By Patrick Larsen
This post includes links to websites created for my Applied Media Analytics course, in which I used the programming language R to find stories in big data.
For this project, I used text of every Presidential Inaugural Address in the history of the United States to try to find interesting trends around major historical events or trends. I used the Bing sentiment lexicon and ggplot2 to get to and display my findings.
For this project, I furthered my experiments with text analysis, this time opting to use data from the Twitter API. Specifically, I was interested in analyzing tweets from and about Kanye West, as this project was being done in the wake of his chaotic return to the social media platform. I used a couple of different sentiment lexicons and visualization packages to find and display my results.
For this project, I scratched the surface of R’s mapping capabilities by comparing the adoption of Naloxone to the number of opioid overdose deaths throughout North Carolina. It took a lot of work in the leaflet package, but it wasn’t terribly difficult to find interesting discrepancies in the data once I got it all working.