Natural Language Experts to Take on Undeciphered Scripts
The National Science Foundation has awarded Kevin Knight and collaborators $1.6 million to develop computational tools to analyze now-unreadable written texts from antiquity.
Knight, a senior research scientist in the Information Science Institute’s and a research associate professor at the USC Viterbi School’s Department of Computer Science, will work with Regina Barzilay of the Massachusetts Institute of Technology Computer Science Artificial Intelligence Lab.
According to Knight, the recently approved project, called DECIPHER, takes on two problems: deciphering ancient texts using computers and training automated language translation systems without using parallel texts.
According to the proposal for the work, “Computational analysis of ancient scripts offers a better understanding of ancient cultures, and unsupervised techniques construct language connections of great interest to historical linguists.
“Applying such techniques to automated language translation offers the chance to bring many more language pairs and domains to the population at large.”
DECIPHER is funded under the American Recovery and Reinvest Act of 2009.