(13-03-2024, 10:21 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view. (13-03-2024, 09:20 AM)Torsten Wrote: You are not allowed to view links. Register or Login to view.The statement "There are words in the manuscript that only appear in a certain section" and the statement "common words can always be found in other sections" are directly contradictory.
Not to my understanding. Common words are common in both senses of the word 'common'. They are frequent and they appear throughout the text.
You are trying to alter the meaning of my words.
My original statement was "if a word is common in one section, it always can also be found in other sections." in the sense that "there are no words that are frequently used in one section, and do not appear in other sections".
(13-03-2024, 10:21 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.Subject-related words are usually not among the most frequent. This is why people analysing texts, and looking for changes (in subject matter or other areas) make an effort to remove such words if they can ("stop word removal").
Neither Montemurro and Zanette nor Sterneck et. al. did remove the most frequent words or words that are equally distributed throughout all sections of the text.
(13-03-2024, 10:21 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.While indeed I don't agree with all arguments of Montemurro and Zanette, it makes the text appear as if it is subject-matter related.
If you do not agree with Montemurro and Zanette's argument, then what their argument suggests should become irrelevant.
Anyway even Sterneck et. al. reported that "The LDA model created one dominant topic ..." [Sterneck et. al.] And also Nick Pellings stated about the Montemurro and Zanette argument: "
Incidentally, M & Z conclude in their paper that results point to a semantic link between the Recipe and Astro sections, and between the Herbal and Pharma sections. Actually, had they been more aware of the codicology analyses that have been done, they would have seen that their results are consistent with the writing phase order. In fact, there are many indications that what I call Voynichese’s ‘container’ layer above evolved during the writing, with the most obvious evolution being between Currier A and Currier B" [You are not allowed to view links. Register or Login to view.].
These findings are consistent with our finding that the token frequencies suggest a continuous evolution from Currier language A to B [see Timm & Schinner 2019, p. 6]. So one possibility is "that in reality the token frequencies might undergo a continuous evolution from Currier language A to B".
Anyway, I have no doubt that in order to prevent a constructive discussion you will continue to disagree with everything I say, regardless of what I say.