+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Analysis of the text (https://www.voynich.ninja/forum-41.html)
+--- Thread: Hypervector Analysis of the Voynich Manuscript (/thread-3314.html)
RE: Hypervector Analysis of the Voynich Manuscript - MarcoP - 10-08-2020
I hope I'll be able to comment more in the next few days. Another very simple addition to this interesting experiment could be having two distinct files for Currier A and B (apologies if this has been mentioned already).
RE: Hypervector Analysis of the Voynich Manuscript - -JKP- - 10-08-2020
Darrin, this is what I would suggest...
Simply eliminate uppercase EVA-A, EVA-O, and EVA-Y. They are very rare and simply not needed for a full-text analysis.
The distinction between EVA-s and EVA-S is important. It needs to be retained. As an example, you could map EVA-S to EVA-u which is also very rare and not needed. Then you have both.
Give me 10 minutes, let me see if I can come up with something rational and unobtrusive...
(10-08-2020, 07:43 AM)MarcoP Wrote: You are not allowed to view links. Register or Login to view.
I hope I'll be able to comment more in the next few days. Another very simple addition to this interesting experiment could be having two distinct files for Currier A and B (apologies if this has been mentioned already).
Yes, good idea. I hadn't even thought that far ahead yet.
RE: Hypervector Analysis of the Voynich Manuscript - -JKP- - 10-08-2020
Okay. Here are some suggestions. What you do is, of course, up to you.
I marked the rare and very rare glyphs with slashes. Some only occur once or twice. It would not significantly change statistics if they were dropped.
The important ones need to be moved into available slots. The most important ones are EVA-S and the benched gallows. There's enough room for them if you move them to lowercase slots where there are rare glyphs.
I tried to change things as little as possible. This does not shuffle the chart, it's the simplest "fix", and it's not really a fix because they are in slots that may not have anything to do with actual character mappings, but at least you don't lose important parts of the Voynich glyph set.
P.S. FYI, in Voynichese, there are numerous instances of eee, eeee, iii, iiii. These are not scribal errors. The Takahashi transcription does not always record them correctly.
RE: Hypervector Analysis of the Voynich Manuscript - dvallis - 10-08-2020
(10-08-2020, 07:21 AM)-JKP- Wrote: You are not allowed to view links. Register or Login to view.I've been giving this some thought.
Would you like me to post an alternate mapping that would enable you to collapse to lowercase without these problems?
Yes, that would be helpful. If you upload a file it I'll run it and post results.
RE: Hypervector Analysis of the Voynich Manuscript - dvallis - 11-08-2020
JKP
Here's my interpretation of your last post for conversion to lower case. Order of operation matters. Let me know if you think it's correct.
Start with TT EVA text
Delete j,u,v,z,'
Delete A,O,Y
E -> e
F -> u
H -> h
I -> c
K -> z
P -> v
S -> j
T -> t
I'll work up a script and post a file with this conversion for review.
RE: Hypervector Analysis of the Voynich Manuscript - dvallis - 11-08-2020
Well, DAMN IT.
The TT EVA "all pages" text You are not allowed to view links. Register or Login to view. is all in lowercase. There's not a single capital letter in there. (?!) That's what I fed into the original comparison after removing folio numbers, periods etc.
<f1r>
<f1r.P1.1;H> fachys.ykal.ar.ataiin.shol.shory.cthres.y.kor.sholdy
<f1r.P1.2;H> sory.ckhar.or.y.kair.chtaiin.shar.are.cthar.cthar.dan
<f1r.P1.3;H> syaiir.sheky.or.ykaiin.shod.cthoary.cthes.daraiin.sa
<f1r.P1.4;H> ooiin.oteey.oteos.roloty.cth*ar.daiin.otaiin.or.okan
<f1r.P1.5;H> dair.y.chear.cthaiin.cphar.cfhaiin
<f1r.T1.6;H> ydaraishy
However, if you click on the individual page for You are not allowed to view links. Register or Login to view. it obviously DOES have capitals.
[font=EVA Hand 1]fachys ykal ar ataiin Shol Shory cThres ykos Sholdy[/font]
[font=EVA Hand 1]sory cKhar or y kaer chtaiin Shar are cThar cThar dan[/font]
[font=EVA Hand 1]syaiir Sheky or ykaiin Shod cThoary cThes daraiin sy[/font]
[font=EVA Hand 1]doiin oteey oteos roloty cTh*ar daiin otaiin or okan[/font]
[font=EVA Hand 1]dair y chear cThaiin cPhar cFhaiin ydaraiShy[/font]
[font=EVA Hand 1]If anyone has a clean version of TT with all caps intact please post it. Otherwise I'll be downloading, and combining the pages.[/font]
RE: Hypervector Analysis of the Voynich Manuscript - dvallis - 11-08-2020
So, I grabbed all the properly capitalized TT files, combined them to a master, normalized (including convert to lower case) and re-plotted. The hypervector is still in the middle of Caucassian languages, which is good. Matches the original data.
I'll use the new capitalized file as a baseline, make the changes as suggested in previous posts and re-plot.
![[Image: Rplot056.jpeg]](http://www.vallisresearch.com/wp-content/uploads/2020/08/Rplot056.jpeg)
RE: Hypervector Analysis of the Voynich Manuscript - -JKP- - 11-08-2020
Sorry, I've been busy with work.
I'm glad you were able to find files with the proper uppercase EVA transliteration.
It's also interesting to see that the VMS still ends up in the same general area on the graph, even though the distinction between EVA-s and EVA-S is pretty important. I guess proportionally the corrections weren't enough to shift it. It's still worth the pain to hunt up the other files. The better the data going in... and you know the rest.
RE: Hypervector Analysis of the Voynich Manuscript - ReneZ - 11-08-2020
(11-08-2020, 05:37 AM)dvallis Wrote: You are not allowed to view links. Register or Login to view.Well, DAMN IT.
The TT EVA "all pages" text You are not allowed to view links. Register or Login to view. is all in lowercase. There's not a single capital letter in there. (?!) That's what I fed into the original comparison after removing folio numbers, periods etc.
...
However, if you click on the individual page for You are not allowed to view links. Register or Login to view. it obviously DOES have capitals.
...
[font=EVA Hand 1]If anyone has a clean version of TT with all caps intact please post it. Otherwise I'll be downloading, and combining the pages.[/font]
Apart from the (lack of) capitalisation, there are minor differences throughout the two versions, as Takeshi has been fixing some errors. The capitalisation is trivial, because all cases of cTh (etc) can be mapped directly to cth (etc) without loss or gain of information.
The perceived problem with Eva-s vs. Eva-sh is not a real problem, especially when one looks at trigrams.
It is exactly the same as in English, where the s in 'she' is a different s than the one in 'son'.
However, English also has words like 'mishap', where the s is like in 'son'. This problem does not happen in Voynichese.
RE: Hypervector Analysis of the Voynich Manuscript - ReneZ - 11-08-2020
Darren, you might want to look at this tool:
You are not allowed to view links. Register or Login to view.
you just need to build the executable from the C source code.
To use all the capabilities will require some learning curve, but to strip the standard transliteration files of all the trimmings and keep the text only, one can use:
ivtt -x7 <infile >outfile
These input files are found through links in the second last columns of this table:
You are not allowed to view links. Register or Login to view.