/
prepare_text.py
112 lines (88 loc) · 4.04 KB
/
prepare_text.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
# Functions used in the file are based on Kaggle sample code found at
# https://github.com/wendykan/DeepLearningMovies
# This file prepares movie reviews for input by loading data from the tsv and performing
# tranformation functions
import pandas as pd
from bs4 import BeautifulSoup
import re
from nltk.corpus import stopwords
import nltk.data
import pickle
# nltk.download()
# Load punkt tokenizer
TOKENIZER = nltk.data.load('tokenizers/punkt/english.pickle')
#TOKENIZER = nltk.data.load('/home/ubuntu/nltk_data/tokenizers/punkt/english.pickle')
# lists used across files
review_list = []
test_review_list = []
def extract_data_from_tsv():
# fetch the reviews from the tsv file, both labeled and unlabeled
labeled_training_data = pd.read_csv("labeledTrainData.tsv", header=0, delimiter="\t", quoting=3) # 25,000 reviews
unlabeled_training_data = pd.read_csv("unlabeledTrainData.tsv", header=0, delimiter="\t", quoting=3) # 50,000 reviews
test_data = pd.read_csv("testData.tsv", header=0, delimiter="\t", quoting=3) # 25, 000 reviews
print "LABELED: ", labeled_training_data['review'].size()
print "UNLABELED: ", unlabeled_training_data['review'].size()
print "TEST: ", test_data['review'].size()
class MovieReview(object):
def __init__(self, mreview):
self.mreview = mreview
self.mreview_clean = None
self.mreview_word_list = []
self.mreview_sentence_list = []
def clean_review(self):
# function to clean the review by stripping html from review text body
self.mreview_clean = BeautifulSoup(self.mreview).get_text()
def remove_punctuation_and_nums(self):
self.mreview_clean = re.sub("[^a-zA-Z]", " ", self.mreview_clean)
def split_review_into_words(self):
# function to split the review text to list of words
self.mreview_word_list = self.mreview_clean.lower().split()
def remove_stop_words(self):
self.mreview_word_list = [word for word in self.mreview_word_list if not word in set(stopwords.words("english"))]
self.mreview_clean = " ".join(self.mreview_word_list)
def split_review_into_sentences(self):
# function to split review into list of sentences
# where each setence is a list of words
extracted_sentences = TOKENIZER.tokenize(self.mreview_clean.strip())
for extracted_sentence in extracted_sentences:
if len(extracted_sentence) > 0:
# extracted_sentence needs to be operated on if stopword or punctuation
# removal is required eventually(not required for word2Vec)
self.mreview_sentence_list.append(extracted_sentence.lower().split())
def main():
global review_list
global test_list
labeled_training_data = pd.read_csv("labeledTrainData.tsv", header=0, delimiter="\t", quoting=3) # 25,000 reviews
# unlabeled_training_data = pd.read_csv("unlabeledTrainData.tsv", header=0, delimiter="\t", quoting=3) # 50,000 reviews
test_data = pd.read_csv("testData.tsv", header=0, delimiter="\t", quoting=3) # 25, 000 reviews
# print "LABELED: ", labeled_training_data['review'].size()
# print "UNLABELED: ", unlabeled_training_data['review'].size()
# print "TEST: ", test_data['review'].size()
for mreview in labeled_training_data["review"]:
mreview_obj = MovieReview(mreview)
mreview_obj.clean_review()
mreview_obj.remove_punctuation_and_nums()
mreview_obj.split_review_into_words()
mreview_obj.remove_stop_words()
review_list.append(mreview_obj.mreview_clean)
for mreview in test_data["review"]:
mreview_obj = MovieReview(mreview)
mreview_obj.clean_review()
mreview_obj.remove_punctuation_and_nums()
mreview_obj.split_review_into_words()
mreview_obj.remove_stop_words()
test_review_list.append(mreview_obj.mreview_clean)
print "Finished"
print "Pickle a list??"
print "Pickle"
with open("review_file_pickle.txt", 'wb') as fp1:
pickle.dump(review_list, fp1)
with open("test_review_file_pickle.txt", 'wb') as fp2:
pickle.dump(test_review_list, fp2)
print "Write to a .py file"
with open('review_file.py', 'w') as f1:
f1.write('review_list = %s' % review_list)
with open('test_review_file.py', 'w') as f2:
f2.write('test_review_list = %s' % test_review_list)
if __name__ == '__main__':
main()