VMs: RE: Re: RE: Re: Information lost!

["Marke Fincher" <markefincher@xxxxxxxxxxxxxxxxxxxxx>]

> It would be very instructive to learn whether 1900 near equal words
> could be explained by "chance" alone.

That's what I'm saying.  They can't.

(1) Create a function which measures the similarity of neighbouring words.
(2) Apply the function to the whole VMs text and note the overall 'score'.
(3) randomise the word order and rescore.

(4) repeat (3) and see how many randomisations it takes before you
    get a similar score (or higher) than the original.

If the number of exact pairs and similar pairs were due to chance alone
then it wont take long to match the original score.

What you'll find is that neither the exact pairs, or the similar pairs are
occurring by chance.

Something is promoting the similarity of neighbouring words.

Marke

P.S.  I stopped my process after 2 million randomisations without coming
close to matching the original score.


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