id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_ia5azdhzengttdfylnfe4qbu6a	Bradley Hauer	Decoding Anagrammed Texts Written in an Unknown Language and Script	2016	12	.pdf	application/pdf	7360	593	62	Decoding Anagrammed Texts Written in an Unknown Language and Script propose a method for deciphering substitution ciphers which is based on Viterbi decoding with mapping probabilities computed with the expectationmaximization (EM) algorithm. The method correctly deciphers 90% of symbols in a 400-letter ciphertext when a trigram character language model In this section, we propose and evaluate three methods for determining the source language of a document enciphered with a monoalphabetic substitution probability of the decipherment according to a bigram character language model derived from a sample document in a given language. We now directly evaluate the three methods described above by applying them to a set of ciphertexts from different languages. The decomposition pattern method makes many fewer errors, with the correct language ranked second in roughly half of those of the resulting decipherment with both characterlevel and word-level language models. Language Text Words Characters words, their distances to the VMS language according to the decomposition pattern method are 0.159,	./cache/work_ia5azdhzengttdfylnfe4qbu6a.pdf	./txt/work_ia5azdhzengttdfylnfe4qbu6a.txt