Miles Efron, assistant professor, has received a Digital Humanities Research Award from Google for his project, "Meeting the Challenge of Language Change in Text Retrieval with Machine Translation Techniques."
The project will enable Efron to develop software that will let people enter a query in contemporary English, and search over English texts throughout history—from Medieval times to the present day.
"Basically, I’m looking at language change as a translation problem," said Efron. In order for older texts to be searchable, contemporary English needs to be translated into language from various historical timeframes. "For this problem we can think of English at different slices of time as different languages," said Efron. "The project will mostly involve training statistical models that assign probabilities of the translation to a word or phrase in a target English language."
The project will also look at how to display results in order to provide the user with the most probable answer to the query. "We can do this in a number of ways," said Efron, "by period, relevance, etc. We’ll have to figure out how to display results so that they are sensible for the user."
The award provides unrestricted grants for one year, with the possibility of renewal for an additional year, as well as access to Google tools, technologies, and expertise.
"This is exciting for me because my work since graduate school has been in statistical approaches to supporting search," said Efron, "but before that I was an English major. This grant lets me go back to work with stuff I like but have drifted away from. I’ve always been interested in changes in language."
The Chronicle of Higher Education has more information on the Google grants.