The purpose of this project is to create a system that will be used to explore COVID-19 political discourse on Twitter. Through NLP methods, we identify and analyze keywords, topics, and overall sentiments from each party, and propose a systematic approach to predict and distinguish a tweet's political stance (left of right leanings) based on its COVID-19 related terms using different classification algorithms on different language models.
Ultimately, the project was used to submit an academic paper which outlines in detail the background, methodology, analysis, and conclusion. In this document, there will be a very rough overview of those sections, but to read specifics, the full paper will be located within the files.
- Python
- Java
Implementation of bag-of-words, bigram, and TF-IDF models was used to identify and analyze tweets collected frrom the first case of COVID (17th of November 2019) until 17th of November, 2020. Specifically, tweets from political figures of right and left stances (Republican and Democratic) were chosen.
In comparing keyword trends and analyzing its sentiments, several distinctions can be made between the Republican and Democratic party regarding their perspectives on COVID-19 related issues. In general, results suggested that Democrats are more concerned with the casualties of the pandemic, and give more medical precautions and recommendations to the public, whereas Republicans are more invested in political responsibilities such as keeping the public updated through media and carefully watching the progress of the virus.
In collaboration with Jiwon Kim, Daniela Puchall, and Jingyi Liang through NYU's NLP (Natural Language Processing) Course.