Re: Natural Language and Perl 6
Not long ago, yary proclaimed... > This is getting more and more off topic, but if you want some lojban > pasers, start at > http://www.lojban.org/tiki/tiki-index.php?page=Dictionaries,+Glossers+and+parsers I have a translation of the Lojban grammar in Perl 6 rules sitting around somewhere, possibly on an old, dead laptop. I've been thinking about reviving that, but haven't been able to find it yet. Maybe I'll just start over. It was quite nice for working with Lojban text. See, not so off-topic after all. :)
Re: Natural Language and Perl 6
On Tue, 3 Aug 2010, Carl Mäsak wrote: Jason (>): No specific tool is best suited for natural language processing. There was apparently a time in which everyone thought that a formal grammar could clearly define any natural language, but I don't think anyone succeeded at creating a complete formal grammar for any language other than something like Esperanto. Even Esperanto is about on the same level of complexity as your regular Indo-European language. Sure, the word-formation is more regular, but the freedom in creating sentences with non-obvious antecedents and all manner of ambiguity, is just as large as in any national language. Now, had you said Lojban, I'd have believed you. :) Or maybe Ithkuil: http://en.wikipedia.org/wiki/Ithkuil :) - | Name: Tim Nelson | Because the Creator is,| | E-mail: wayl...@wayland.id.au| I am | - BEGIN GEEK CODE BLOCK Version 3.12 GCS d+++ s+: a- C++$ U+++$ P+++$ L+++ E- W+ N+ w--- V- PE(+) Y+>++ PGP->+++ R(+) !tv b++ DI D G+ e++> h! y- -END GEEK CODE BLOCK-
Re: Natural Language and Perl 6
On Sun, Aug 1, 2010 at 6:46 AM, Timothy S. Nelson wrote: >Hi. I'm wondering if any thought has been given to natural language > processing with Perl 6 grammars. > > Yes. ;) -- Aaron Sherman Email or GTalk: a...@ajs.com http://www.ajs.com/~ajs
Re: Natural Language and Perl 6
This is getting more and more off topic, but if you want some lojban pasers, start at http://www.lojban.org/tiki/tiki-index.php?page=Dictionaries,+Glossers+and+parsers -y On Mon, Aug 2, 2010 at 3:58 PM, Carl Mäsak wrote: > Jason (>): >> No specific tool is best suited for natural language processing. There was >> apparently a time in which everyone thought that a formal grammar could >> clearly define any natural language, but I don't think anyone succeeded at >> creating a complete formal grammar for any language other than something >> like Esperanto. > > Even Esperanto is about on the same level of complexity as your > regular Indo-European language. Sure, the word-formation is more > regular, but the freedom in creating sentences with non-obvious > antecedents and all manner of ambiguity, is just as large as in any > national language. > > Now, had you said Lojban, I'd have believed you. :) > > // Carl >
Re: Natural Language and Perl 6
Jason (>): > No specific tool is best suited for natural language processing. There was > apparently a time in which everyone thought that a formal grammar could > clearly define any natural language, but I don't think anyone succeeded at > creating a complete formal grammar for any language other than something > like Esperanto. Even Esperanto is about on the same level of complexity as your regular Indo-European language. Sure, the word-formation is more regular, but the freedom in creating sentences with non-obvious antecedents and all manner of ambiguity, is just as large as in any national language. Now, had you said Lojban, I'd have believed you. :) // Carl
Re: Natural Language and Perl 6
On Sun, Aug 1, 2010 at 5:46 AM, Timothy S. Nelson wrote: >Hi. I'm wondering if any thought has been given to natural language > processing with Perl 6 grammars. > No specific tool is best suited for natural language processing. There was apparently a time in which everyone thought that a formal grammar could clearly define any natural language, but I don't think anyone succeeded at creating a complete formal grammar for any language other than something like Esperanto. Modern NLP seems to be focused on (empirical based) stochastic models or other statistical models. Most languages can be used to build and represent such systems. That being said, there are sub-fields where it might help, such as stemming, POS tagging, or text generation. Perl 6 grammars seem best suited to reduce the workload to build and manipulate parse trees or lattice models. Depending on what task is at hand, this may be of no use. I am finishing a masters in NLP and after all this work, p6 grammars wouldn't have helped much. -Jason "s1n" Switzer
Re: Natural Language and Perl 6
Hello On 01/08/2010 11:46, Timothy S. Nelson wrote: > Hi. I'm wondering if any thought has been given to natural language > processing with Perl 6 grammars. I Think Perl 6 grammars can implement the most advanced parsing algorithms like Generic LR, that that will not really solve the problem of NLP. Unfortunately. Natural language is too ambiguous and needs special treatment. At least, this is my opinion on working on NLP for the Portuguese for about en years ;) Cheers > > :) > > > - > | Name: Tim Nelson | Because the Creator is,| > | E-mail: wayl...@wayland.id.au| I am | > - > > BEGIN GEEK CODE BLOCK > Version 3.12 > GCS d+++ s+: a- C++$ U+++$ P+++$ L+++ E- W+ N+ w--- V- PE(+) Y+>++ > PGP->+++ R(+) !tv b++ DI D G+ e++> h! y- > -END GEEK CODE BLOCK- > -- Alberto Simões