Title: Integration of Computational Methods to Understand the Spread of Climate Change Beliefs On Social Media

Presented by Uwaila Ekhator, Computing PhD student, Data Science emphasis

Abstract

Climate change continues to be a global challenge, requiring urgent action. However, climate misinformation and skepticism have become widespread, as social media platforms accelerate its spread, thereby undermining the seriousness of climate change.

This necessitates the need to understand the social learning strategies (SLS) behind the spread of climate skepticism. SLS are the biases that influence what, when, and whom individuals copy. The spread of climate beliefs may differ in communities of climate believers and skeptics, but identifying these communities using traditional community detection algorithms is challenging because of the need for network data, which is not easily accessible.

Stance detection (SD) can be utilized to overcome this challenge. SD utilizes the content of the tweet to infer the stance of the author. This paper reviews existing SLS in the domain of climate change communication, a promising empirical research direction to study SLS called ”generative inference”, and stance detection as a community detection method for identifying climate skeptics and believers.

Finally, we identify open problems, highlighting the need for empirical studies on cultural evolution in the domain of climate change communication.

Committee: Dr. Vicken Hillis (Advisor), Dr. Francesca Spezzano, Dr. Matt Williamson

CompEE: Dr. Edoardo Serra