def _get_morph():
morph = majka.Majka(os.path.join(NLP_DATA_PATH, 'majka.w-lt'))
morph.tags = False
morph.first_only = True
morph.negative = "ne"
morph.flags |= majka.IGNORE_CASE # ignore the word case whatsoever
return morph
def convert_words_into_lemmas(text, text_language):
try:
morph = majka.Majka("../majka/" + text_language + ".w-lt")
except:
return text
result = ""
for word in text.split():
lemma_result = morph.find(word)
if len(lemma_result) is 0:
result += (" " + word)
else:
result += (" " + lemma_result[0]['lemma'])
return result
def init_lemmatizer(self, majka_path):
lemmatizer = majka.Majka('{}\\{}'.format(majka_path, 'majka.w-lt'))
lemmatizer.first_only = True
lemmatizer.tags = False
lemmatizer.negative = 'ne'
return lemmatizer
import psycopg2
import os
import re
import sys
import majka
import json
import time
morph_cz = majka.Majka('/home/jhu/PycharmProjects/ConEv/nlp/majka.w-lt')
morph_en = majka.Majka('/home/jhu/PycharmProjects/ConEv/nlp/w-lt.en.fsa')
morph_de = majka.Majka('/home/jhu/PycharmProjects/ConEv/nlp/w-lt.ger.fsa')
# https://nlp.fi.muni.cz/czech-morphology-analyser/
db_connection = psycopg2.connect(
"dbname=ConEv user=postgres password=forward host=127.0.0.1")
db = db_connection.cursor()
def find_word_attrs(word, language):
#-----------------------------------------------------------------------------------------------CZ
morph_cz.flags |= majka.ADD_DIACRITICS # find word forms with diacritics
morph_cz.flags |= majka.DISALLOW_LOWERCASE # do not enable to find lowercase variants
morph_cz.flags |= majka.IGNORE_CASE # ignore the word case whatsoever
morph_cz.flags = 0 # unset all flags
morph_cz.tags = False # return just the lemma, do not process the tags
morph_cz.tags = True # turn tag processing back on (default)
morph_cz.compact_tag = True # return tag in compact form (as returned by Majka)
morph_cz.compact_tag = False # do not return compact tag (default)
import majka
lemmatizer = majka.Majka('majka.w-lt')
lemmatizer.first_only = True
lemmatizer.tags = False
lemmatizer.negative = 'ne'
def lemmatize(word):
lemma = lemmatizer.find(word)
if len(lemma) > 0:
return lemma[0]['lemma']
else:
return word