Overview

My research mainly lies in the applied machine learning and natural language processing domains. I primarily build and apply tools to collect data on or model human behaviors. For example, I developed a variation of the public goods game and deployed it to collect participant data. More recently, I have been developing tools to apply Large Language Models, such as Meta’s Llama-2, to perform named-individual identification and stance detection on tweets. Along with this, I am working on ways to identify unique groups of users based on what they tweet, and then identify ways to identify pairs of similar users across these groups.

Current Projects

I am currently working on comparing various ways individuals reacted to Bernie Sanders dropping out of the 2020 Democratic Primary (and thus the race for president as a whole). This involves topic modelling, as well as developing tools to identify pairs of similar users across these groups. Along with this, I am investigating the emergence, engagement with, and discussions of conspiracy theories online.

Publications

Posters

Previous Research Projects

While completing my bachelors degree, I had the opportunity to volunteer as a research assistant in the Center for Applied Moral Psychology under the supervision of Dr. Azim Shariff and Dr. Joe Green. In this project, I developed a variation of the public good game using the oTree framework and python. Once built I deployed it to collect data from participants and wrote a report on my findings, which can be read here.