NADIA

From UNL Wiki
(Difference between revisions)
Jump to: navigation, search
(Structure)
(Requisites)
 
(5 intermediate revisions by one user not shown)
Line 4: Line 4:
 
The project NADIA has two main goals:
 
The project NADIA has two main goals:
 
#To review UNL-NL mappings provided in generation dictionary projects (such as [[MIR]]).
 
#To review UNL-NL mappings provided in generation dictionary projects (such as [[MIR]]).
#To provide morphological (such as gender, number, inflectional paradigm, etc) and syntactic information (transitivity, subcategorization frame, etc) to natural language entries derived from generation dictionaries
+
#To provide morphological information (such as gender, number, inflectional paradigm, etc) and syntactic information (transitivity, subcategorization frame, etc) to natural language entries derived from generation dictionaries
  
== The repository ==
+
== Repository ==
NADIA is language dependent. Every language has its own set of entries to be addressed. The list of entries is derived from [[GD|generation dictionaries]].
+
NADIA is language dependent. Every language has its own set of entries to be addressed. The list of entries is derived from the corresponding MIR projects, as indicated below:
  
== Structure ==
 
NADIA is divided into 6 different subprojects according to the source of the entries.
 
 
{|border="1" align="center" cellpadding="2"
 
{|border="1" align="center" cellpadding="2"
 
!Subproject
 
!Subproject
Line 34: Line 32:
 
|}
 
|}
  
== Methodology ==
+
== Requisites ==
NADIA is open to all languages, except to English, for which it is expected to be already finished. As a derivative project, NADIA depends on the results of MIR, and it is open only when the language reaches at least 90% of the corresponding level in MIR<ref>For instance, NADIA-A1 for French is open only after French achieves 90% in MIR-A1.</ref>. As in any NLization project, users are expected to provide the features for natural language lemmas that were linked to UW's.  
+
NADIA is open to all languages, except to English, for which it is expected to be already finished. As a derivative project, NADIA depends on the results of MIR, and it is open only when the language reaches at least 90% of the corresponding level in MIR<ref>For instance, NADIA-A1 for French is open only after French achieves 90% in MIR-A1.</ref>:
 +
*[[MIR-A1]] is required for NADIA-A1;
 +
*[[MIR-A2]] is required for NADIA-A2;
 +
*[[MIR-B1]] is required for NADIA-B1;
 +
*[[MIR-B2]] is required for NADIA-B2;
 +
*[[MIR-C1]] is required for NADIA-C1;
 +
*[[MIR-C2]] is required for NADIA-C2.
  
 +
== Instructions ==
 +
In NADIA, users are expected to provide the features for natural language lemmas that were linked to UW's in the project MIR, according to the following instructions:
 +
#Entries must be treated only if lexical units. If the lemma deriving from MIR is not a lexical unit, users are expected to report the problem and not to provide any information.
 +
#Entries must be treated only if lemmas. If the entry deriving from MIR is not a lemma, users are expected to report the problem and not to provide any information.
 +
#Entries must be corrected, in case of spelling problems. If the entry deriving from MIR has spelling problems, users are expected to correct it and to treat it normally.
 +
#Entries must be addressed thoroughly. Users are expected to provide all the morphological, syntactic and semantic information concerning each entry.
 +
#Entries must be split, in case they have different morphosyntactic behavior (gender, number, inflectional paradigm, subcategorization frame).
  
 
== Notes ==
 
== Notes ==
 
<references />
 
<references />

Latest revision as of 16:23, 24 September 2013

The project NADIA (NAtural language DIctionary for UNL-NL mAppings is devoted to the creation of NL dictionaries for entries derived from generation dictionaries.

Contents

Goal

The project NADIA has two main goals:

  1. To review UNL-NL mappings provided in generation dictionary projects (such as MIR).
  2. To provide morphological information (such as gender, number, inflectional paradigm, etc) and syntactic information (transitivity, subcategorization frame, etc) to natural language entries derived from generation dictionaries

Repository

NADIA is language dependent. Every language has its own set of entries to be addressed. The list of entries is derived from the corresponding MIR projects, as indicated below:

Subproject Source
NADIA-A1 MIR-A1
NADIA-A2 MIR-A2
NADIA-B1 MIR-B1
NADIA-B2 MIR-B2
NADIA-C1 MIR-C1
NADIA-C2 MIR-C2

Requisites

NADIA is open to all languages, except to English, for which it is expected to be already finished. As a derivative project, NADIA depends on the results of MIR, and it is open only when the language reaches at least 90% of the corresponding level in MIR[1]:

  • MIR-A1 is required for NADIA-A1;
  • MIR-A2 is required for NADIA-A2;
  • MIR-B1 is required for NADIA-B1;
  • MIR-B2 is required for NADIA-B2;
  • MIR-C1 is required for NADIA-C1;
  • MIR-C2 is required for NADIA-C2.

Instructions

In NADIA, users are expected to provide the features for natural language lemmas that were linked to UW's in the project MIR, according to the following instructions:

  1. Entries must be treated only if lexical units. If the lemma deriving from MIR is not a lexical unit, users are expected to report the problem and not to provide any information.
  2. Entries must be treated only if lemmas. If the entry deriving from MIR is not a lemma, users are expected to report the problem and not to provide any information.
  3. Entries must be corrected, in case of spelling problems. If the entry deriving from MIR has spelling problems, users are expected to correct it and to treat it normally.
  4. Entries must be addressed thoroughly. Users are expected to provide all the morphological, syntactic and semantic information concerning each entry.
  5. Entries must be split, in case they have different morphosyntactic behavior (gender, number, inflectional paradigm, subcategorization frame).

Notes

  1. For instance, NADIA-A1 for French is open only after French achieves 90% in MIR-A1.
Software