Softcomputing in identification of the origin of Voynich manuscript by comparison with ancient dialects

Abstract

The Voynich manuscript is a more than 600-year-old historical manuscript. It is considered one of the most mysterious books in the world. Over the last 100 years, this book has resisted attempts to decipher its content; hence, it is written in unidentified language. Since the discovery of the manuscript, many known and unknown cryptographers have unsuccessfully tried to decipher this book. Also, many mathematical methods have been implemented to determine whether it is a fraudulent historical text or an authentic text containing valuable information. This article aims to show the use of deep learning networks and classical methods to measure the similarity between the individual characters of the alphabet and between other alphabets and Voynich. The first part of the article demonstrates the effectiveness of our method in determining the similarities between individual characters of the Voynich alphabet. In the second part, we find the similarity between the Voynich Manuscript and other individual alphabet sets (languages). In other words, this article shows another possible direction in the research of Voynich manuscript to identify the language dialect family from which Voynich manuscript can theoretically come.