NLP recipe 1: processing tweets with spaCy

recipe_tweet.py

from __future__ import unicode_literals, print_function

#pip install spacy && python -m spacy download en
import spacy

'''
NLP tasks:
1.tokenization
2.part of speech tagging
3.chunking
'''
#load the language model for English
nlp = spacy.load('en_core_web_sm')

# Link to tweet: https://twitter.com/dataandme/status/989938791744987137
doc1 = nlp(u"⭐️ intro to a 🌟 tool! \"🕵️ RegExplain\" by @grrrck")

# test tokenization and POS tagging
for token in doc1:
    if not token.is_punct: # filter punctuation
        print(token.text, token.pos_)
        
# test noun chunks
for chunk in doc1.noun_chunks:
    print(chunk.text)