This project provides several NLP tools such as a dependency parser, a semantic role labeler, a penn-to-dependency converter, a prop-to-dependency converter, and a morphological analyzer. All tools are written in Java and developed by the Computational Language and EducAtion Research (CLEAR) group at the University of Colorado at Boulder.

Important Notice

The ClearParser project is going to move to a bigger project called ClearNLP during the Q2 or Q3 of 2012. Here are expected changes in the ClearNLP project.

• Improved dependency parser.
• Improved semantic role labeler.
• Constituent-to-dependency converter supports both the CoNLL and Stanford dependencies.
• Built-in part-of-speech tagger.
• LibLinear for NLP applications.

Please send an email to choijd<at>colorado<dot>edu if you wish to be notified for the ClearNLP project.

Installation

1. Download the ClearParser system file, and uncompress it.
• tar -zxvf clearparser-sys.tar.gz
2. Download the latest version of the ClearParser jar-file and put it under the 'lib' directory.
• cp clearparser-version.jar lib/
3. Add all jar files in 'lib/' to your classpath. If you are using the Bash shell, it can be something like this.

CLEAR_PATH="path of the [clearparser] directory"
export CLASSPATH=$CLEAR_PATH/lib/opennlp-maxent-3.0.1-incubating.jar:$CLEAR_PATH/lib/opennlp-tools-1.5.1.jar:$CLEAR_PATH/lib/args4j-2.0.12.jar:$CLEAR_PATH/lib/commons-compress-1.1.jar:$CLEAR_PATH/lib/hppc-0.4.0.jar:$CLEAR_PATH/lib/clearparser-version.jar:.

Clear Dependency Parser

Clear dependency parser uses two transition-based dependency parsing algorithms, shift-eager algorithm (Choi & Nicolov, 2009) and shift-pop algorithm (default algorithm; Choi & Palmer, 2011a). These algorithms show near state-of-the-art performance in both speed and accuracy.

• Guidelines

Clear Semantic Role Labeler

Clear semantic role labeler is a dependency-based SRL system. The labeler uses a transition-based SRL algorithm and shows near state-of-the-art performance in accuracy (Choi & Palmer, 2011b).

• Guidelines

Clear Penn-to-Dependency Converter

Clear penn-to-dependency converter takes Penn Treebank style phrase structure trees and generates CoNLL style dependency trees. The converter is up-to-date for the new Treebank guidelines (as in 2011), and shows robust results across different corpora (Choi & Palmer 2010).

• Guidelines

Clear Prop-to-Dependency Converter

Clear prop-to-dependency converter takes PropBank instances and generates CoNLL style dependency trees with semantic roles. The converter is up-to-date for the new PropBank guidelines (as in 2011).

• Guidelines

Clear Morphological Analyzer

Clear morphological analyzer takes a pair of word-form and its part-of-speech tag, and generates the lemma of the word-form. It is a dictionary-based analyzer developed on top of the WordNet morphy.

• Guidelines

References

• K-best, Locally Pruned, Transition-based Dependency Parsing Using Robust Risk Minimization, Jinho D. Choi, Nicolas Nicolov, Collections of Recent Advances in Natural Language Processing V, 205-216, John Benjamins, Amsterdam & Philadelphia, 2009

• Robust Constituent-to-Dependency Conversion for English, Jinho D. Choi, Martha Palmer, Proceedings of the 9th International Workshop on Treebanks and Linguistic Theories (TLT'9), 55-66, Tartu, Estonia, 2010

• Getting the Most out of Transition-based Dependency Parsing, Jinho D. Choi, Martha Palmer, Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies (ACL:HLT'11), 687-692, Portland, Oregon, 2011a

• Transition-based Semantic Role Labeling Using Predicate Argument Clustering, Jinho D. Choi, Martha Palmer, Proceedings of ACL workshop on Relational Models of Semantics (RELMS'11), 37-45, Portland, Oregon, 2011b