-
Notifications
You must be signed in to change notification settings - Fork 0
/
auxiliar_module.py
61 lines (55 loc) · 1.47 KB
/
auxiliar_module.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
from nltk.stem import WordNetLemmatizer
import nltk.tokenize
from nltk.corpus import stopwords
import numpy as np
stopwords = stopwords.words('english')
lemmatizer = WordNetLemmatizer()
f=open('emojis','r')
emojis=f.read().split(',')
emojis=[emojis[0].split(' '),emojis[1].split(' ')]
f.close()
def remove_emojis(tweet,i):
aux=tweet
emoji_score=0
for emoji in emojis[i]:
if aux.startswith(emoji) or aux.endswith(emoji) or (' '+emoji+' ') in aux:
emoji_score+=1
aux = aux.replace(emoji,'')
if i==0:
emoji_score*=-1
return aux,emoji_score
def clean_tweet(tweet,rmquery=''):
aux=tweet.lower()
if rmquery:
aux=aux.replace(rmquery.lower(),'')
if tweet.startswith('RT '):
aux=aux[3:]
usus=aux.split('@')
aux=usus[0]
for w in usus[1:]:
try:
aux+=w[w.index(' ')+1:]
except(ValueError):
pass
tweet_words=[]
for w in nltk.word_tokenize(aux):
splittedw = w.split('-')
is_composed = len(splittedw)>1
if is_composed:
for splittedpart in splittedw:
is_composed = is_composed and splittedpart.isalpha()
if (w.isalpha() or is_composed) and w not in stopwords and 'http' not in w and len(w)>1:
lemma=lemmatizer.lemmatize(w)
tweet_words.append(lemma)
return tweet_words
def rellenar_arrays(t,pals):
tset=np.zeros((len(t),len(pals)),dtype=bool)
for i in range(len(t)):
auxset=set(t[i])
for w in auxset:
try:
ind=pals.index(w)
tset[i,ind]=True
except(ValueError):
pass
return tset