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

Re: VMs: New computer algorithm for language learning

Wow! If jargon generation is a sign of growth in a field of endeavor, NLP
has grown a lot since I messed with it in the early 1980's :-)
There is a trial algorithm available on the Z. Solan site, might have
enough moxie to do the job. Does anyone have a Linux machine to try it?


> On Thu, 1 Sep 2005, Greg Stachowski wrote:
>> I wonder how this would work with the VMS - it's interesting, anyway.
>> ...
>> To quote:
>> "Cornell University and Tel Aviv University researchers have developed a
>> method for enabling a computer program to scan text in any of a number
>> of
>> languages, including English and Chinese, and autonomously and without
>> previous information infer the underlying rules of grammar."
> While agreeing with the stipulations from Elmar and Heike, I think it
> would be very interesting to see what this algorithm - called ADIOS or
> Automatic Distillation of Structure - makes of both the text of the VMs,
> and the set of VMs words.  Of course, assuming the material is encrypted,
> not merely encoded, something which we usually do assume, that might well
> have some influence on the applicability of the algorithm as well.
> I don't think it's claimed that this approach makes sense of the input,
> only that it produces a generative grammar of the input - a pattern
> analysis.
> Patented doesn't necessarily mean available only at high cost, and however
> it is available, I suspect the authors would be as interested in seeing
> what it made of the VMs as anyone else.  They're plainly interested in
> wide applications, even non-linguistic ones, e.g.,
> "In addition to child-directed language, the algorithm has been tested on
> the full text of the Bible in several languages, on artificial
> context-free languages with thousands of rules and on musical notation. It
> also has been applied to biological data, such as nucleotide base pairs
> and amino acid sequences. In analyzing proteins, for example, the
> algorithm was able to extract from amino acid sequences patterns that were
> highly correlated with the functional properties of the proteins."
> Contacts:
> http://www.tau.ac.il/~zsolan/
> http://kybele.psych.cornell.edu/~edelman/adios-nips-workshop.pdf
> (Or search on adios horn ruppin.)
> Perhaps one or more of the more notable members of the Voynich fraternity
> would like to write to the ADIOS project?
> ______________________________________________________________________
> To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
> unsubscribe vms-list

Dr. Don Latham
17850 Six Mile Road
POB 134
Huson, MT, 59846
VOX 406-626-4304
FAX 406-626-5979

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