Decoding Anagrammed Texts Written in an Unknown Language and Script

Abstract

Algorithmic decipherment is a prime example of a truly unsupervised problem. The first step in the decipherment process is the identification of the encrypted language. We propose three methods for determining the source language of a document enciphered with a monoalphabetic substitution cipher. The best method achieves 97% accuracy on 380 languages. We then present an approach to decoding anagrammed substitution ciphers, in which the letters within words have been arbitrarily transposed. It obtains the average decryption word accuracy of 93% on a set of 50 ciphertexts in 5 languages. Finally, we report the results on the Voynich manuscript, an unsolved fifteenth century cipher, which suggest Hebrew as the language of the document.

Decipherment of Historical Manuscript Images

“…Finally, to realize a fully-automatic camera-phone decipherment app, we need to lift several assumptions we made in the paper. These include knowing the plaintext language and cipher system [12], [17], [18], pre-processing images to remove margins and non-cipher text, knowing the cipher alphabet size, and cipher-specific setting of segmentation parameters. Figure 12: 50 restarts of LM-GMM on Borg dataset initialized from 3-stage decipherment with random noise (+).…”

A Hidden Markov Model for the Linguistic Analysis of the Voynich Manuscript

“…Other authors, as most researchers in the past, think that the text is enciphered in some way. Hauer and Kondrak assumed that the manuscript was written in some abjad's alphabet using some transposition of letters or anagramming [38]. These assumptions are very strong and their conclusions of a relation to Hebrew have been widely dismissed.…”

A Hidden Markov Model for the Linguistic Analysis of the Voynich’s Manuscript

“…Other authors, as most researchers in the past, think that the text is enciphered in some way. Hauer and Kondrak assumed that the manuscript is written in some Abjad's alphabet using some transposition of letters or anagramming [22]. These assumptions are very strong and their conclusions of a relation to Hebrew have been widely dismissed.…”