Aditi Khare
Pronouns: she/her/hers
Research Mentor(s): Christopher Fariss, Assistant Professor
Research Mentor School/College/Department: Political Science, College of Literature, Science, and the Arts
Presentation Date: Thursday, April 22, 2021
Session: Session 4 (2pm-2:50pm)
Breakout Room: Room 18
Presenter: 1
Abstract
In recent decades, whether climate change is real has become more and more of a partisan issue among everyday people. Over 97% of the scientific community, however, agrees that the climate has been changing due to human activity. In order to see if and how people’s views change when they actually experience a natural disaster that is likely due to climate change, we conduct an analysis of over 100,000 tweets before and after Hurricane Harvey.The analysis consists of labelling each tweet as part of one of six categories (climate change, Hurricane Harvey, sports, politics, policing, and other). Then, if the tweet relates to climate, the disposition of the tweet’s author towards climate change was also noted. Currently, in order to make the analysis of data more efficient, several machine learning classifiers are being built in the programming languages python and R. Once these classifier models are created, we will benchmark them and then use the best performing model to label new tweets at a much faster pace than a team of human coders. With the fully labeled corpus of climate change tweets, the study will be able to shed light on behavioral and political changes before and after climate change-induced events. Keywords: climate change, hurricane, Harvey, twitter, politics
Authors: Aditi Khare, Chris Fariss
Research Method: Computer Programming
