Data Science Professor Helps Uncover Cosmic Mysteries of Nuclear Physics
An NJIT business school professor is teaching graduate students how to apply data science to unique fields, starting with cutting-edge problems in nuclear physics.
Dantong Yu, associate professor of business data science, is co-principal investigator for a $250,000 grant derived from a federal $5.7 million project applying data science to understand what happens when particles collide at high velocity.
Such collisions are what scientists think happened at the start of the universe. Studying them now will help scientists to "probe the internal structure and forces of protons and neutrons that compose the atomic nucleus," the U.S. Department of Energy stated.
Yu is collaborating with scientists from Fermi National Laboratory, Los Alamos National Laboratory and Massachusetts Institute of Technology. He has experience working with such experts because he was previously employed in the Department of Energy's Brookhaven National Laboratory, which is not part of the grant but is providing data for the experiment.
Their work will be tested in an experiment called sPHENIX using Brookhaven's Relativistic Heavy Ion Collider.
"This is basically cross-disciplinary research," Yu explained. "When you develop one solution from one field, you can translate or migrate it to the new field. We constantly see this."
Let data talk and drive the decision-making. That data speaks up, that data tells a story.
Data science can be applied to business management, finance, healthcare, marketing, natural sciences, social media and many other problems. Where students in NJIT's Ying Wu College of Computing focus on the technology itself, those in Martin Tuchman School of Management study it to learn which problems it can solve and how to solve them, he noted.
"We data scientists can study novel techniques and understand new machine learning models, for example graph neural networks, to sort of verify and hone their basic algorithms in one field and then translate it into another field," he continued.
"We actually look at the data with a reasonable understanding of the domain knowledge. Let data talk and drive the decision-making. That data speaks up, that data tells a story. So this is data-driven decision making. This is to me urgent and most fascinating."
In one example, lives could have been saved if more people understood data science two years ago when the COVID-19 pandemic started, Yu said. "Definitely they could have coped with this COVID-19 pandemic better, because by using data science we understand how society functions, how people actually travel from one area of interest to another area of interest, from one census block to another census block, and who actually stayed home, who has to go out to work. Then you can understand how the different groups of people cope with COVID-19 much better."
"We are training our data scientists how to interact with different domains. That's important. I think that's the reason why I am here."