How to get past form of a Irregular Verb using WordNet

[Naveed]

Dear All,

I am looking to obtain irregular form of Verb using the WordNet. For
example, I have a verb 'bind' whose past form is 'bound'.

Is there any way I can obtain the past form of given verb using
wordnet.

Any help/pointer in this regard will be more than welcome.

Thanks

[Sandra Derbring]

Hi Naveed,

I found your problem interesting, so I did a little research. I don't know if it will be possible to solve but I was thinking that you somehow could reverse the morphy function, that takes an inflected form and makes it a lemma. There are rules to make plural forms into present ("es" becomes "e" etc.), but there is also a file with exceptions (verb.exc), and maybe you could use it and write an own function for it?

I send you some links I browsed while looking for some clues:
http://wordnet.princeton.edu/wordnet/man/morphy.7WN.html
http://wordnet.princeton.edu/wordnet/man/wndb.5WN.html
http://opensource.ebswift.com/WordNet.Net/Forum/viewtopic.php?f=2&t=9

I hope you'll be able to make something out of it. If you do, please share what you got.

Best,
Sandra

2009/8/28 Naveed <acq...@googlemail.com>

[Naveed]

Thanks for the reply. Is there any way I can get the past form a verb
using wordnet package in NLTK.???

Any body would like to answer this.

Thanks

On Aug 28, 9:28 pm, Sandra Derbring <sandra.derbr...@gmail.com> wrote:
> Hi Naveed,
>
> I found your problem interesting, so I did a little research. I don't know
> if it will be possible to solve but I was thinking that you somehow could
> reverse the morphy function, that takes an inflected form and makes it a
> lemma. There are rules to make plural forms into present ("es" becomes "e"
> etc.), but there is also a file with exceptions (verb.exc), and maybe you
> could use it and write an own function for it?
>

> I send you some links I browsed while looking for some clues:http://wordnet.princeton.edu/wordnet/man/morphy.7WN.htmlhttp://wordnet.princeton.edu/wordnet/man/wndb.5WN.htmlhttp://opensource.ebswift.com/WordNet.Net/Forum/viewtopic.php?f=2&t=9

>
> I hope you'll be able to make something out of it. If you do, please share
> what you got.
>
> Best,
> Sandra
>

> 2009/8/28 Naveed <acq0...@googlemail.com>

[Steven Bird]

2009/9/1 Naveed <acq...@googlemail.com>:

>
> Thanks for the reply. Is there any way I can get the past form a verb
> using wordnet package in NLTK.???
>
> Any body would like to answer this.

Hi Naveed -- did you see Sandra Derbring's suggestion?

http://groups.google.com/group/nltk-users/browse_thread/thread/a13ab258881359c0?hl=en

You probably want a morphological analyzer (like morphy) but which can
be run in reverse. Unfortunately NLTK doesn't have one of those yet.
Here's a starting point:

http://www.stanford.edu/~laurik/fsmbook/home.html

-Steven Bird

[Chris]

Nodebox Linguistics (http://nodebox.net/code/index.php/Linguistics)
has something like this, but its vocabulary seems limited.

>>> import en
>>> en.verb.past('swim')
'swam'

Unfortunately, "bind" doesn't appear to be in it's dictionary.

Regards,
Chris

On Sep 1, 6:54 am, Steven Bird <stevenbi...@gmail.com> wrote:
> 2009/9/1 Naveed <acq0...@googlemail.com>:

>
>
>
> > Thanks for the reply. Is there any way I can get the past form a verb
> > using wordnet package in NLTK.???
>
> > Any body would like to answer this.
>
> Hi Naveed -- did you see Sandra Derbring's suggestion?
>

> http://groups.google.com/group/nltk-users/browse_thread/thread/a13ab2...

[Shel]

Hello,

I was poking around looking for ways to generate inflected forms of a
lemma, and came across this old thread. I don't really understand
everything at the links in the conversation, and am new to this stuff,
but I would like to experiment with generating inflected forms from a
lemma.

I have put together a bunch of inflected forms labeled with POS
together with their lemmas from the Brown corpus to play around with,
and I was thinking that as a first step maybe I could do some kind of
clustering analysis that would group lemmas that have similar
inflectional patterns (Does that seem reasonable/possible?)

Unfortunately I have no idea how to actually *do* the clustering...
Was just looking at the basic documentation of nltk's cluster package
(http://docs.huihoo.com/nltk/0.9.5/api/nltk.cluster-module.html) and
haven't been able to get the example code to run.

When I try to run:
vectors = [array(f) for f in [[3, 3], [1, 2], [4, 2], [4, 0]]]

I get:
NameError: name 'array' is not defined

Any thoughts are appreciated.

Thanks,
Shel

[Shel]

Ah, it runs if I import numpy and use "numpy.array" instead of just
"array." Of course now I'm getting "cluster is not defined" when I
try to run the next line of the sample code, despite importing
nltk.cluster and referring to nltk.cluster, but will poke around some
more.