DECODE: Automatisk avkodning av historiska manuskript

Tidsperiod: 2015-01-01 till 2017-12-31

Projektledare: Beata Megyesi, Beáta Megyesi

Finansiär: Vetenskapsrådet

Bidragstyp: Projektbidrag

Budget: 6 042 000 SEK

Thousands of ciphers, enciphered historical manuscripts, are buried in libraries and archives. Our project will develop computer-aided tools for automatic and semi-automatic decoding of historical source material by cross-disciplinary research involving computer science, language technology, linguistics and philology. We will carry out basic research needed to put together a set of useful software tools, and to apply those tools to ciphers of historical interest. The project involves systematic automatic detection of various cipher types, the development of algorithms for (semi-)automatic decryption of different types of ciphers, and the creation of language models and pattern dictionaries for early variants of twenty European languages.Decoding historical manuscripts is valuable not only to historians and the public but also for computer science and language technology research through the development of new techniques for automatic analysis and decryption of ciphers, and modeling old variants of European languages. For linguistics, the project provides significantly new material, previously unknown texts hidden behind the ciphers, and new tools that can be used to analyze historical texts for different languages. The project is also step towards the (semi-)automatic decoding of older, undeciphered writing systems from antiquity, such as Linear A, Proto-Elamite, or the Indus script, puzzles that scholars have been trying to solve for decades.