Communications of the ACM
The Grooviest Words of Medieval Times
Gelila Tilahun created software that helps in the dating of texts from 1066 to the 1400s.
Credit: Diana Tyszko
The University of Toronto's Center for Medieval Studies is using new software to determine the dates of British documents from approximately 1066 to the 1400s.
The British did not date medieval documents, and writing styles and seals are not helpful due to the absence of original parchment. Toronto's Gelila Tilahun developed the software, which looks for popular words or phrases, and uses the Documents of Early England Data Set Project database as a source.
When given an undated text, the program aggregates the probability of occurrence of words and phrases at each time period and then estimates the date to be the time value that maximizes the aggregated probabilities.
"Some words and phrases eventually die and others continue on," Tilahun notes. "These words have their own life. It's amazing how we can decipher the date of a document based on the evolution of word usage."
The researchers plan to make the program available to users online. Tilahun says the researchers also are working on tweaking the algorithm to determine where the document is from.
From University of Toronto
View Full Article
Abstracts Copyright © 2013 Information Inc., Bethesda, Maryland, USA
No entries found