-
Notifications
You must be signed in to change notification settings - Fork 0
/
Posi-Nega-Neutra_Tagged-Sentence-Parsing.py
executable file
·121 lines (115 loc) · 3.99 KB
/
Posi-Nega-Neutra_Tagged-Sentence-Parsing.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
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
#-*- coding: utf-8 -*-
####################################
########## Training Data ##########
####################################
def PosiNegaSentences(comment):
'''
Input is "word word <positive> word word word </positive> <negative> word word word </negative> word ..."
Returns a list of sentences in the format ['word word word',1] if positive or ['word word word', 0] if negative
'''
comment = comment.split()
# The comment has been reduced to a list of words including the tags
sentences = []
sentence = []
att = 0
positive = False
negative = False
for current_word in comment:
skip = False
if (current_word == '<positive>'):
positive = True
negative = False
skip = True
if (current_word == '</positive>'):
positive = True
skip = True
if (current_word == '<negative>'):
positive = False
negative = True
skip = True
if (current_word == '</negative>'):
negative = True
skip = True
if not skip:
if positive or negative:
sentence.append(current_word)
else:
if len(sentence)>0:
if positive and not negative:
att = 1
sentence = " ".join(sentence).strip()
sentences.append([sentence,att])
positive = False
elif negative and not positive:
att = 0
sentence = " ".join(sentence).strip()
sentences.append([sentence,att])
negative = False
sentence = []
return sentences
def PosiNegaNeutraSentences(comment):
'''
Input is "<neutral> word word </neutral> <positive> word word word </positive> <negative> word word word </negative> word ..."
Returns a list of sentences in the format
['word word word', 1] if positive
or ['word word word', 0] if neutral
or ['word word word', -1] if negative
'''
comment = comment.split()
# The comment has been reduced to a list of words including the tags
sentences = []
sentence = []
att = 0
positive = False
negative = False
neutral = False
for current_word in comment:
skip = False
if (current_word == '<positive>'):
positive = True
negative = False
neutral = False
skip = True
if (current_word == '</positive>'):
positive = True
skip = True
if (current_word == '<negative>'):
positive = False
neutral = False
negative = True
skip = True
if (current_word == '</negative>'):
negative = True
skip = True
if (current_word == '<neutral>'):
positive = False
neutral = True
negative = False
skip = True
if (current_word == '</neutral>'):
neutral = True
skip = True
if not skip:
if positive or negative or neutral:
sentence.append(current_word)
else:
if len(sentence)>0:
if positive and not negative and not neutral:
att = 1
sentence = " ".join(sentence).strip()
sentences.append([sentence,att])
positive = False
elif negative and not positive and not neutral:
att = -1
sentence = " ".join(sentence).strip()
sentences.append([sentence,att])
negative = False
elif neutral and not positive and not negative:
att = 0
sentence = " ".join(sentence).strip()
sentences.append([sentence,att])
neutral = False
sentence = []
return sentences
if __name__ == '__main__':
pass