def __init__(self, korean_corpus_worker, korean_corpus_splitter, token_min,
korean_stop_words, **kwargs):
# Complete
try:
# -------------
# Twitter
self.tagger = corpus.Okt()
# self.tagger_options = dict(stem=bool(kwargs.get('stem', True)))
# -------------
# Mecab
# self.tagger = corpus.Mecab()
# self.tagger_options = dict(flatten=bool(kwargs.get('flatten', True)))
except:
say.error('Could not load korean corpus')
raise KoreanCorpusException(
TextRankException('Could not load korean corpus'))
self.korean_corpus_tags = korean_corpus_worker
self.delimiters = korean_corpus_splitter
self.min_token_length = token_min
self.keyword_stop_lists = korean_stop_words
self.sentence_splitter = self.__sentences_splitter()
# Configurations Korean Corpus
# Option [1. mecab, 2. twitter]
# self.pos = lambda sentence: self.tagger.pos(sentence, **self.tagger_options)
self.pos = lambda sentence: self.tagger.pos(sentence, stem=False)
def tagger_load(self):
name = self.tagger_name
if name == 'Okt':
self.tagger = tag.Okt()
elif name == 'Kkma':
self.tagger = tag.Kkma()
elif name == 'Komoran':
self.tagger = tag.Komoran()
else:
self.tagger = tag.Hannanum()
print("load tagger")
def __init__(self, pre_trained=True, analyzer='Hannanum'):
self.pre_trained = pre_trained
if analyzer == 'Hannanum':
self.analyzer = tag.Hannanum()
elif analyzer == 'Kkma':
self.analyzer = tag.Kkma()
elif analyzer == 'Komoran':
self.analyzer = tag.Komoran()
elif analyzer == 'Mecab':
self.analyzer = tag.Mecab()
elif analyzer == 'Okt':
self.analyzer = tag.Okt()
else:
if pre_trained == False:
pass
else:
print('Enter a valid KoNLPy analyzer name.\n\tavailable: Hannanum, Kkma, Komoran, Mecab, Okt')
self.WordExtractor = WordExtractor(min_frequency=0)
self.noun_extractor = LRNounExtractor(verbose=False)
self.word_score = {}
def __init__(self):
import konlpy
from konlpy import tag
print("Initialize Okt from konlpy ({})".format(konlpy.__version__),
file=sys.stderr)
self._mdl = tag.Okt()
def decompose(titles):
okt = tag.Okt()
for title in titles:
print(okt.morphs(title))
def test(titles):
test_title_list = titles[:20]
okt = tag.Okt()
for title in test_title_list:
print(okt.pos(title))