tea Party in the House: A Hierarchical Ideal Point Topic Model and Its Application to Republican Legislators in the 112th Congress , 2015, ACL.