Joey Livorno | gil15@pitt.edu | 4.24.2020
This is the repository for Joey's term project, Analysis of Trump Tweets. The goal of this project was to conduct an analysis of Donald Trump's Twitter feed and discover some trends in his opinions from the past 10 years. Specifically, twitter data from Donald Trump and a random subset of users was used to find overarching trends in his tweeting habits, conduct a sentiment analysis, and construct a machine learning model that predicts whether or not a given tweet was made by Donald Trump.
- README.md: the markdown file that acts as the directory of the repository
- LICENSE.md: contains the license for my data
- presentation.pptx: an oral presentation of my project accompanied by a slideshow
- project_plan.md: the plan of how I will complete my project frmo beginning to end
- project_report.md: chronological reports of the progression of my project
- final_report.md: the final report of my entire project
- code: the directory in which I keep my project code file (for an nbviewer version of my code, click here)
- data: the directory in which I keep my data
- images the directory containing the images that were generated from my project file
The data I use in my research came from several different sources. The data from Trump comes from the Trump Twitter Archive, which is a free database of the tweets of Donald Trump and many other prominent American political figures. The website was made by Brendan Brown. I also got a set of 200,000 randomly generated tweets from followthehashtag.com. This too was free to use.
I received regular feedback from my peers in Data Science for Linguists which can be found here. Thank you to everyone who contributed.