stanford-parser を Homebrew, MacPorts でインストール

プロジェクトリンク

• ホームページ
• リポジトリ
• パッケージマネージャページ

コマンドとエイリアス

• lexparser-gui.sh
• lexparser-lang-train-test.sh
• lexparser-lang.sh
• lexparser.sh

プロジェクトの歴史

The CoreNLP release history identifies Stanford Parser, released on 5 December 2002, as the earliest open-source software now comprising CoreNLP. The parser documentation says the original version was mainly written by Dan Klein, with support code and linguistic grammar development by Christopher Manning, followed by extensive contributions from other Stanford NLP researchers.

採用の歴史

The parser became widely packaged because it exposed statistical parsing models through a Java distribution, command-line scripts, GUI tooling, and language wrappers. The official parser page even notes Homebrew installation for OS X, reflecting its role as a standalone CLI package before and alongside the unified CoreNLP distribution.

使われ方

Users download the parser distribution and language models, then run parser scripts or Java classes to produce phrase-structure and dependency parses. The package includes PCFG, lexicalized dependency, lexicalized PCFG, shift-reduce, and neural dependency parser components depending on distribution and model selection.

パッケージ好きにとっての重要性

Package maintainers care about Stanford Parser because it represents the classic heavy NLP CLI package: model jars, Java launch scripts, language-specific assets, and optional GUI behavior. A system package makes it easier to call from shell scripts and wrapper libraries without manually assembling each release.

タイムライン

• 2002: Stanford Parser is released as open-source software.
• 2003: Klein and Manning publish the cited ACL paper for accurate unlexicalized parsing.
• 2014: Parser releases add shift-reduce and dependency improvements in the CoreNLP-era tooling.
• 2020: Standalone parser 4.2.0 retrains English models with treebank fixes.

Related projects

• Stanford Parser is related to CoreNLP, Stanford POS Tagger, Stanford NER, shift-reduce constituency parsing, neural dependency parsing, Tregex/Tsurgeon, and language wrappers for Java, PHP, Python/Jython, Ruby, and .NET.

ソース

• https://nlp.stanford.edu/software/lex-parser.shtml: Stanford parser software page with download, release history, wrappers, and Homebrew note.
• https://stanfordnlp.github.io/CoreNLP/history.html: official release history naming Stanford Parser as the 2002 origin.
• https://stanfordnlp.github.io/CoreNLP/tools_lex_parser.html: official parser overview, package contents, authorship, and parser variants.