Computer algorithm to decipher ancient texts

BEERSHEBA, Israel, (Reuters) – Researchers in Israel  say they have developed a computer programme that can decipher  previously unreadable ancient texts and possibly lead the way to  a Google-like search engine for historical documents.

The programme uses a pattern recognition algorithm similar  to those law enforcement agencies have adopted to identify and  compare fingerprints.

But in this case, the programme identifies letters, words  and even handwriting styles, saving historians and liturgists  hours of sitting and studying each manuscript.

By recognising such patterns, the computer can recreate with  high accuracy portions of texts that faded over time or even  those written over by later scribes, said Itay Bar-Yosef, one of  the researchers from Ben-Gurion University of the Negev.

“The more texts the programme analyses, the smarter and more  accurate it gets,” Bar-Yosef said.