[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: VMs: Proving Gibberish



Zitat von 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
>

Is it really?

I mean, yeah, doublets and triplets appear to be more common in the VM than in 
natural languages, but do their frequencies really exceed the mere statistics? 
Especially when we take into account that some words are more frequent than 
others?

I'm just asking, I haven't seen the numbers.

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