21-03-2025, 01:21 PM
I thought most things where explained in the presentation.
I use theories from books on development of a Turing machine learning. It's what pop culture calls Chatgtp today. But its old news in tech world.
So I wanted to make a model on how to learn a machine to analyse unknown language. To do that I used a four digit code system + a letter. What I did next was to divide the spreadsheet in categories as the number of words increased. Then I thought that I saw a pattern; that the first letters had a meaning, like a category. Then one thing led to another, and I started to use human intuition and brute force to test names and purpose. So far the model seem to work.
And I can only repeat that this is mainly a language model building project, not a voynich dream project on what someone wish the book to be about. This is because when u are using a preprogrammed model on previous not analysised text, anything can happen, and in worst case, the model starts to produce gibberish and will be stopped.
I go as new knowledge shows, and the newest addition, 67r2, shows no Zodiac, but a calendar. Next week I will analyse the new words and use sub cross method to identity new words from 67r2. This version of 67r2 is therefore just a draft that I only use to identify thematic direction.
If anyone want my guess of that this book is about, my guess it's that it could be a school book on plants, how to plant them, where to find them and what to use them to.
And, yes, incl. water plants.
I use theories from books on development of a Turing machine learning. It's what pop culture calls Chatgtp today. But its old news in tech world.
So I wanted to make a model on how to learn a machine to analyse unknown language. To do that I used a four digit code system + a letter. What I did next was to divide the spreadsheet in categories as the number of words increased. Then I thought that I saw a pattern; that the first letters had a meaning, like a category. Then one thing led to another, and I started to use human intuition and brute force to test names and purpose. So far the model seem to work.
And I can only repeat that this is mainly a language model building project, not a voynich dream project on what someone wish the book to be about. This is because when u are using a preprogrammed model on previous not analysised text, anything can happen, and in worst case, the model starts to produce gibberish and will be stopped.
I go as new knowledge shows, and the newest addition, 67r2, shows no Zodiac, but a calendar. Next week I will analyse the new words and use sub cross method to identity new words from 67r2. This version of 67r2 is therefore just a draft that I only use to identify thematic direction.
If anyone want my guess of that this book is about, my guess it's that it could be a school book on plants, how to plant them, where to find them and what to use them to.
And, yes, incl. water plants.