Zahra Zarei Ardestani, PhD candidate
Recalling an idyllic upbringing in a small town in Iran, Zahra Zarei Ardestani said she learned at an early age how pushing technology on a community without considering social and cultural impact can lead to unintended consequences.
When she was younger, people in the town were still largely accustomed to a traditional way of life. There were few plastic containers, so there wasn’t much trash. Homes were made with local materials. But as she got older, she noticed more and more use of plastics, and home construction that used modern, imported materials that were not always environmentally friendly. It had an impact on how the town looked, but it also affected residents’ way of life, and even their culture.
“We think that we are doing a good thing for our society, but it often doesn’t end up very well,” Ardestani said.
Seeking out a place where she could combine her background in engineering with global development, Zahra found WPI’s Social Science and Policy Studies (SSPS) department, which this year celebrates half a century of infusing the university’s hallmark project-based STEM education with a sense of doing social good.
“I really wanted to do something that involves society with technology,” she said. “It should be consistent with the culture. At the end of the day, we don’t do engineering to just do engineering—we do engineering to improve people’s lives.”
She received her master’s degree from WPI this past spring, and has stayed on as a PhD student. She is working with Robert Krueger, professor and department head of SSPS, on a three-year project to use machine learning to help communities lift themselves out of poverty.
Ardestani and Krueger will use machine learning—a form of artificial intelligence—to sift through vast amounts of data collected in several countries that allowed people to define their own levels of poverty. The clusters and patterns that emerge may be able to take indicators of poverty and transform them into economic opportunities.
“We hope that by feeding the model the data that is directly collected from people,” Ardestani said, “it will produce better results and, ultimately, solutions.”