Research

Current Projects

1. Co-evolving Bounded Confidence Opinion Dynamics Model

Currently, I am working on a research project on how opinions of people are influenced by their social connections using an opinion dynamics model in which people have “selective biases” and tend to only consider those opinions which are close to their own. A flavor of this model can be seen in this blog post of mine. More specifically, I am working on a co-evolving bounded confidence opinion dynamics model on networks, where individuals are able to rewire away from others that may have too different an opinion. My mentors are Professor Mason Porter and Dr. Michelle Feng.

While this model in particular is on the more theoretical/experimental side of mathematical social science, opinion dynamics is a very rich and applicable field and definitely deserves attention. Understanding the way people’s opinions change can help us understand real-life phenomena like protests and political choices.

This was my first project that I started during my undergrad, and I have learned many interesting things and a ton of useful lessons, especially on modeling, writing good simulation code, and implementing reproducible simulations. If you are interested in chatting about this, feel free to shoot me a message.

2. Patterns of commute flows on an urban rail network

Traditionally, people use models like the gravity model or radiation model to analyze and predict flows between origin-destination pairs. With my advisor Dr. Eduardo Lopez, we are working on using a new approach to analyze flows on networks that can add to our current understanding of how things move in a spatial setting. While the method is general, I am currently focusing on commute flows in particular.

3. Mathematical framework for measuring diversity and shared identity in a group

My collaborators at the QSIDE Institute (H. Z. Brooks, B. Sandstede, C. M. Smith, C. M. Topaz) and I are working on developing a network-based framework for measuring the degree of diversity and identity shared-ness in a group. We are also interested in seeing, through analysis of data, whether a diverse group is more effective in certain work context.