(24-04-2025, 09:25 AM)oshfdk wrote: You are not allowed to view links. Register or Login to view.Hi and welcome!
requirements.txt is missing from the repository
Could you give some high level plaintext description of what this is? I found your other posts that explains the idea: You are not allowed to view links. Register or Login to view.
But I'm not sure I understand it. Is the analysis based on word token distributions?
Also, how was this seed 1405 selected?
Hi, and thanks for your message!
I’ve just added the requirements.txt file to the repository, so the full pipeline is now reproducible and ready to run:
You are not allowed to view links. Register or Login to view.
What the project does (in plain terms)
This pipeline tests whether the Voynich Manuscript’s botanical section exhibits a statistically significant cyclical structure.
It does not assume the text is a phonetic language yet, but treats the glyph sequences as symbolic patterns, and applies:
- Token distribution analysis
- Lexical entropy per folio
- Topic modeling (LDA)
- Supervised classification of four synthetic “phases”
- Permutation tests and ablations to verify the structure
The cyclical pattern is robust and reproducible — it emerges even when phase labels are shuffled or angle features are removed (and disappears when phase order is broken). This confirms that the structure is real, not random. Why is this happening? Who knows...
The agronomic interpretation — linking these phases to sowing, growth, flowering, and harvest — is still a hypothesis. However, the fact that it correlates with known plant phenology and historical agricultural calendars gives this hypothesis strong support, thanks to the statistical foundation.
Why seed
1405
?
[font=.AppleSystemUIFont]The seed 1405 is arbitrary and used solely to ensure full reproducibility. Any other fixed number would work — this one was chosen to match the year 1405, which aligns with the possible calendar framework used in the analysis.[/font]
Let me know if you’d like any details or to try running the model. Happy to collaborate or improve it further!
Best,
O. Cho