Skip to content

Latest commit

 

History

History
72 lines (51 loc) · 1.84 KB

README.md

File metadata and controls

72 lines (51 loc) · 1.84 KB
Raw

Pipeline NLP

Pipeline NLP is a toolkit to demo the different between syntatic information and semantic information. See attachment - pdf file - to get more information or check out my blog.

Functionality

1) SRL (Semantic Role Labeling)
2) POS Tagging (Part of Speech Tagging) 
3) Syntactic Parsing

What's New

1.2

  • Reorganize every single function to moudle
  • Added pos.POSTagger class to get the result of pos tagging.
  • Added srl.SRLTagger class to get the result of srl.
  • Added syntax.SyntaxTree class to get the result of syntax.
  • Remove stanfordcore dependency. POS and Syntaxtree are performed using practNLPTools.
  • Remove requirments.txt file.
  • Remove demo.py file. The functions in this file move to module - core.

1.1

  • Added pos function.
  • Added ner function.
  • Added srl function.
  • Added syntax function.

Requirements

Pipleline NLP has been tested on Python 2.7.

Python Dependencies

Basic Usages

Import module as first step

>>> from core import *

predict POS(part-of-speech) tagging from sentence:

>>> POSTagger.getInfo("how are you?")
[('how', 'WRB'), ('are', 'VBP'), ('you', 'PRP'), ('?', '.')]

predict SRL(semantic role labeling) from sentence:

>>> SRL_Tagger.getInfo("I ate an apple")
[{'A1': 'an apple', 'A0': 'I', 'V': 'ate'}]
>>> SRL_Tagger.getInfo("An apple was eaten by me")
[{'A1': 'An apple', 'A0': 'by me', 'V': 'eaten'}]

predict syntatic parsing tree from sentence:

>>> Syntax_Tree.getInfo("I ate an apple")
'(S1(S(NP(PRP I))(VP(VBD ate)(NP(DT an)(NN apple)))))'
>>> Syntax_Tree.getInfo("An apple was eaten by me")
'(S1(S(NP(DT An)(NN apple))(VP(VBD was)(VP(VBN eaten)(PP(IN by)(NP(PRP me)))))))'