https://github.com/touhi99/parsing

Parsing Course Project - A CYK implementation with CNF
https://github.com/touhi99/parsing

cky cnf cyk-algorithm nltk parsing

Last synced: about 4 hours ago
JSON representation

Parsing Course Project - A CYK implementation with CNF

• Host: GitHub
• URL: https://github.com/touhi99/parsing
• Owner: touhi99
• Created: 2017-12-25T17:33:34.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2018-06-22T10:27:28.000Z (over 6 years ago)
• Last Synced: 2023-11-26T21:28:19.821Z (12 months ago)
• Topics: cky, cnf, cyk-algorithm, nltk, parsing
• Language: Python
• Homepage:
• Size: 5.86 KB
• Stars: 0
• Watchers: 2
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
CYK Parsing Implementation

Programming Language- Python 3.6.1

Overview:

CKY.py

* Class ConverCNF()

   * startSymbolAdd() - function adds a new start symbol if existing symbol already on the right hand side of the rules

   * eliminateEpsilon() - if rule contains any ‘ε’ in the right hand side, it removes and restructure the grammar accordingly

   * eliminateVariableUnit() - This function removes single production non-terminal from the right hand side

   * moveTerminalToUnits() - this function add unit non-terminal for terminals

   * replaceLongProd() - replace any non-terminal production more than 2 are adjusted in the form of A -> B C

* Class CYK()

   * readGrammar() - read input grammar, add the rules as dictionary, where key goes the left-hand side rule and value contains right hand-side rules in a list

   * readString() - read input string from the text files

   * readOutput() - print the modified grammar & CKY table

   * parser() - CKY parser algorithm

Dependencies:

re - should be included with python by default

NLTK-

* Pip3 install nltk

Usage:

* On terminal where the files exist, write command

	Python3 CYK.py  

* Sample grammar file as included

   * First line contains the start symbol

   * Epsilon defines with ‘ε’ symbol

   * Terminals are separated by single-quote (‘’) e.g. ‘John’, ‘man’

   * Multiple terminals and non-terminals are separated by whitespace

* Sample string file as included- each line contains separate string