def getLemmatization(example_sent):
word_tokens = Tokenization.tokenizationProcess(example_sent)
filtered_sentence = []
for w in word_tokens:
filtered_sentence.append(nltk.ISRIStemmer().suf32(w))
return filtered_sentence
def defineStemming(example_sent):
word_tokens = Tokenization.tokenizationProcess(example_sent)
filtered_sentence = []
ps = nltk.ISRIStemmer()
for w in word_tokens:
filtered_sentence.append(ps.stem(w))
return filtered_sentence
class Normalizer :
stopword_list = open ('res/stop_words.txt','r').read().strip('\n')
stemmer = nltk.ISRIStemmer()
def tokenize_text(self,text):
tokens = nltk.word_tokenize(text)
tokens = [token.strip() for token in tokens]
return tokens
def stemming_text(self,text):
tokens = self.tokenize_text(text)
filtered_tokens = [self.stemmer.stem(token) for token in tokens]
filtered_text = ' '.join(filtered_tokens)
return filtered_text
def remove_special_characters(self,text):
def remove_conjunction(token):
if token[0] == 'و' and len(token) > 4:
t = token[1:len(token)]
return t
else:
return token
tokens = self.tokenize_text(text)
pattern = re.compile(r'[?|.|:|;|,|"|\d|$|&|*|%|@|(|)|~]'.format(re.escape(string.punctuation)))
filtered_tokens = filter(None, [pattern.sub('',remove_conjunction(token)) for token in tokens])
filtered_text = ' '.join(filtered_tokens)
return filtered_text
def remove_stopwords(self,text):
tokens = self.tokenize_text(text)
filtered_tokens = [token for token in tokens if token not in self.stopword_list]
filtered_text = ' '.join(filtered_tokens)
return filtered_text
def remove_repeated_characters(self,text):
tokens = self.tokenize_text(text)
repeat_pattern = re.compile(r'(\w*)(\w)\2(\w*)')
match_substitution = r'\1\2\3'
def replace(old_word):
if wn.synsets(old_word):
return old_word
new_word = repeat_pattern.sub(match_substitution, old_word)
return replace(new_word) if new_word != old_word else new_word
correct_tokens = [replace(word) for word in tokens]
filtered_text = ' '.join(correct_tokens)
return filtered_text
def normalize_corpus(self,corpus):
normalized_corpus = []
for text in corpus:
text = self.remove_special_characters(text)
text = self.remove_stopwords(text)
text = self.remove_repeated_characters(text)
text = self.stemming_text(text)
text = self.remove_stopwords(text)
normalized_corpus.append(text)
return normalized_corpus
def getLemmatization(example_sent):
lemmatizer = WordNetLemmatizer()
word_tokens = Tokenization.tokenizationProcess(example_sent)
lemmatized = []
for word in word_tokens:
lemmatized.append(nltk.ISRIStemmer().suf32(word))
return lemmatized
def resolve_txt(root, info):
text = info.context.get('txt') #.encode('utf8')
text = re.sub('[%s]' % re.escape(string.punctuation), '', text)
text = re.sub('\w*\d\w*', '', text)
text = re.sub('[‘’“”«»…]', '', text)
text = re.sub("-_،؟", '', text)
lemmatizer = WordNetLemmatizer()
word_tokens = word_tokenize(text)
lemmatized = []
for word in word_tokens:
lemmatized.append(nltk.ISRIStemmer().suf32(word))
return {"result": lemmatized}
def __init__(self, raw_data=None):
self.analyzer = pyaramorph.Analyzer()
self.stemmer = nltk.ISRIStemmer()
self.lemmatizer = nltk.WordNetLemmatizer()
self.segmenter = nltk.data.load("tokenizers/punkt/english.pickle")
if raw_data is not None:
self.raw_data = raw_data
self.org_data = raw_data
self.analyze_text()
self.ambig()
self.load_corpus('Tashkeela')
self.select_cand()
self.print_result()
def root(word):
st = nltk.ISRIStemmer()
tokens = nltk.word_tokenize(word)
result = [st.stem(w) for w in tokens]
return result
import re # for pre-processing text
import string # for pre-processing text
import nltk # for processing texts
from nltk.corpus import stopwords
nltk.download('punkt')
nltk.download('stopwords')
nltk.download('wordnet')
# define arabic punctuations
arabic_punctuations = '''`÷×؛<>_()*&^%][ـ،/:"؟.,'{}~¦+|!”…“–ـ'''
english_punctuations = string.punctuation
punctuations_list = arabic_punctuations + english_punctuations
st = nltk.ISRIStemmer()
def clean_text(text):
"""
DESCRIPTION:
This function to clean text
INPUT:
text: string
OUTPUT:
text: string after clean it
"""
# remove english letters
text = re.sub("[a-zA-Z]", " ", str(text))
# remove \n from text
text = re.sub('\n', ' ', text)
# remove number
text = re.sub(r'\d+', '', text)
def __init__(self):
self.analyzer = pyaramorph.Analyzer()
self.stemmer = nltk.ISRIStemmer()
self.lemmatizer = nltk.WordNetLemmatizer()
self.segmenter = nltk.data.load("tokenizers/punkt/english.pickle")
