lemma_sentence = []
lemma_dict = {}
if __name__ == '__main__':
ZEMBEREK_PATH: str = join('zemberek-full.jar')
startJVM( # dogru olan
getDefaultJVMPath(), '-ea', '-Djava.class.path=%s' % (ZEMBEREK_PATH))
morphology = JClass(
'zemberek.morphology.TurkishMorphology').createWithDefaults()
WordAnalysis = JClass('zemberek.morphology.analysis.WordAnalysis')
lemmas = {}
for sentence in sentence_tokens:
if sentence != '':
analysis: java.util.ArrayList = (
morphology.analyzeAndDisambiguate(sentence).bestAnalysis())
lemma = []
for i, analysis in enumerate(analysis, start=1):
lemma.append(f'{str(analysis.getLemmas()[0])}')
lemma_dict[f'{str(analysis.getLemmas()[0])}'] = lemma_dict.get(
f'{str(analysis.getLemmas()[0])}', 0.0) + 1.0
lemmas_list.append(f'{str(analysis.getLemmas()[0])}')
#print(f'\nFull sentence with POS tags: {" ".join(lemma)}')
lemma_sentence.append(f'{" ".join(lemma)}')
# %% ngram
bigram = ngrams(lemmas_list, 2)
bigram_freq = Counter(bigram)
