def removeEmphasis(sentence):
sentence = sentence.split()
no_emphasis = [remove_emphasis(x) for x in sentence]
return (no_emphasis)
file = path.join('data','nsk_scrape.xlsx')
xl = pd.ExcelFile(file)
df = xl.parse('Sheet1')
df.head()
corpus = []
STOPWORDS = set(stopwords.words('greek'))
#Επεξεργασία ΓΝΩΜΟΔΟΤΗΣΕΩΝ
print(df.shape[0])
for i in range(0, df.shape[0]):
subject = re.sub(r"\d+", '', df['Concultatory'][i],flags=re.I)
subject = re.sub(r"[-,()/@\'?\.$%_+\d]", '', df['Concultatory'][i],flags=re.I)
stemmer = gr_stemm.GreekStemmer()
subject = subject.split()
subject = [remove_emphasis(x) for x in subject]
subject = [x.upper() for x in subject]
subject = [stemmer.stem(word) for word in subject if not word in STOPWORDS and len(word)>=3]
subject = [x.lower() for x in subject]
subject = " ".join(subject)
corpus.append(subject)
#words_ = word_tokenize(subject)
corpus=pd.DataFrame(corpus, columns=['Concultatory'])
corpus.head()
result = corpus.join(df[['Status']])
result.groupby(['Status']).size()
result.head()
def removeEmphasis(word):
word = word.split()
no_emphasis = [remove_emphasis(x) for x in word]
return (no_emphasis)