[Corpora-List] ANC, FROWN, Fuzzy Logic

[Chris Brew]

Hi Linda and Rob,

Rob's point about language being fundamentally a corpus is 
interesting. I think there are several things at issue here.

- how should we, as scientists, proceed in trying to
  derive objective and generalizable knowledge about
  language from corpora? As Rob says, the debate 
  includes serious discussion of whether the object 
  of study is to be the corpus itself or some idealized
  extension of the same to include, for example, all the
  words that an educated speaker knows. 


- even if we reach consensus on how to do science over
  corpora (not that likely, but there is some measure
  of agreement), we might disagree on the extent to which it
  is a nice account of how language users and language 
  learners act to treat them as "little scientists" who
  process input in much the same way as we do in our experiments.

- once we have decided what to try and explain, what kind of
  models we should use? See for example 

Sarah Bunin Benor and Roger Levy. 2006. The Chicken or the Egg? A
Probabilistic Analysis of English Binomials. Language 82(2):233-278

   which tries out various options for explaining the behaviour
   of expressions like "thunder and lightning".


Chris

On Sat, Jul 22, 2006 at 01:26:21PM +0800, Rob Freeman wrote:
> Hi Linda,
> 
> Re. your third point. As far as I know fuzzy logic is just a way of keeping 
> track of uncertain qualities, it does not explain the underlying uncertainty. 
> I don't see it being of much use for a fundamental understanding.
> 
> For a better hint at a fundamental understanding I would look at this article 
> by Greg Chaitin on limits of uncertainty in all logical systems:
> 
> http://www.cs.auckland.ac.nz/CDMTCS/chaitin/cmu.html
> 
> I think Chaitin's article provides a better idea of the underlying problem 
> with language (for which the solution is to understand language to be 
> fundamentally a corpus and not a logical system of rules and classes over 
> that corpus.)
> 
> Of course this perspective of language (as fundamentally a corpus and not a 
> logical system of rules and classes over that corpus) is highly controversial 
> and not fully understood even on this list, which is why most of the debate 
> here is still about how to find or impose underlying systems of rules or 
> classes (annotation schemes, disambiguation or categorization of 
> "word-senses"), which we then face dealing with as "fuzzy" qualities.
> 
> -Rob Freeman
> 
> On Saturday 22 July 2006 05:10, Linda Bawcom wrote:
> > Dear all,
> > I'm resending the following because I have received no response, which is
> > unusual for our friendly group. My apologies if it was, in fact, received.
> > Kindest regards,
> > Linda
> >
> > Dear Friends, Colleagues and List Members,
> >
> > I need to impose upon you once again.  I have three questions:
> >
> > 1) Can the ANC be used with Wordsmith? Only a program called "Gate" is
> > listed on the web site and I don't understand enough about XML, filenames, 
> > or markups to know if the information given means it can be used with it.
> > The ANC publisher has not gotten back to me (just so you know I've done my
> > homework!).
> >
> > My problem is that my small corpus is from American newspapers, but  I used
> > the BNC as a monitor corpus and I can just hear a question arise regarding
> > that as I  sit down to defend my doctoral thesis. So I  figure I had best
> > compare my findings to an American corpus.
> >
> > 2) Try as I may on the ICAME site, I can not seem to get myself to a link
> > that would allow me to download (purchase?) FROWN (with the exception of
> > the manual). Can someone who is probably more patient send me that link?
> >
> > 3) This is a fun (?) one  for the comp. linguists and math whizzes.  Is
> > fuzzy logic ever used to measure semantic similarity, perhaps, for example,
> > with regard to color spectrum categorization? I am thinking in particular
> > of this in relation to  Vantage Theory, but would be curious about any
> > other broader use, albeit the algorithm is tweaked a bit.  If you do have
> > an answer and it needs some explanation, try to pretend you have to explain
> > it to a five year old :-).  You'll probably notice my ignorance just from
> > the question. Unfortunately, my tutorial regarding fuzzy logic with a math
> > teacher at UHD where I teach won't begin until October ( but if you don't
> > feel like making the explanation so simple, just pretend I know everything
> > and I'll have the math teacher [attempt to] explain it to me. He does
> > projects for NASA here in Houston, so I imagine he'll understand!).
> >
> > In the end, it (fuzzy logic) may be too far afield from my  doctoral
> > research regarding  (lexical) priming, lexical choice, and how we
> > categorize synonyms, but I would like to see if I should rule it out with
> > regard to empirical support.
> >
> > Kindest regards,
> > Linda Bawcom
> 

--