项目作者: StarlangSoftware

项目描述 :
Annotated constituency treebank library
高级语言: Python
项目地址: git://github.com/StarlangSoftware/AnnotatedTree-Py.git
创建时间: 2020-02-02T07:40:07Z
项目社区:https://github.com/StarlangSoftware/AnnotatedTree-Py

开源协议:GNU General Public License v3.0

下载


Constituency TreeBank

A treebank is a corpus where the sentences in each language are syntactically (if necessary morphologically) annotated. In the treebanks, the syntactic annotation usually follows constituent and/or dependency structure.

Treebanks annotated for the syntactic or semantic structures of the sentences are essential for developing state-of-the-art statistical natural language processing (NLP) systems including part-of-speech-taggers, syntactic parsers, and machine translation systems. There are two main groups of syntactic treebanks, namely treebanks annotated for constituency (phrase structure) and the ones that are annotated for dependency structure.

Data Format

We extend the original format with the relevant information, given between curly braces. For example, the word ‘problem’ in a sentence in the standard Penn Treebank notation, may be represented in the data format provided below:

  1. (NN problem)

After all levels of processing are finished, the data structure stored for the same word has the following form in the system.

  1. (NN {turkish=sorunu} {english=problem}
  2. {morphologicalAnalysis=sorun+NOUN+A3SG+PNON+ACC}
  3. {metaMorphemes=sorun+yH}
  4. {semantics=TUR10-0703650})

As is self-explanatory, ‘turkish’ tag shows the original Turkish word; ‘morphologicalanalysis’ tag shows the correct morphological parse of that word; ‘semantics’ tag shows the ID of the correct sense of that word; ‘namedEntity’ tag shows the named entity tag of that word; ‘propbank’ tag shows the semantic role of that word for the verb synset id (frame id in the frame file) which is also given in that tag.

Video Lectures

For Developers

You can also see Cython, Java, C++, C, Js, Swift, or C# repository.

Requirements

Python

To check if you have a compatible version of Python installed, use the following command:

  1. python -V

You can find the latest version of Python here.

Git

Install the latest version of Git.

Pip Install

  1. pip3 install NlpToolkit-AnnotatedTree

Download Code

In order to work on code, create a fork from GitHub page.
Use Git for cloning the code to your local or below line for Ubuntu:

  1. git clone <your-fork-git-link>

A directory called DataStructure will be created. Or you can use below link for exploring the code:

  1. git clone https://github.com/starlangsoftware/AnnotatedTree-Py.git

Open project with Pycharm IDE

Steps for opening the cloned project:

  • Start IDE
  • Select File | Open from main menu
  • Choose AnnotatedTree-Py file
  • Select open as project option
  • Couple of seconds, dependencies will be downloaded.

Detailed Description

TreeBankDrawable

To load an annotated TreeBank:

  1. TreeBankDrawable(folder: str, String pattern: str)
  2. a = TreeBankDrawable("/Turkish-Phrase", ".train")
  3. TreeBankDrawable(folder: str)
  4. a = new TreeBankDrawable("/Turkish-Phrase")

To access all the trees in a TreeBankDrawable:

  1. for i in range(a.sentenceCount()):
  2. parseTree = a.get(i);
  3. ....
  4. }

ParseTreeDrawable

To load a saved ParseTreeDrawable:

  1. ParseTreeDrawable(file: str)

is used. Usually it is more useful to load TreeBankDrawable as explained above than to load ParseTree one by one.

To find the node number of a ParseTreeDrawable:

  1. nodeCount() -> int

the leaf number of a ParseTreeDrawable:

  1. leafCount() -> int

the word count in a ParseTreeDrawable:

  1. wordCount(excludeStopWords: bool) -> int

above methods can be used.

LayerInfo

Information of an annotated word is kept in LayerInfo class. To access the morphological analysis
of the annotated word:

  1. getMorphologicalParseAt(index: int) -> MorphologicalParse

meaning of an annotated word:

  1. getSemanticAt(self, index: int) -> str

the shallow parse tag (e.g., subject, indirect object etc.) of annotated word:

  1. getShallowParseAt(self, index: int) -> str

the argument tag of the annotated word:

  1. getArgumentAt(self, index: int) -> Argument

the word count in a node:

  1. getNumberOfWords(self) -> int

Cite

  1. @inproceedings{yildiz-etal-2014-constructing,
  2. title = "Constructing a {T}urkish-{E}nglish Parallel {T}ree{B}ank",
  3. author = {Y{\i}ld{\i}z, Olcay Taner and
  4. Solak, Ercan and
  5. G{\"o}rg{\"u}n, Onur and
  6. Ehsani, Razieh},
  7. booktitle = "Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)",
  8. month = jun,
  9. year = "2014",
  10. address = "Baltimore, Maryland",
  11. publisher = "Association for Computational Linguistics",
  12. url = "https://www.aclweb.org/anthology/P14-2019",
  13. doi = "10.3115/v1/P14-2019",
  14. pages = "112--117",
  15. }