W dniu 2013-06-23 12:01, Milos Sramek pisze:
> Hi,
>
> I am new in this list. I would like to find some linguists to create a
> meaningful set of grammar rules for Slovak (so that we can say that LT
> is supported in Slovak), but first I have to learn how to create the LT
> rules by myself.
>
> Since Slovak is rich in forms, I think that the following problem can
> occur quite often:
>
> It is a common error in Slovak that the pronoun 'koho' (meaning 'whom')
> is incorrectly used instead of 'čí' ('whose'). 'koho' is in this case
> used without inflection and should be replaced by a proper inflected
> form of 'čí', which depends on a matching noun.
>
> The rule to identify the incorrect text is:
> <token>koho</token>
> <token postag="S...." postag_regexp="yes"></token>
>
> where the POS tag means: a noun (S) in any gender, number and case. (the
> last 3 dots).
> (it looks like the POS tags are language specific, aren't they?)
They are.
> In the suggestion tag it is thus necessary to apply the noun parameters
> specified by the last three dots to the pronoun 'čí', so that it matches
> the noun. In the tutorial http://www.languagetool.org/development/ there
> is an example, how to do that between two verbs:
> <match no="1" postag="verb:.*perf">kierować</match>
>
> I tried:
> <suggestion> <match no="2" postag="PA...">čí</match> \2</suggestion>.
> which suggests all possible forms of 'čí', and
> <suggestion> <match no="2" postag="S....">čí</match> \2</suggestion>.
> which does not suggest anything.
>
> Is there a way how to do that for the pair noun,pronoun?
Yes, but you need to use a regular expression to change the POS tag of
the matched word (use replace patterns).
Here's an example from the Polish file:
<message>Ten wyraz powinien być w mianowniku: <suggestion>
<match no="2" postag="(adj|ppas|pact):sg:inst.*(:pos)"
postag_regexp="yes" postag_replace="$1:sg:.*nom.*:n1\.n2.*$2"></match>
</suggestion> (domyślnym podmiotem jest „to”).
</message>
Now, as you can see the postag is specified with a regular expression,
and you will need to use "S(.*)" or something like this. Then, simply
use "PA$1" in postag_replace. Of course, I am not entirely sure if all
attributes of nouns are fine for pronouns in Slovak, but if they are, it
should work just like this. If not, you need to add more structure to
your replacement patterns.
Best,
Marcin
------------------------------------------------------------------------------
This SF.net email is sponsored by Windows:
Build for Windows Store.
http://p.sf.net/sfu/windows-dev2dev
_______________________________________________
Languagetool-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/languagetool-devel
