United States Presidential Debates - A Machine Learning and Multivariate Purview

Dr. T. Leo Alexander, C. Amelia Betsy

Abstract: Presidential Debate is like a Job Interview, the voters watch out for a trait in the candidates. The entire venture was done using the transcripts of all the candidates. The purpose of this paper is to highlight the significance of the topics spoken and how it affects their victory. Only explicit debates (with the moderator) from 1960 have been taken into consideration

Keywords: Candidates, CART- Classification and Regression Tree, Cluster analysis, Principal Components, Text Mining