twitter.pos('우리아이오아이는 이뻐요')

twitter.pos('우리아이오아이는 이뻐요')

twitter.pos('트와이스tt는 좋아요')

from ckonlpy.tag import Twitter

twitter.add_dictionary('아이오아이', 'Noun')
twitter.add_dictionary(['트와이스', 'tt'], 'Noun')

twitter.add_dictionary('lovit', 'Name', force=True)

twitter.template_tagger.templates

twitter.template_tagger.add_a_template(('Noun', 'Noun', 'Josa'))

my_weights = [
    ('num_nouns', -0.1),
    ('num_words', -0.2),
    ('no_noun', -1),
    ('len_sum_of_nouns', 0.2)
]

def my_evaluate_function(candidate):
    num_nouns = len([word for word, pos, begin, e in candidate if pos == 'Noun'])
    num_words = len(candidate)
    has_no_nouns = (num_nouns == 0)
    len_sum_of_nouns = 0 if has_no_nouns else sum(
        (len(word) for word, pos, _, _ in candidate if pos == 'Noun'))

    scores = (num_nouns, num_words, has_no_nouns, len_sum_of_nouns)
    score = sum((score * weight for score, (_, weight) in zip(scores, my_weights)))
    return score

twitter.set_evaluator(my_weights, my_evaluate_function)

from ckonlpy.tag import Postprocessor

passwords = {'아이오아이', ('정말', 'Noun')}
postprocessor = Postprocessor(twitter, passwords = passwords)
postprocessor.pos('우리아이오아이는 정말 이뻐요')
# [('아이오아이', 'Noun'), ('정말', 'Noun')]

stopwords = {'는'}
postprocessor = Postprocessor(twitter, stopwords = stopwords)
postprocessor.pos('우리아이오아이는 정말 이뻐요')
# [('우리', 'Modifier'), ('아이오아이', 'Noun'), ('정말', 'Noun'), ('이뻐', 'Adjective'), ('요', 'Eomi')]

passtags = {'Noun'}
postprocessor = Postprocessor(twitter, passtags = passtags)
postprocessor.pos('우리아이오아이는 정말 이뻐요')
# [('아이오아이', 'Noun'), ('정말', 'Noun')]

replace = {'아이오아이': '아이돌', ('이뻐', 'Adjective'): '예쁘다'}
postprocessor = Postprocessor(twitter, replace = replace)
postprocessor.pos('우리아이오아이는 정말 이뻐요')
# [('우리', 'Modifier'), ('아이돌', 'Noun'), ('는', 'Josa'), ('정말', 'Noun'), ('예쁘다', 'Adjective'), ('요', 'Eomi')]

ngrams = [(('미스', '함무라비'), 'Noun'), ('바람', '의', '나라')]
postprocessor = Postprocessor(twitter, ngrams = ngrams)
postprocessor.pos('미스 함무라비는 재밌는 드라마입니다')
# [('미스 - 함무라비', 'Noun'), ('는', 'Josa'), ('재밌는', 'Adjective'), ('드라마', 'Noun'), ('입니', 'Adjective'), ('다', 'Eomi')]

from ckonlpy.utils import load_wordset

passwords = load_wordset('./passwords.txt')
print(passwords) # {('아이오아이', 'Noun'), '아이오아이', '공연'}

stopwords = load_wordset('./stopwords.txt')
print(stopwords) # {'은', '는', ('이', 'Josa')}

from ckonlpy.utils import load_replace_wordpair

replace = load_replace_wordpair('./replacewords.txt')
print(replace) # {'아빠': '아버지', ('엄마', 'Noun'): '어머니'}

from ckonlpy.utils import load_ngram

ngrams = load_ngram('./ngrams.txt')
print(ngrams) # [('바람', '의', '나라'), (('미스', '함무라비'), 'Noun')]