An Embedding Model for Predicting Roll-Call Votes

Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 2066–2070, Austin, Texas, November 1-5, 2016. c⃝2016 Association for Computational Linguistics An Embedding Model for Predicting Roll-Call Votes Peter E. Kraft Hirsh Jain Alexander M. Rush School of Engineering and Applied Science, Harvard University {pkraft, hirshjain}@college.harvard.edu, srush@seas.harvard.edu Abstract We develop a novel embedding-based model for predicting legislative roll-call votes from bill text. The model introduces multidimen- sional ideal vectors for legislators as an alter- native to single dimensional ideal point mod- els for quantitatively analyzing roll-call data. These vectors are learned to correspond with pre-trained word embeddings which allows us to analyze which features in a bill text are most predictive of political support. Our model is quite simple, while at the same time allowing us to successfully predict legislator votes on specific bills with higher accuracy than past methods. 1 Introduction Quantitative analysis of political data can contribute to our understanding of governments. One impor- tant source of such data is roll-call votes, records of how legislators vote on bills. Analysis of roll-call data can reveal interesting information about legisla- tors (such as political leanings and ideological clus- ters) and can also allow prediction of future votes (Clinton, 2012). Previous work on analyzing roll-call votes has chiefly involved positioning congresspeople on ideal point models. Ideal point models assume all legisla- tors and bills can be plotted as single points in one- dimensional “political space.” The closer a particu- lar bill’s position is to a particular congressperson’s, the more utility the congressperson is expected to derive from the bill. Initial work on ideal point models focused on using them to test theories about legislative behavior, such as predicting that the rel- ative differences between ideal points of congress- people of different parties, and thus party polar- ization, would increase over time (McCarty, 2001). Ideal point models are often created using Bayesian techniques over large amounts of roll-call data (Clin- ton et al., 2004; Jackman, 2001). However, these models are not used to make predictions. They are trained using the complete vote matrix for the bill, which indicates how each congressperson voted on each bill. Therefore, they cannot say anything about how congresspeople will vote on a new bill, as until some congresspeople have voted on the bill its ideal point is not known. We target this vote prediction problem: given the text of a bill and a congressperson, can we indepen- dently predict how each congressperson will vote on the bill? The first prior attempt at this task was made by Gerrish and Blei (2011) who create an ideal point topic model which integrates a topic model similar to LDA for the bill text with an ideal point model for the congresspeople. They use variational inference to approximate the posterior distribution of the top- ics and ideal points, predicting with a linear model. Gerrish and Blei (2012) further extend this work with an issue-adjusted model, a similar model that modifies congressperson ideal points based on top- ics …

Content truncated to 500 words