Re: CYK & Context-Free Expressions

Rock Brentwood <markwh04@xxxxxxxxx> writes:

On Apr 26, 11:14 am, Chris F Clark <c...@xxxxxxxxxxxxxxxxxxxx> wrote:
If you can show how to create a derivation tree using your model, then
you can call that process "parsing",

Don't you think it's a little disingenuous to reply with a suggestion
directly to the very thing being suggested?

I don't understand that sentence. However, my intention is not to be
disingenious. I have tried to understand your postings, but find them
too terse and notationally dense (particularly with notations and
terms) that I have no prior background in. Then, you draw
conclussions that seem contrary to material that I do think I
understand and you state that the common definitions upon which the
results I understand are based are flawed.

The particular example the is most striking is your remarks on the
complexity of "parsing". If I understand you right, you have made
remarks about being able to parse in linear time (over what I assume
are general context free grammars), and yet at the same time refer to
the n**3 bound for Earley parsing. I'm assuming that this is because
there is something buried in your works that has escaped me.

So, what I'm trying to understand is how you can make both claims.
And, in particular, what your definition of "parsing" is that it
allows you to make both statements and how (or whether it covers the
aspects of "parsing" that are relevant to me). In particular, I am
interested in "parsing" as it imposes a structure upon an input
sentence, because that is what allows one to take unstructured input
and structure it--the structure that is relevant for me is a
derivation tree, since that tree captures the nature of the inputs I
wish to stucture.

There are lots of examples
like that in the archive you're discussing!

What archive is this? If you mean the web pages pointed to in your
previous post, I followed the ones that reported the results you
published at the RelMiCS conference. However, I only found only what
appeared to me to be slide sets with little explanatory material.
And, perhaps, for the researchers you were sharing the results with
the notation was obvious. It wasn't for me. If I recall correctly, I
got to about slide 3 before I came to a slide with notation that
wasn't clear and for which I could find no explanation. Somewhere
between the 6th and 15th page (I can't recall how far I got before
coming to the point I had to give up), the notation became completely
inscrutable to me due to the notation that had not been explained and
for which I could not guess a reasonable interpretation.

If you want, I can come up with specific examples. Perhaps, if you
explain enough of the notation, I will be able to follow the points
you are making, and your results will be as obvious as you claim.
That would be a welcome result.

Sincerely,
-Chris

******************************************************************************
Chris Clark Internet: christopher.f.clark@xxxxxxxxxxxxxxxxxxxxxx
Compiler Resources, Inc. or: compres@xxxxxxxxxxxxx
23 Bailey Rd Web Site: http://world.std.com/~compres
Berlin, MA 01503 voice: (508) 435-5016
USA fax: (978) 838-0263 (24 hours)
------------------------------------------------------------------------------

.

Relevant Pages

  • Re: Val() versus CLng, CInt or CByte
    ... common usage of Val it seems to me), you might be surprised by the returns when the string happens to look like an exponential notation. ... when parsing a character string which does "look like" exponential notation. ... I think Donald was referring to surprise someone might get parsing a set of code numbers where the digits in front of a letter were being sought and where the code numbers were something like this... ...
    (microsoft.public.vb.general.discussion)
  • Re: About MyClass#my_method
    ... > I always see the # notation in the docs to design instance_methods. ... class Test end ... The def keywork would allow to make the distinction while parsing. ...
    (comp.lang.ruby)
  • Re: Val() versus CLng, CInt or CByte
    ... common usage of Val it seems to me), you might be surprised by the returns when the string happens to look like an exponential notation. ... parsing a character string which does "look like" exponential notation. ... I think Donald was referring to surprise someone might get parsing a set of code numbers where the digits in front of a letter were being sought and where the code numbers were something like this... ...
    (microsoft.public.vb.general.discussion)
  • Re: What is the convention for Command Prompt window arguments
    ... I think your earlier suggestion to check both makes sense ... Well, I don't know that it is written in stone anywhere, but the notation ... You may want to try googling for it. ...
    (microsoft.public.dotnet.languages.vc)
  • Re: C1 Notation
    ... > Upon printing after editing a file I now get a C1 notation in the ... > upper left corner of a slide with a plain page behind the slide with ... Are you printing slides, handouts, or notepages? ...
    (microsoft.public.powerpoint)