Hometown: De Soto, Kansas
Program: Machine translation of languages
Research Group:Dan Gildea
Why did you pick Rochester to do your graduate work in Computer Science?
I chose Rochester because my area is natural language processing, and Rochester had the best mix of experts in that area as well as in adjacent areas, such as linguistics and brain/cognitive science.
How did you choose your area of research?
I knew when I applied that I wanted to work on machine translation. That hasn't changed since arriving here, though I have focused more specifically on unsupervised translation.
I'm working on ways to enable computers to translate between human languages without needing millions of example professional translations by learning each language independently and then figuring out how to go between them. I developed a system that can use spelling information and cognates to improve translation for related languages, improving accuracy by around 10%. This paper was accepted for publication by the Association for Computational Linguistics and I had the opportunity to present it in Melbourne, Australia.
What have you enjoyed most about your time in the CS department at Rochester?
I really enjoy the close-knit atmosphere of everyone in the department; the grad students regularly hang out together, which has helped me preserve a work-life balance. I am also involved in the community as an assistant coach for a local FIRST Lego League robotics team of 3rd-5th graders. I've found that URCS is the perfect size for working on a PhD: large enough that I can find an expert on whatever problem I want to work on, but small enough that I can know almost everybody.
My current plan is to enter industry after graduation, though I have not completely written off entering academia.