Date of Graduation


Level of Access

Restricted Access Thesis

Department or Program

Computer Science

First Advisor

Mohammad Irfan


A senator’s decision to vote yea or nay on a bill comes as a result of their ideologies, agendas, and their interactions with other legislators. There have been many multidisciplinary studies into modeling and predicting voting behaviors based on historical roll-call data. One salient model in political science is the ideal point model, which assigns each senator and bill a number in low-dimensional Euclidean space (often the real line). Usually, Democratic senators and liberal bills get negative numbers, whereas Republican senators and conservative bills get positive numbers. These points then allow for prediction of future voting behavior. In this paper, we extend the classical ideal point model with network-structured interactions among senators. In contrast to the ideal point model’s prediction of individual voting behavior, ours predicts joint voting behaviors in a game-theoretic fashion. In addition, our model inherits a parameter from ideal point models that reflects the characteristics of a bill. This allows our model to outperform previous models that solely focus on the networked interactions among senators with no bill-specific parameters. We focus on two fundamental questions: learning the parameters of the model from real-world data and computing stable outcomes of the model in order to predict joint voting behavior. We demonstrate the effectiveness of our model through experimental studies based on data from the 114th U.S. Congress.