German Noun Decompounding Uima Annotator

Pre-requirements

• Google Web 1T 5-gram, 10 European Languages Version 1 (http://www.ldc.upenn.edu/Catalog/CatalogEntry.jsp?catalogId=LDC2009T25)
• Java
• Maven

Installation

Before you can use the component you need to index the Google Web1T corpus with Lucene. Extract all German n-grams archives and put them in one folder on your medium. In the /bin directory you will find an executable called web1TLuceneIndexer. Run:

./bin/web1TLuceneIndexer.sh 
    --web1t PATH/TO/FOLDER/WITH/ALL/EXTRACTED/N-GRAM/FILES 
    --outputPath PAHT/TO/LUCENE/INDEX/FOLDER

CAUTION: This task takes at least 5 hours and will write 25 GB of data to the medium.

The task will generate the index including all data. The extracted archives are not needed anymore.

More Information

More information can also be found in the /doc folder.