[Attempto] relation(var1,of,var2) with var2 inexistant.

[Kaarel Kaljurand]

Hi,

yes, seems to be a bug. I could simplify it further to "John prefers
to eat his own apple."

$ ./ape.exe -text "John likes to eat his own apple." -solo drspp

[A,B]
predicate(A,like,named(John),B)-1/2
   B
   [C,D]
   relation(C,of,E)-1/5
   object(C,apple,countable,na,eq,1)-1/7
   predicate(D,eat,named(John),C)-1/4

Best,
Kaarel

On Thu, May 31, 2018 at 4:51 PM Norbert E. Fuchs <fuchs at ifi.uzh.ch> wrote:
>
>
>
> > On 31 May 2018, at 13:54 , Pierre-Alexandre Voye <ontologiae at gmail.com> wrote:
> >
> > Working around some sentences taken from "ACE described by itself", i get a strange result : the second variable to which relation refer doesn't exist.
> >
> > Here the two sentences, I simplified :
> >
> > There are some users.
> > They prefer to use a formal method in their own natural language.
> >
> > The drs resulted :
> >
> > drspp>[A,B,C]
> > object(A,user,countable,na,geq,2)-1/4
> > predicate(B,prefer,A,C)-2/2
> >    C
> >    [D,E,F]
> >    object(D,method,countable,na,eq,1)-2/7
> >    property(D,formal,pos)-2/6
> >    relation(E,of,G)-2/9
> >    object(E,language,countable,na,eq,1)-2/12
> >    property(E,natural,pos)-2/11
> >    predicate(F,use,A,D)-2/4
> >    modifier_pp(F,in,E)-2/8
> > </drspp>
> >
> > We can see that the variable G, used in relation(E,of,G)-2/9 doesn't exist.
>
>
> Pierre-Alexandre
>
> As it seems you have found an error – possibly caused by an incorrect resolution of the anaphor "their own" – located within the compacted subordinated sentence "They prefer to use ..." (see Construction Rules, Section 3.4.4.5).
>
> This error disappears when you rephrase the sentence as
>
>         They prefer that they use a formal method in their own natural language.
>
> which is not as compact but generates the expected DRS.
>
> Almost 10 years after the publication of "ACE can be described by itself" it is no longer possible to know whether this error existed already in the version of APE that we had in 2009.
>
> Best regards.
>
>    --- nef
>
> _______________________________________________
> attempto mailing list
> attempto at lists.ifi.uzh.ch
> https://lists.ifi.uzh.ch/listinfo/attempto