! apt-get update
! apt-get install g++ openjdk-8-jdk 
! pip3  install  nltk konlpy

import nltk
nltk.download('punkt')
nltk.download('tagsets')
nltk.download('averaged_perceptron_tagger')
text_eng  = " Don't hesitate to ask questions"
text_kor  = """삼성 갤럭시(GalaxyNote)노트의 신형을 홍보 합니다. 
홍보:유관순   031-478-2311  010-8888-9999.
삼성 페이지 https://www.samsung.com/sec/index.html"""

text_kor

from nltk import sent_tokenize, word_tokenize, FreqDist
sent_tokenize(text_kor)

tokens = word_tokenize(text_kor)
tokens

dict(FreqDist(tokens))

text_kor

import re
tokenizer = re.compile(r'[가-힣]+')
tokenizer.findall(text_kor)

tokenizer = re.compile(r'[0-9]{3}-[0-9]{3,4}-[0-9]{4}')
tokenizer.findall(text_kor)

tokenizer = re.compile(r'\d{3}-\d{3,4}-\d{4}')
tokenizer.findall(text_kor)

tokenizer = re.compile(r'[^ 가-힣]+')
tokenizer.findall(text_kor)

tokenizer.sub("", text_kor)#.split(" ")

tokenizer = re.compile(r'https://[w]{3}.[A-z]+.[./A-z]+')
tokenizer.findall(text_kor)

text_eng = text_eng.lower()
text_eng

from nltk.tokenize import TreebankWordTokenizer
tokenizer = TreebankWordTokenizer()
token     = tokenizer.tokenize(text_eng)
token

from nltk import pos_tag
pos_tag(token)

import nltk.help as nltk_help
nltk_help.upenn_tagset('PRP')  # 대명사

nltk_help.upenn_tagset('JJ')  # 형용사

from konlpy.tag import Okt
twitter = Okt()

# Stemming
text = "파이썬을 활용하여 자연어 분석 특강입니다"
print(twitter.pos(text, stem="true")) 

print(twitter.pos(text))

%%time
from konlpy.tag import Kkma
kkma = Kkma()
print(kkma.pos(text))

%%time
from konlpy.tag import Hannanum
han = Hannanum()
print(han.pos(text))