Varun Jog puts a modern spin on the old proverb “people are known by the company they keep” by applying advanced mathematical tools to analyze interpersonal connections within social networks.
People’s social interactions dramatically shape how they see the world. Even if someone initially doesn’t hold strong feelings about a topic, conversations with friends, colleagues and acquaintances can cause extreme viewpoints to develop.
“Community dynamics push and polarize opinions,” says Jog, who in fall 2016 joined the Department of Electrical and Computer Engineering at the University of Wisconsin-Madison as an assistant professor. Jog is also a fellow at the Grainger Institute for Engineering, where he is a faculty member in the Computational Science and Engineering thrust area.
Jog hopes to understand how this process plays out, and to develop models to predict which direction public sentiment might swing.
“Opinions evolve over time,” says Jog. “Each person is kind of independent, but once opinions start getting influenced by social networks, they start getting more extreme.”
Understanding the dynamics within social networks is much more complicated than simply scrolling through a list of someone’s Twitter followers. Picking out communities from complicated interconnected webs of interactions requires advanced mathematical techniques.
Furthermore, some people within a network exert outsized influence on a community as a whole. Identifying the opinion leaders can be useful to predict how public attitudes may form, or make marketing more effective.
“One way to market could be to give out free samples, but how do you select who receives the product?” says Jog. “I’m interested in if there’s a principled approach to finding these key people.”
Learning about networks also offers promise to the medical field. Understanding interpersonal connections can help predict how diseases spread, which is one research focus of Jog’s collaborator, Po-Ling Loh, another recent addition to the electrical and computer engineering faculty. They plan to work together on projects that play to both of their strengths as mathematically minded engineers.
Jog looks forward to working with multiple faculty members within the ECE department, as well as across campus. In fact, the opportunity to do multidisciplinary research is one of Jog’s favorite aspects of UW-Madison.
“There’s this collaborative atmosphere here. People work together a lot, even outside their area,” Jog says.
Jog also values the opportunity to train future engineers, and says teaching is probably the most satisfying thing he has ever done.
Throughout his training, Jog hasn’t limited his scope. Instead he’s always tackled challenging questions about unwieldy datasets. Jog started investigating networks during his time as a postdoctoral scholar at the University of Pennsylvania. Before then, his PhD studies at UC Berkeley focused on more abstract mathematical concepts related to information theory and geometry. In the future, he will continue to seek out interesting problems from across a wide array of disciplines.
“Eventually, I’m interested in working on questions related to genetics,” says Jog. “Information about genes makes up some of the most challenging kinds of big data.”
Author: Sam Million-Weaver