Re: VMs: word database and binomial distribution

[Knox Mix <knoxmix@xxxxxxxxxxxxx>]

Thanks, looks like a good place to play.

We must be doing something differently. Compare my plot with yours.

http://home.earthlink.net/~knoxmix/id21.html

I have not tried to determine why this list gives a less extended right 
leg than most of the meaningful documents I have looked at. The Towneley 
Plays and Liber Salomonis are exceptional so perhaps the VMS is, also. 
However, I am not able to compare the curves mathematically. A true 
binomial distribution of unique words could mean something entirely 
different than an apparent but not binomial distribution.

Ciao ...... Knox


Eric wrote:

> First, for all of you text analysis people out there,
> I found this database of English words (in case not
> already known):
>
> http://www.psy.uwa.edu.au/mrcdatabase/uwa_mrc.htm
>
> There is an online interface to the database and you
> can download the whole database and search
> application. It is very extensive and includes useful
> items like # of syllables, # of phonemes and phonemic
> syllabic transcription.
>
> I found it while playing around with ideas of
> investigating syllabic and phonemic properties of
> English text, especially in regards to what Stolfi had
> reported years ago about the VMS word length:
>
> http://www.dcc.unicamp.br/~stolfi/voynich/00-12-21-word-length-distr/
>
> I was wondering perhaps if that was syllabic or
> phonemic in structure (all VMS phonemes written as two
> characters or all syllables being two characters for
> instance). While plotting those, I also plotted word
> length and... I got a binomial plot for it???. I used
> the summary data from this page:
>
> http://www.psy.uwa.edu.au/mrcdatabase/mrc2.html#NLET
>
> There are some anomolies of the data in the database
> (search for all one letter words for instance), but it
> doesn't look systemic. The plot is much sharper than
> what Stolfi shows for "English" and "Latin" - maybe
> because the database has a much larger sample of
> words. My first reaction is to take that to mean the
> binomial nature of the VMS pointed out by Stolfi is
> unique only in that it occurs in a small data sample.
>
> Thanks,
> Eric

______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list