This project provides an Arabic Dependency Parser trained on data processed by Farasa toolkit using MIT RBGParser v1.1. For more details about the RBGParser v1.1 See Readme.md
To run the project from the command line, go to the dist folder and type the following:
java -jar "RBGParserWrapper.jar" -i <inputFile> -o <outputFile>
For more details about using the Parser:
java -jar "RBGParserWrapper.jar" -h
The Dependency Parser uses:
- Farasa Arabic Segmenter
- Farasa Arabic POS Tagger
- The pre-trained model: model
-
Download the tools and place the jars "FarasaSegmenterJar.jar" and "FarasaPOSJar.jar" in the "./dist/lib" folder.
-
Download the model and place it in "./src/data/" folder.
Otherwise, for building the project from source. Place the jars in "./libs" folder
If you have any inquiry, please contact ([email protected]) ([email protected]) or ([email protected]).
URL for the project and the latest news and downloads http://farasa.qcri.org/
Where to download the latest version, https://github.com/qcri/Farasa
QCRI FARASA package for processing Arabic text is being made
public for research purpose only.
For non-research use, please contact:
Kareem M. Darwish <[email protected]>
Hamdy Mubarak <[email protected]>
Ahmed Abdelali <[email protected]>
This library is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Copyright 2015-2018 QCRI