def get_bigram_list(full_sentence_list, stem=False):
sentence_stream = [doc.split(" ") for doc in full_sentence_list]
#print(sentence_stream)
stemmer = RafiStemmer()
bigram = Phrases(sentence_stream, min_count=2, threshold=5, delimiter=b'_')
bigram_phraser = Phraser(bigram)
bigram_list = []
#print(bigram_phraser)
for sent in sentence_stream:
tokens_ = bigram_phraser[sent]
for each_bigram in tokens_:
if each_bigram.count('_') == 1:
#print(each_bigram)
if stem == True:
bigram_list.append(stemmer.stem_word(each_bigram))
else:
bigram_list.append(each_bigram)
bigram_count_list = []
for each_unique_bigram in set(bigram_list):
bigram_count_list.append(
[each_unique_bigram,
bigram_list.count(each_unique_bigram)])
return (bigram_count_list)
def stemmer(words):
steams = []
stemmer = RafiStemmer()
for i in words:
steams.append(stemmer.stem_word(i))
return steams
def preprocess_documents(doc):
preprocessed_list_of_docs = []
stemmer = RafiStemmer()
stop_words = load_stop_word()
preprocessed_docs = []
doc_token = []
if isinstance(doc, str):
for token in punctuation_remover(doc).split():
if token not in stop_words and len(token) >= 3:
if len(stemmer.stem_word(token)) >= 2:
doc_token.append(stemmer.stem_word(token))
return doc_token
def preprocess_documents(doc):
preprocessed_list_of_docs = []
stemmer = RafiStemmer()
stop_words = load_stop_word()
preprocessed_docs = []
doc_token = []
if isinstance(doc, str):
for token in punctuation_remover(doc).split():
if token not in stop_words and len(token) >= 3:
if len(stemmer.stem_word(token)) >= 2:
doc_token.append(stemmer.stem_word(token))
return doc_token
header_list = list(range(0, len(dictionary)-1))
all_list = [header_list]
for each_list in bow_corpus:
temp_list = [0]*len(dictionary)
for each_tuple in each_list:
temp_list[each_tuple[0]] = each_tuple[1]
all_list.append(temp_list)
minimal_all_list = []
minimal_header_list = []
for i in range(len(dictionary)):
minimal_header_list.append(dictionary[i])
minimal_all_list.append(minimal_header_list)
for each_mini_list in all_list[1:]:
minimal_all_list.append(each_mini_list)
return(minimal_all_list)
from bengali_stemmer.rafikamal2014 import RafiStemmer
stemmer = RafiStemmer()
stemmed_word = stemmer.stem_word('বাংলায়')
print(stemmed_word)
for each_bigram in tokens_:
if '_' in each_bigram:
#print(each_bigram)
bigram_list.append(each_bigram)
# for each_trigram in trigram_tokens_:
# #print(each_trigram)
# if each_trigram.count('_') == 2:
# trigram_list.append(each_trigram)
#print(len(bigram_list))
stemmed_bigram_count_list = []
stemmer = RafiStemmer()
stemmed_bigram_list = [
stemmer.stem_word(each_non_stemmed_bigram)
for each_non_stemmed_bigram in bigram_list
]
for each_unique_stemmed_bigram in set(stemmed_bigram_list):
stemmed_bigram_count_list.append([
each_unique_stemmed_bigram,
stemmed_bigram_list.count(each_unique_stemmed_bigram)
])
bigram_count_list = []
for each_unique_bigram in set(bigram_list):
bigram_count_list.append(
[each_unique_bigram,
from nltk.stem import PorterStemmer
from nltk.tokenize import word_tokenize
from bengali_stemmer.rafikamal2014 import RafiStemmer
"""
Stem is the form of a word before any inflectional affixes are added
it is like word
"""
ps=PorterStemmer()
example_words=["করাল","কইরালাইতেছে","নাম্বারটা","উচ্চারন","করিজাইও"]
stemmer = RafiStemmer()
for w in example_words:
print(stemmer.stem_word(w))