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

Re: VMs: Hamptonese

Has anyone tried any of Sukhotin's algorithms besides
the vowel/consonant classification? Specifically the
aggregation one he mentions? He coins a term for
"internal stability" which, to me, seems like it
should be inverted.

Pardon if I'm on an old subject that's been
investigated - I'm reading through the archives but
haven't got through 1992 yet (there was a lot of good
stuff in that year! doesn't seem like we've gotten far
since then...).

> 	VMs relevance.  He has used a Hidden Markov Model
> (HMM) to analyze Hamptonese.  I don't recall that
> anyone has tried this on the VMs.  I myself don't
> have
> experience with HMMs.  
> 	In the paper he discusses what the HMM showed for
> single glyphs.  It seems to show that there are in
> fact
> no vowels and consonants in Hamptonese.  The
> Sukhotin
> algorithm shows vowels and consonants in Hamptonese,
> but Jacques reminded me that it does not find
> "vowels";
> it separates symbols into two sets, the member of
> each
> tending to occur next to the members of the other. 
> But
> it seems to me that is also what the HMM would find!
> The results aren't completely unambiguous, so I am
> thinking it over.  

Do you Yahoo!?
Yahoo! Tax Center - File online by April 15th
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list