RE: VMs: Proving Gibberish

["Marke Fincher" <markefincher@xxxxxxxxxxxxxxxxxxxxx>]

...sorry, misunderstanding there.
My example was for a specific word (w) with probability (1/p).
So the probability of a double of that _particular_ word (w.w) is 1/(p*p).

If you take most common words from the VMS and assume it's probability is
p=NumInstances/TotalWordsInSample, the expected proportion for it's double
would be 1/(p*p)  {assuming a random independent process creates the words}.
But in practice the number of doubles is usually way over this, and even
more so for the triples.

Marke

-----Original Message-----
From: owner-vms-list@xxxxxxxxxxx [mailto:owner-vms-list@xxxxxxxxxxx]On
Behalf Of elvogt@xxxxxxxxxxx
Sent: 17 August 2004 13:33
To: vms-list@xxxxxxxxxxx
Subject: RE: VMs: Proving Gibberish


Zitat von Marke Fincher <markefincher@xxxxxxxxxxxxxxxxxxxxx>:

> ...
> If the grille WAS moved randomly over the table I don't think it would
> produce
> the number of exact word repetitions we see.  If the process that
generates
> one word is identical but independent to the generation of neighbouring
> words
> the expected proportion of doubled words would be 1/(p*p) and tripled
words
> 1/(p*p*p) (given the chance 1/p of generating a particular word).

Uhm... I disagree. If you've got a vocabulary of p words (all of equal
frequency), then the probability for any particular word to show up in a
given
spot is 1/p, and thus the probability for any sport to be occupied by the
same
word as the previous spot is 1/p as well.

Uneven frequencies would certainly shift this value. (Off the top of my head
I'd assume it'd raise the probability for word doublets, but I'm not sure
about
this.)

But of course this is somewhat moot, because AIUI Rugg never claimed to have
completely random tables/grilles, but rather modified them to fit VM
characteristics.

Cheers,

   E.


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

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