We like to do some ad-hoc text analysis from time to time to break things up a bit and work with new tools and software. We’ve done some similar things with Twitter #hashtag text analysis titled Michigan Lean Startup Conf. Twitter Visualizations.
In the spirit of the upcoming election and debates, I thought it would be interesting to put out some something to summarize the words used by both of the candidates in the 2012 Second Presidential Debate on October 16, 2012. We grabbed the text from here. We’re not diving into anything overly complex here but it does put last night’s debate in a different context that we found interesting.
The way the graphic turned out is interesting: president, governor, jobs, thats people.
Link to the WordCloud: http://www.solidlogic.com/wp-content/uploads/2012/10/wordcloud_debate_transcript.png
2012 Second Presidential Debate Word Cloud
How to build a word cloud
The easiest way to build a word cloud is to use one of the great free online tools like Wordle to build the graphic. If you need a more customized approach or need to create something like this in software, you can use several software tools to make it a lot easier. More details to come on the methods and code behind this later on but its based on Python and R, both of which we use quite a bit for data analysis and development projects. The code for this was created by myself and our CIO, Michael Bommarito. Its based on some of the work he’s previously made available here: Wordcloud of the Arizona et al. v. United States opinion and Archiving Tweets with Python.
To get customized analysis like this, or to ask us anything else please use the contact us.