project consists of German, English, and Romanian Lexicons, and of the bilingual connections between them. The lexicons are encoded in such a way that can be used in several application types (Machine Translation, etc). The first main purpose of these lexicons is to be used in student projects. One of the requirements of these lexicons is to have a standard structure, so that they fit in the actual stage of Natural Language Processing (NLP) applications. The lexicons contain only lexemes. The structure chosen is a slightly modified PAROLE/SIMPLE
According to the starting purpose of the lexicons, G.E.R.L.
contains the following information
The starting point in creating the G.E.R.L. structure is the PAROLE/SIMPLE DTD
. The original structure was simplified according to the specification needed, and several features were added so that some problems, that appeared due to the Romanian language, are solved (e.g. the definite article is added at the end of the word). The G.E.R.L. structure is composed of a morphological layer, a syntactic layer, a semantic layer and a multilingual one (as SIMPLE/PAROLE structure). The G.E.R.L. structure can be seen here
, some GUI Snapshots of the tool for adding, deleting, updating, and consulting lexicon entries can be found.
you can find existing publications about G.E.R.L.
To be updated
26 Aug 2005