The live demo is attached above. You could watch it with Internet browser in a full screen mode.
Since the demo is recorded using release version 0.1.0, the results are a bit lower than the latest release.
Look forward to your review and rating!
Change Log
Summary
This tool can convert CCG derivations to PTB trees using Max Entropy model implemented by OpenNLP, as well as visualizing the tree graphs. The main technical innovation presented here is the effective conversion method which achieves a F score over 95%.
Features:
System requirements:
Successfully tested on:
Sample input:
CCG derivations and PTB trees. One sentence a line.
Install:
On Linux (Ubantu):
On Windows (7):
Run
First: cd to the folder.
GUI:
Command line:
java -cp sweetonionccg2ptb*.jar integration.CCG2PTBConverter -trainccg sampledata/ccg_sample_train -trainptb sampledata/ptb_sample_train -trainpos sampledatapos/ptb_sample_train_lapos_auto -model1 models/ccg2ptb/m1-4-200-lapos-valid-all.bin -model2 models/ccg2ptb/m2-4-200-7-300-lapos-valid-all.bin -cutoff1 4 -cutoff2 7 -iteration1 200 -iteration2 300
java -cp sweetonionccg2ptb*.jar integration.CCG2PTBConverter -testccg sampledata/ccg_sample_test -testpos sampledatapos/ptb_sample_test_lapos_auto -model1 models/ccg2ptb/m1-4-200-lapos-valid-all.bin -model2 models/ccg2ptb/m2-4-200-7-300-lapos-valid-all.bin -resultfile results/rs_lapos_4_200_7_300_sampletest
Use EVALB evaluation script.
Source Code:
Source code could be checked out by "svn checkout svn://svn.code.sf.net/p/ccg2ptb/code/trunk ccg2ptb-code"
Citation:
Xiaotian Zhang, Hai Zhao, Cong Hui. A Machine Learning Approach to Convert CCGbank to Penn Treebank. The Demo Session at the 24th International Conference on Computational Linguistics.
Screenshots:
License:
The software is available for non-commercial purposes under the the Apache License, Version 2.0
Welcome your new ideas, suggestions, criticism and improvements!
Contact:
sweetonion at users.sourceforge.net