Skip to content

Commit

Permalink
Begin updating the core lex-parser page
Browse files Browse the repository at this point in the history
  • Loading branch information
AngledLuffa committed Nov 15, 2023
1 parent 53a1d63 commit 4f55ebb
Showing 1 changed file with 21 additions and 28 deletions.
49 changes: 21 additions & 28 deletions _pages/tools_lex_parser.md
Original file line number Diff line number Diff line change
Expand Up @@ -8,7 +8,7 @@ toc: true
parent: Additional Tools
---

## About
### About

A natural language parser is a program that works out the grammatical
**structure of sentences** , for instance, which groups of words go together
Expand Down Expand Up @@ -49,7 +49,7 @@ also been used for other languages, such as Italian, Bulgarian, and
Portuguese.

The parser provides [Universal Dependencies (v1) and Stanford
Dependencies](stanford-dependencies.html) output as well as phrase structure
Dependencies](https://nlp.stanford.edu/software/stanford-dependencies.html) output as well as phrase structure
trees. Typed dependencies are otherwise known **grammatical relations**. This
style of output is available only for English and Chinese. For more details,
please refer to the [Stanford Dependencies webpage](stanford-
Expand All @@ -62,13 +62,13 @@ to update to it.).
#### Shift-reduce constituency parser

As of version 3.4 in 2014, the parser includes the code necessary to run a
[shift reduce parser](srparser.html), a much faster constituent parser with
[shift reduce parser](tools_srparser.md), a much faster constituent parser with
competitive accuracy. Models for this parser are linked below.

#### Neural-network dependency parser

In version 3.5.0 (October 2014) we released a [high-performance dependency
parser](nndep.html) powered by a neural network. The parser outputs typed
parser](pytorch-depparse.md) powered by a neural network. The parser outputs typed
dependency parses for English and Chinese. The models for this parser are
included in the general Stanford Parser models package.

Expand Down Expand Up @@ -116,7 +116,7 @@ commercial license, then you need a license to both the Stanford Parser and
the Stanford POS tagger. Or you can get the whole bundle of Stanford CoreNLP.)
If you don't need a commercial license, but would like to support maintenance
of these tools, we welcome gift funding: use [this
form](http://giving.stanford.edu/goto/writeingift) and write "Stanford NLP
form](http://giving.stanford.edu/) and write "Stanford NLP
Group open source software" in the Special Instructions.

### Citing the Stanford Parser
Expand All @@ -140,11 +140,9 @@ version of the parser being used!
> _Proceedings of ACL 2013_
> _For the Shift-Reduce Constituency parser (starting at version 3.2):_
> This parser was written by John Bauer. You can thank him and cite the web
> page describing it:
> [`https://nlp.stanford.edu/software/srparser.html`](https://nlp.stanford.edu/software/srparser.html).
> You can also cite the original research papers of others mentioned on that
> page.
> This parser was written by John Bauer. You can thank him and [cite the web
> page describing it](tools_srparser.md)
> You can also cite the original research papers of others mentioned on that page.
> _For the PCFG parser (which also does POS tagging):_
> Dan Klein and Christopher D. Manning. 2003. [Accurate Unlexicalized
Expand Down Expand Up @@ -186,25 +184,25 @@ version of the parser being used!
> _For the Chinese Parser:_
> Roger Levy and Christopher D. Manning. 2003\. [Is it harder to parse
> Chinese, or the Chinese Treebank?](/pubs/acl2003-chinese.pdf) _ACL 2003_ ,
> Chinese, or the Chinese Treebank?](http://nlp.stanford.edu/pubs/acl2003-chinese.pdf) _ACL 2003_ ,
> pp. 439-446.
> _For the Chinese Stanford Dependencies:_
> Pi-Chuan Chang, Huihsin Tseng, Dan Jurafsky, and Christopher D. Manning.
> 2009\. [Discriminative Reordering with Chinese Grammatical Relations
> Features](/pubs/ssst09-chang.pdf). In _Proceedings of the Third Workshop on
> Features](http://nlp.stanford.edu/pubs/ssst09-chang.pdf). In _Proceedings of the Third Workshop on
> Syntax and Structure in Statistical Translation_.
> _For the Arabic parser:_
> Spence Green and Christopher D. Manning. 2010\. [Better Arabic Parsing:
> Baselines, Evaluations, and Analysis](/pubs/coling2010-arabic.pdf). In
> Baselines, Evaluations, and Analysis](http://nlp.stanford.edu/pubs/coling2010-arabic.pdf). In
> _COLING 2010_.
> _For the French parser:_
> Spence Green, Marie-Catherine de Marneffe, John Bauer, and Christopher D.
> Manning. 2010\. [Multiword Expression Identification with Tree Substitution
> Grammars: A Parsing _tour de force_ with
> French.](/pubs/green+demarneffe+bauer+manning.emnlp11.pdf). In _EMNLP 2011_.
> French.](http://nlp.stanford.edu/pubs/green+demarneffe+bauer+manning.emnlp11.pdf). In _EMNLP 2011_.
> _For the Spanish parser:_
> Most of the work on Spanish was by Jon Gauthier. There is no published
Expand All @@ -215,26 +213,21 @@ version of the parser being used!

1. If you're new to parsing, you can start by running the GUI to try out the parser. Scripts are included for linux (lexparser-gui.sh) and Windows (lexparser-gui.bat).
2. Take a look at the Javadoc `lexparser` package documentation and `LexicalizedParser` class documentation. (Point your web browser at the `index.html` file in the included `javadoc` directory and navigate to those items.)
3. Look at the [parser FAQ](parser-faq.html) for answers to common questions.
3. Look at the [parser FAQ](tools_parser_faq.md) for answers to common questions.
4. If none of that helps, please see [our email guidelines](email.html) for instructions on how to reach us for further assistance.



### Download

**[Download Stanford Parser version 4.2.0](stanford-parser-4.2.0.zip)**
The standard download includes models for Arabic, Chinese, English, French,
[The standard CoreNLP download](index.md) includes the code for the lexicalized parser and all of its siblings described above.

This standard download includes models for Arabic, Chinese, English, French,
German, and Spanish. There are additional models we do not release with the
standalone parser, including shift-reduce models, that can be found in the
models jars for each language. Below are links to those jars.

**[Arabic Models](stanford-corenlp-4.2.0-models-arabic.jar)** **[Chinese
Models](stanford-corenlp-4.2.0-models-chinese.jar)** **[English
Models](stanford-corenlp-4.2.0-models-english.jar)** **[French
Models](stanford-corenlp-4.2.0-models-french.jar)** **[German
Models](stanford-corenlp-4.2.0-models-german.jar)** **[Spanish
Models](stanford-corenlp-4.2.0-models-spanish.jar)**
models jars for each language.

Older versions are linked below.

### Extensions: Packages by others using the parser

Expand Down Expand Up @@ -271,8 +264,8 @@ Models](stanford-corenlp-4.2.0-models-spanish.jar)**
### Release history


| | | |
|---|---|---|
| Version | Date | Description | Models |
|---|---|---|---|
| <a href="stanford-parser-4.2.0.zip">Version 4.2.0</a> | 2020-11-17 | Retrain English models with treebank fixes | <a href="stanford-corenlp-4.2.0-models-arabic.jar">arabic  </a> <a href="stanford-corenlp-4.2.0-models-chinese.jar">chinese  </a> <a href="stanford-corenlp-4.2.0-models-english.jar">english  </a> <a href="stanford-corenlp-4.2.0-models-french.jar">french  </a> <a href="stanford-corenlp-4.2.0-models-german.jar">german  </a> <a href="stanford-corenlp-4.2.0-models-spanish.jar">spanish</a> |
| <a href="stanford-parser-4.0.0.zip">Version 4.0.0</a> | 2020-05-22 | Model tokenization updated to UDv2.0 | <a href="stanford-corenlp-4.0.0-models-arabic.jar">arabic  </a> <a href="stanford-corenlp-4.0.0-models-chinese.jar">chinese  </a> <a href="stanford-corenlp-4.0.0-models-english.jar">english  </a> <a href="stanford-corenlp-4.0.0-models-french.jar">french  </a> <a href="stanford-corenlp-4.0.0-models-german.jar">german  </a> <a href="stanford-corenlp-4.0.0-models-spanish.jar">spanish</a> |
| <a href="stanford-parser-full-2018-10-17.zip">Version 3.9.2</a> | 2018-10-17 | Updated for compatibility | <a href="stanford-arabic-corenlp-2018-10-05-models.jar">arabic  </a> <a href="stanford-chinese-corenlp-2018-10-05-models.jar">chinese  </a> <a href="stanford-english-corenlp-2018-10-05-models.jar">english  </a> <a href="stanford-french-corenlp-2018-10-05-models.jar">french  </a> <a href="stanford-german-corenlp-2018-10-05-models.jar">german  </a> <a href="stanford-spanish-corenlp-2018-10-05-models.jar">spanish</a> |
Expand Down

0 comments on commit 4f55ebb

Please sign in to comment.