Political Dimensionality Estimation Using a Probabilistic Graphical Model

  • Yoad Lewenberg
  • Yoram Bachrach
  • Lucas Bordeaux
  • Pushmeet Kohli

Proceedings of the Thirty-Second Conference on Uncertainty in Artificial Intelligence, (UAI), June 25-29, 2016, New York City, NY |

Published by AUAI Press

This paper attempts to move beyond the left-right characterization of political ideologies. We propose a trait based probabilistic model for estimating the manifold of political opinion. We demonstrate the efficacy of our model on two novel and large scale datasets of public opinion. Our experiments show that although the political spectrum is richer than a simple left-right structure, peoples’ opinions on seemingly unrelated political issues are very correlated, so fewer than 10 dimensions are enough to represent peoples’ entire political opinion.