Trump - Political Science

I've created 3 mini-projects to analyse Trump's tweets

1. Text analysis of Trump's tweets

• to understand the tweeting habits/ pattern/ characteristics on different devices
• learn to create a wordcloud
• use natural language processing technique to identify the sentiment of words used on different devices

2. Strategy for Trump's US election 2016

• to identify the change of words used before and after US election and understand how Trump changed his strategy to defend his position

3. who_was_tweeting_for_Trump.ipynb

• learn to handle imbalance dataset
• leverage the first mini-project - Text analysis of Trump's tweets to predict which tweets were written by Trump himself

Installation

Install jupyter notebook (please see https://jupyter.org/install for the details)

Required libraries

pip install pandas
pip install matplotlib
pip install seaborn
pip install nrclex
pip install nltk
pip install sklearn
pip install wordcloud

Dataset

You can download it in the following website: https://www.thetrumparchive.com/

Run

git clone https://github.com/miki-lwy/Trump
cd Trump
jupyter notebook

Reference

[code in R] http://varianceexplained.org/r/trump-tweets/ [code in R] http://varianceexplained.org/r/trump-followup/