Method · NLP (Creutz & Lagus)Unsupervised morphology (Morfessor Baseline)
How it works
Compared with: Latin of the manuscript's genres, slot-grammar text and the manuscript shuffled across pages, all in the same lines.
Runs and results 10 runs · 6 shown
| Run | Mystery | Measured | Result | Against | Date |
|---|
| RUN-200 | Voynich | Share stems with 3plus endings | 0.179 | Latin 0.176 · slot 0.155 · shuffled 0.178 | 22 Oct |
| RUN-199 | Voynich | Paradigm mean endings | 3.11 | Latin 1.96 · slot 2.38 · shuffled 3.16 | 22 Oct |
| RUN-198 | Voynich | Morphs per word | 2.35 | Latin 1.9 · slot 2.83 · shuffled 2.32 | 22 Oct |
| RUN-197 | Voynich | Morph types | 673.0 | Latin 4314.0 · slot 573.0 · shuffled 680.0 | 22 Oct |
| RUN-196 | Voynich | Types | 7473.0 | Latin 10069.0 · slot 12685.0 · shuffled 7473.0 | 22 Oct |
| RUN-100 | Voynich | Share stems with 3plus endings | 0.181 | Latin 0.167 · slot 0.168 · shuffled 0.193 | 22 Oct |
Applied to
3 datasetsExploration split, Latin genre texts, ZL3b transcription
