+ LoPar

Description

LoPar is an implementation of a parser for head-lexicalised probabilistic context-free grammars (see Carroll/Rooth). (URL: http://www.ims.uni-stuttgart.de/projekte/gramotron/SOFTWARE/LoPar-en.html)

LoPar can be used for

• symbolic parsing with a context-free grammar (CFG)
• statistical parsing with PCFGs and head-lexicalised PCFGs
• training of PCFGs and HLPCFGs
• Viterbi parsing
• chunking
• tagging
• and more

You can download LoPar as a gzip-compressed tar file for either Sun/Solaris or Intel/Linux platforms. It is freely available for education, research and other non-commercial purposes. There is also a paper on LoPar and an English grammar as well as a German grammar.

Please send comments, suggestions and bug reports to Helmut.Schmid@ims.uni-stuttgart.de

License

unknown

Downloads

• lopar-3.0.solaris.tar.gz: LoPar for Solaris
• lopar-3.0.linux.tar.gz: LoPar for Linux
• German-Chunker.tar.gz: LoPar grammar for chunking german
• English-HLPCFG.tar.gz: LoPar grammar for parsing english

See here for a german grammar. Contact Sabine.Schulte@ims.uni-stuttgart.de (Sabine Schulte im Walde).

Documetation

• Helmut Schmid and Sabine Schulte im Walde (2002): Robust German Noun Chunking with a Probabilistic Context-Free Grammar. Proceedings of the 18th International Conference on Computational Linguistics (COLING-00). Saarbrücken, Germany. (PS)
• Glenn Carroll, and Mats Rooth (1998): Valence Induction with a Head-lexicalized PCFG, Empirical Methods in NLP workshop, Granada, 1998. (PS)
• Sabine Schulte im Walde (2000): The German Statistical Grammar Model: Development, Training and Linguistic Exploitation. Institute for Computational Linguistics, University of Stuttgart. (PS)
• Helmut Schmid (2000): LoPar: Design and Implementation. Institute for Computational Linguistics, University of Stuttgart. (PS).

Evaluation on the HeiseCorpus

• heise-chunks.gz
• heise-chunk-trees.gz