At 10:30 PM +0200 1/31/08, Avril Styrman wrote:
Lainaus Pat Hayes
<phayes@xxxxxxx>:
> >All the parsed results can be compared, and
> >probablilities given to each individual X and Y in
> >"X is Y". And this is exactly the same as
making
> >a 'genuine' ontology with the subclass relations.
>
> No, it is absolutely nothing like making a genuine ontology.
We have reached the not-too, not-too situation.
But - and I'm sorry to pull rank on you here, but I don't have
time to do anything more collegial - I do know something about what Im
talking about, and it is quite apparent that you don't. (Attended one
seminar?? Give me a break. If you had read one textbook I might be
more impressed. NLP isn't my main field, but I work alongside people
who wrote most of the main textbooks, and Ive chaired more research
meetings on NLP and Meaning than I can now recall.)
> >The is-a relationship is a part
of the solution.
>
> No, the problem is to find the ontological content in the free
NL
> text. Such text contains all sorts of information, some of which
can
> be possibly expressed as subclass relations, and a much
smaller
> fraction of which can be usefully so expressed. But there are
no
> simple heuristics for reliably recognizing such information based
on
> the form of the NL sentences.
If is-a is not one sort of ontological content, then what
is?
Of course subclass information is ontological content. Nothing I
have said even suggests that I would say it was not.
A large number of simple "X is Y"
NL sentences gives a quite
good approximation that "X is a subclass of Y"
OK, if you believe this, find me some. Actual examples from the
Web.
Here are a few sentences I found starting from the Google news
page that match your pattern.
He is
a great American hero and an extraordinary leader.
California is the largest prize among the 24
states that hold nominating contests on "Super Tuesday,"
February 5
The Air is 0.16 of an inch thick at its thinnest point.
Activiation
of the phones is crucial to Apple's bottom line because it gets
a 10 percent of the cut in monthly fees charged by
AT&T.
Calendar
year '08 is what Steve referenced in his keynote.
The point
that he made was that the worldwide market for total cell phones
is somewhere around 1 billion and our objective of getting 1% of
it would yield 10 million units across the calendar year,
A price
cut is a possibility, agrees Mike Abramsky, an analyst with RBC
Capital
, similarly as
using Google as means to separate which one is the right word:
'successor' of 'sucesor'. Of course there are more complex
tasks too.
> I'm sorry, but it is clear that you know almost nothing about
the
> actual NLP field. There is no point in continuing this
discussion.
> Read some textbooks on the state of the art on NL
comprehension
> systems.
> >PNL is a general scheme. It does not solve all the
> >problems of AI, but it is a good suggestion of
> >the general guidelines.
>
> The general guidelines for solving all the problems of AI??
I'm
> afraid I am simply laughing at this point.
The general guideline is: use web as the source of information
instead of hard-coding it, which is done with Cyc.
That is not a research guideline. If you knew anything about AI
you not make such foolish observations.
> >One of Zadeh's ideas was to
convert NL into PNL.
>
> And the best of luck to you and to Lotfi in actually managing to
do
> this.
It doesn't take a Phd in NLP to understand that NLP is
the
the right way to do the
conversion.
NLP is not a method or a technique, it is the name of an open
research problem area.
I've only been to one
seminar about NLP, but reading many books doesn't change
the points made above.
It might inform them past the point where they are totally
disconnected from the real world, however.
The main point was the web is
more
likely to be the repository of common
sense than Cyc.
This is false on many grounds. First, common sense is not
commonly written down, even in English. Some of it may *never* be
written down because we all learn it (or are born with it) before
becoming literate. Second, it is not currently or for the forseeable
future possible to extract meaning from free NL text without already
having a CYC-scale ontology to help one comprehend the NL sentences
with. Third, although there may indeed be a great deal of information
on the Web, there is no way to distinguish facts from opinion or sense
from nonsense, unless one already has a great deal of common sense
(more than many adult humans) to help one do the sorting out.
And
this doesn't change the fact that the internal model
of a NL parser can be called an ontology
too.
You can call a horse a car, but that has nothing to do with the
facts. Parsers do not include ontologies.
Pat
Avril
--
---------------------------------------------------------------------
IHMC(850)434 8903 or (650)494 3973 home
40 South Alcaniz St.(850)202 4416 office
Pensacola(850)202 4440 fax
FL 32502(850)291 0667 cell
phayesAT-SIGNihmc.us
http://www.ihmc.us/users/phayes
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Subscribe/Config: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (01)