top of page



Enan Srivastava, Hai Le


Given a political speech, will we be able to tell which side of the aisle it supports? We are interested in how the languages of the two major parties of the United States have changed over the years and are looking to develop an NLP model to classify the partisanship of the words that members of Congress use to assert their ideology. We use the dataset “Congressional Record for the 43rd-114th Congresses” from the Stanford’s Social Science Data Collection. The dataset includes speeches given in both chambers of Congress which we only analyze from 1981 to 2017. The goal is to determine the party, Democratic or Republican, the speech belongs to.

Screen Shot 2022-06-03 at 11.31.35 AM.png
github URL
bottom of page