MU engineer uses machine learning to translate historical script

Researchers receive grant to devise a way to decipher and digitize 17th century records.

Sept. 1, 2020

Praveen Rao

Praveen Rao

When Viviana Grieco, associate professor in history and Latin American and Latinx Studies at the University of Missouri-Kansas City, needed help decoding a collection of 17th century notary records from Argentina, she collaborated with a University of Missouri engineer.

Grieco and Praveen Rao, an associate professor of electrical engineering and computer science and health management and informatics at Mizzou, are using machine learning to translate the texts. The two researchers received a one-year, $100,000 grant from the National Endowment for the Humanities to devise a way to decipher and digitize the collection.

“Our goal is to take this archive, which contains images of scanned, handwritten records, and make it into something meaningful for historians and students,” Rao said. “And we want to do that in a machine-driven manner.”

Grieco and Rao worked together when Rao was a faculty member at UMKC. There, they received seed money to get this project started.

Rao relocated to Mizzou early this year, but the University of Missouri System’s focus on collaborative research helped keep the team together.

Rao said he hopes the project inspires more faculty in computer science and engineering to collaborate with social sciences such as history.

“I wouldn’t have thought five years ago that I would be looking at historical collections,” he said. “But when an opportunity comes to work on a fun and interesting problem, you jump on it. I think you’re going to see more of this in the future. Our UM System, as a whole, supports that.”

Read more from the College of Engineering

Subscribe to

Show Me Mizzou

Stay up-to-date with the latest news by subscribing to the Show Me Mizzou newsletter.