Presidential Candidate Deep Dive: "What are they saying?"

Hillary Clinton and Donald Trump have logged countless hours in front of a podium the past several months. This provides an opportunity to use natural language processing to drill down on their messages.

Using a sentiment analysis toolkit which employs three sentiment dictionaries , I examined one speech for each candidate. Given the range of topics being discussed in rallies and with the press, I thought it best to look at both candidate's Super Tuesday victory speeches. Below is a quick look at their positive and negative sentiments throughout their delivery. Values greater than zero indicate positive sentiment and below zero indicate negative.

61% of Trumps speech has positive emotional content, where as Clinton's possessed 77%. To further elaborate, we can look at the primary emotions being conveyed in the language of the speech below.

The largest splits between Trump and Clinton came from positive and negative language. Trump's speech included a much higher proportion of the negative emotions using language indicative of fear and sadness, whereas Clinton had a higher percent of joy and trust in her speech.

We will take another look at these splits in the debates in a few months and see how each candidate's emotional message has evolved.

Featured Posts
Recent Posts
Archive
Search By Tags
Follow Us
  • Facebook Basic Square
  • Twitter Basic Square
  • Google+ Basic Square

Visualizations by Spencer Davison and inspired by d3js.org, Tableau Public, bl.ocks.org and various other sites in the analytics community.