keywords/input.py at master · matthewberryman/keywords · GitHub

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
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
#!/usr/local/bin/python3

import random, itertools, sys
from textblob import TextBlob


def load_set(filenames): # for loading a test or training set
  set = []
  for filename in filenames:
    for line in open(filename,'r'):
      line = line.rstrip()
      if line.endswith(',-1'):
        set.append((line[0:-3],'neg'))
      elif line.endswith(',1'):
        set.append((line[0:-2],'pos'))
      elif line.endswith(',0'):
        set.append((line[0:-2],'neu'))
  return set


def read_anzns(filename):
  title = ''
  seen = set()
  storing = False
  fulltext = ''
  year = ''
  location = ''
  store = {}

  for line in open(filename,'r'):
    line = line.rstrip()
    if line.startswith('Full text:'):
      storing = True
      fulltext = line[10:].lower()
    else:
      if storing:
        fulltext += ' ' + line
    if line == '':
      storing = False
#    if line.startswith('Subject:') or line.startswith('Credit:'):
#      storing = False
#      fulltext = ''

    if line.startswith('Title:'):
      title = line[7:]

    if line.startswith('Publication year:'):
      year = line[18:]

    if line.startswith('Place of publication:'):
      location = line[22:]

    if len(fulltext) > 1 and not storing:
        if title not in seen:
          print(location)
          by_year = {}
          try:
            by_year = store[location]
          except KeyError:
            store[location] = {}
          try:
            by_year[year].append(fulltext)
          except KeyError:
            by_year[year] = [fulltext]
          store[location] = by_year
          fulltext = ''
        else:
          seen.add(title)
        fulltext = ''
  return store

# Next function needs further testing.
def factiva(filename):
  storing = False
  dateLineNext = False
  fulltext = ''
  year = ''
  for line in open(filename,'r'):
    if line.endswith('words'):
      dateLineNext = True
      storing = True

    if dateLineNext:
      year = line[-4:]
      dateLineNext = False;

    if storing:
      fulltext += line

    if line.startswith('Document'):
      storing = False
      store.append((year,fulltext))
      fulltext = ''

    return store

# Generate input data

if __name__ == '__main__':

  filename = ''
  if len(sys.argv) > 1:
    filename = sys.argv[1]
  else:
    filename = 'anzns.txt'
  articles = read_anzns(filename)

  wollongong_articles = []
  kiama_articles = []
  other_articles = []

  wollongong_test = open('wollongong_test.txt','w')
  wollongong_training = open('wollongong_training.txt','w')
  kiama_test = open('kiama_test.txt','w')
  kiama_training = open('kiama_training.txt','w')
  other_test = open('other_test.txt','w')
  other_training = open('other_training.txt','w')

  for location in articles.keys():
    if location.startswith('Wollongong'):
      for article in [item for sublist in articles[location].values() for item in sublist]:
        wollongong_articles.append(article)
    elif location.startswith('Kiama'):
      for article in [item for sublist in articles[location].values() for item in sublist]:
        kiama_articles.append(article)
    else:
      for article in [item for sublist in articles[location].values() for item in sublist]:
        other_articles.append(article)

  for i in range(88):
    wollongong_article = TextBlob(random.choice(wollongong_articles))
    for sentence in wollongong_article.sentences:
      if 'NBN' in sentence:
        if i < 44:
          wollongong_test.write(str(sentence) + ',\n')
        if i > 44:
          wollongong_training.write(str(sentence) + ',\n')

  for i in range(min(len(kiama_articles),88)):
    print(str(i))
    kiama_article = TextBlob(kiama_articles[i])
    for sentence in kiama_article.sentences:
      if 'NBN' in sentence:
        if i < 44:
          kiama_test.write(str(sentence) + ',\n')
        if i > 44:
          kiama_training.write(str(sentence) + ',\n')

  for i in range(min(len(other_articles),88)):
    other_article = TextBlob(random.choice(other_articles))
    for sentence in other_article.raw_sentences:
      if 'NBN' in sentence:
        if i < 44:
          other_test.write(str(sentence) + ',\n')
        if i > 44:
          other_training.write(str(sentence) + ',\n')