[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki


[ Back to the navigation ]

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
user:zeman:interset:common-problems [2008/04/04 15:59]
zeman Refined future model.
user:zeman:interset:common-problems [2008/04/04 16:03] (current)
zeman Refined future model.
Line 52: Line 52:
  
 Chinese 的 (de) has a part of speech of its own in the Sinica treebank, ''​DE''​. The easiest approach is to decode it as a particle and remember its special nature using a new ''​subpos''​ value, or just storing it in the ''​other''​ feature. However, //​de//'​s usage could be compared to that of conjunctions. (It's not a coordinative conjunction,​ though. It connects two elements with different roles: often a possessor and the possessed object, e.g. 我的腦海.) Chinese 的 (de) has a part of speech of its own in the Sinica treebank, ''​DE''​. The easiest approach is to decode it as a particle and remember its special nature using a new ''​subpos''​ value, or just storing it in the ''​other''​ feature. However, //​de//'​s usage could be compared to that of conjunctions. (It's not a coordinative conjunction,​ though. It connects two elements with different roles: often a possessor and the possessed object, e.g. 我的腦海.)
 +
  
  
Line 94: Line 95:
 Or the alternating features would directly reference the array of substructures containing their alternating values. No additional technical feature would be needed. There could be more than one independent arrays of substructures if needed. The main structure would look the same as today, including the possibility that a feature value is an array reference. However, the referenced array would not contain directly the values of the referencing feature. Instead, it would contain (references to) substructures (hashes), each of them defining a single value for this and possibly other features. Or the alternating features would directly reference the array of substructures containing their alternating values. No additional technical feature would be needed. There could be more than one independent arrays of substructures if needed. The main structure would look the same as today, including the possibility that a feature value is an array reference. However, the referenced array would not contain directly the values of the referencing feature. Instead, it would contain (references to) substructures (hashes), each of them defining a single value for this and possibly other features.
  
-<code perl>@alt =+<code perl>@alt1 =
 ( (
     {'​gender'​ => "​masc",​ '​animateness'​ => "​inan"​},​     {'​gender'​ => "​masc",​ '​animateness'​ => "​inan"​},​
     {'​gender'​ => "​fem"​}     {'​gender'​ => "​fem"​}
 +);
 +@alt2 =
 +(
 +    {'​case'​ => "​nom"​},​
 +    {'​case'​ => "​acc"​},​
 +    {'​case'​ => "​voc"​}
 ); );
 %f = %f =
 ( (
     '​pos' ​         => "​adj",​     '​pos' ​         => "​adj",​
-    '​gender' ​      => \@alt+    '​gender' ​      => \@alt1
-    '​animateness' ​ => \@alt,+    '​animateness' ​ => \@alt1,
     '​number' ​      => "​plu",​     '​number' ​      => "​plu",​
-    '​case' ​        ​=> ​"​nom"​,+    '​case' ​        ​=> ​\@alt2,
     '​degree' ​      => "​pos",​     '​degree' ​      => "​pos",​
     '​negativeness'​ => "​pos",​     '​negativeness'​ => "​pos",​
 );</​code>​ );</​code>​

[ Back to the navigation ] [ Back to the content ]