add_additional_ne.py
9.63 KB
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
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
import argparse
import os
import re
REGEXPS = {
'EMAIL': '^[a-zA-Z0-9_.+-]+@[a-zA-Z0-9-]+\.[a-zA-Z0-9-.]+$',
'URL': '^(?:(?:https?|ftp|sftp|file)://|(www|ftp)\.)[-A-Za-z0-9+&@#/%?=~_|$!:,.;()]*$',
'IP': '^(?:(?:25[0-5]|2[0-4]\d|1?\d?\d)(?:\.(?!$)|$)){4}$',
'IBAN': '^(AD\d{10}[A-Z0-9]{12}|AE\d{21}|AL\d{10}[A-Z0-9]{16}|AT\d{18}|AZ\d{2}[A-Z]{4}[A-Z0-9]{20}|BA\d{18}|'
'BE\d{14}|BG\d{2}[A-Z]{4}\d{6}[A-Z0-9]{8}|BH\d{2}[A-Z]{4}[A-Z0-9]{14}|BR\d{25}[A-Z]{1}[A-Z0-9]{1}|'
'BY\d{2}[A-Z0-9]{4}\d{4}[A-Z0-9]{16}|CH\d{7}[A-Z0-9]{12}|CR\d{20}|CY\d{10}[A-Z0-9]{16}|CZ\d{22}|DE\d{20}|'
'DK\d{16}|DO\d{2}[A-Z0-9]{4}\d{20}|EE\d{18}|EG\d{27}|ES\d{22}|FI\d{16}|FO\d{16}|FR\d{12}[A-Z0-9]{11}\d{2}|'
'GB\d{2}[A-Z]{4}\d{14}|GE\d{2}[A-Z]{2}\d{16}|GI\d{2}[A-Z]{4}[A-Z0-9]{15}|GL\d{16}|GR\d{9}[A-Z0-9]{16}|'
'GT\d{2}[A-Z0-9]{24}|HR\d{19}|HU\d{26}|IE\d{2}[A-Z]{4}\d{14}|IL\d{21}|IQ\d{2}[A-Z]{4}\d{15}|IS\d{24}|'
'IT\d{2}[A-Z]{1}\d{10}[A-Z0-9]{12}|JO\d{2}[A-Z]{4}\d{4}[A-Z0-9]{18}|KW\d{2}[A-Z]{4}[A-Z0-9]{22}|'
'KZ\d{5}[A-Z0-9]{13}|LB\d{6}[A-Z0-9]{20}|LC\d{2}[A-Z]{4}[A-Z0-9]{24}|LI\d{7}[A-Z0-9]{12}|LT\d{18}|'
'LU\d{5}[A-Z0-9]{13}|LV\d{2}[A-Z]{4}[A-Z0-9]{13}|MC\d{12}[A-Z0-9]{11}\d{2}|MD\d{2}[A-Z0-9]{20}|ME\d{20}|'
'MK\d{5}[A-Z0-9]{10}\d{2}|MR\d{25}|MT\d{2}[A-Z]{4}\d{5}[A-Z0-9]{18}|MU\d{2}[A-Z]{4}\d{19}[A-Z]{3}|'
'NL\d{2}[A-Z]{4}\d{10}|NO\d{13}|PK\d{2}[A-Z]{4}[A-Z0-9]{16}|PL\d{26}|PS\d{2}[A-Z]{4}[A-Z0-9]{21}|PT\d{23}|'
'QA\d{2}[A-Z]{4}[A-Z0-9]{21}|RO\d{2}[A-Z]{4}[A-Z0-9]{16}|RS\d{20}|SA\d{4}[A-Z0-9]{18}|'
'SC\d{2}[A-Z]{4}\d{20}[A-Z]{3}|SE\d{22}|SI\d{17}|SK\d{22}|SM\d{2}[A-Z]{1}\d{10}[A-Z0-9]{12}|ST\d{23}|'
'SV\d{2}[A-Z]{4}\d{20}|TL\d{21}|TN\d{22}|TR\d{8}[A-Z0-9]{16}|UA\d{8}[A-Z0-9]{19}|VA\d{20}|'
'VG\d{2}[A-Z]{4}\d{16}|XK\d{18})$',
# # vat/tax number https://www.oreilly.com/library/view/regular-expressions-cookbook/9781449327453/ch04s21.html
# 'ID': '^((AT)?U[0-9]{8}|(BE)?0[0-9]{9}|(BG)?[0-9]{9,10}|(CY)?[0-9]{8}L|(CZ)?[0-9]{8,10}|(DE)?[0-9]{9}|(DK)?[0-9]{8}|'
# '(EE)?[0-9]{9}|(EL|GR)?[0-9]{9}|(ES)?[0-9A-Z][0-9]{7}[0-9A-Z]|(FI)?[0-9]{8}|(FR)?[0-9A-Z]{2}[0-9]{9}|'
# '(GB)?([0-9]{9}([0-9]{3})?|[A-Z]{2}[0-9]{3})|(HU)?[0-9]{8}|(IE)?[0-9]S[0-9]{5}L|(IT)?[0-9]{11}|'
# '(LT)?([0-9]{9}|[0-9]{12})|(LU)?[0-9]{8}|(LV)?[0-9]{11}|(MT)?[0-9]{8}|(NL)?[0-9]{9}B[0-9]{2}|(PL)?[0-9]{10}|'
# '(PT)?[0-9]{9}|(RO)?[0-9]{2,10}|(SE)?[0-9]{12}|(SI)?[0-9]{8}|(SK)?[0-9]{10})|'
# # personal identification https://ipsec.pl/european-personal-data-regexp-patterns.html
# '[0-9]{12}|[0-9]{10}|[A-Za-z0-9+/]{22}[A-Za-z0-9+/=][A-Za-z0-9+/=]|'
# '[0-9]{2}\.?[0-9]{2}\.?[0-9]{2}-[0-9]{3}\.?[0-9]{2}|[0-9]{2}[0,1,2,4][0-9][0-9]{2}[0-9]{4}|'
# '[0-9]{2}[0,1,5][0-9][0-9]{2}/?[0-9]{4}|[A-Z]{2}[0-9]{6}|[0-9]{2}[0,1][0-9][0-9]{2}-[0-9]{4}|'
# '[1-6][0-9]{2}[1,2][0-9][0-9]{2}[0-9]{4}|[A-Z]{2}?[ ]?[0-9]{2}[ ]?[0-9]{4}[ ]?[0-9]{4}[ ]?[0-9]{4}[ ]?'
# '[0-9]{4}[ ]?[0-9]{4}|[0-9]{2}\.?[0,1][0-9]\.?[0-9]{2}[-+A][0-9]{3}[A-Z]|'
# '[1,2][ ]?[0-9]{2}[ ]?[0,1,2,3,5][0-9][ ]?[0-9A-Z]{5}[ ]?[0-9]{3}[ ]?[0-9]{2}|'
# '[0-9]{2}[0,1][0-9][0-9]{2}-[A-Z]-[0-9]{5}|[0-9]{3}/?[0-9]{4}/?[0-9]{4}|'
# '[0-9]{2}[0-9]{2}[0,1][0-9][0-9]{2}[A-Z][0-9]{2}[0-9]|[A-Z][ -]?[0-9]{6}|[0-9]{3}[ ]?[0-9]{3}[ ][0-9]{3}|'
# '[1-8][ ]?[0-9]{2}[0,1][0-9][0-9]{2}[ ]?[0-9]{4}|[0-9]{7}[A-Z]W?|'
# '[A-Z]{6}[0-9]{2}[A-E,H,L,M,P,R-T][0-9]{2}[A-Z0-9]{5}|'
# '[0-9]{2}[0,1][0-9][0-9]-[0-9]{5}|[3-6][0-9]{2}[0,1][0-9][0-9]{2}[0-9]{4}|[0-9]{9}|'
# '[0-9]{2}[0,1][0-9][0-9]{2}[ ]?[0-9]{5}|[0-9]{4}[0-3]{1}[0-9}{1}[0-9]{5}|'
# '[1-8][0-9]{2}[0,1][0-9][0-9]{2}[0-9]{6}|[0-9,X,M,L,K,Y][0-9]{7}[A-Z]|[0-9]{2}[0-1][0-9][0-9]{2}[-+][0-9]{4}|'
# '[0-9]{3}\.?[0-9]{2}\.?[0-9]{3}\.?[0-9]{3}|756\.?[0-9]{4}\.?[0-9]{4}\.?[0-9]{2}|'
# '[A-CEGHJ-PR-TW-Z][A-CEGHJ-NPR-TW-Z]{1}[0-9]{6}[A-DFM]?|^([ACEHJLMOPRSW-Yacehjlmoprsw-y]'
# '[A-CEGHJ-NPRSTW-Za-ceghj-nprstw-z]|[Bb][A-CEHJ-NPRSTW-Za-cehj-nprstw-z]|[Gg][ACEGHJ-NPRSTW-Zaceghj-nprstw-z]'
# '|[Kk][A-CEGHJ-MPRSTW-Za-ceghj-mprstw-z]|[Nn][A-CEGHJLMNPRSW-Za-ceghjlmnprsw-z]|'
# '[Tt][A-CEGHJ-MPRSTW-Za-ceghj-mprstw-z]|[Zz][A-CEGHJ-NPRSTW-Ya-ceghj-nprstw-y])[0-9]{6}[A-Da-d ]?$|'
# '[0-9]{3}[ -]?[0-9]{3}[ -]?[0-9]{4}$',
# # https://ihateregex.io/expr/phone/
'PHONE': '^[\+]?[(]?[0-9]{3}[)]?[-\s\.]?[0-9]{3}[-\s\.]?[0-9]{4,6}$',
# 'GPS': '^(-?[1-8]?\d(?:\.\d{1,18})?|90(?:\.0{1,18})?),\s+?(-?(?:1[0-7]|[1-9])?\d(?:\.\d{1,18})?|180(?:\.0{1,18})?)$'
}
def main():
args = parse_arguments()
if not args.input or not args.output:
print('Error: Input and output must be selected!')
add_ne_by_year(args.input, args.output)
def parse_arguments():
parser = argparse.ArgumentParser(description='Add additional NE to the corpora.')
parser.add_argument('-o', '--output', help='output directory')
required_arguments = parser.add_argument_group('required arguments')
required_arguments.add_argument('-i', '--input', help='corpora root directory', required=True)
return parser.parse_args()
def add_ne_by_year(root_directory, out_corpora_directory):
for root, dirs, files in os.walk(root_directory):
for filename in files:
if filename.endswith('.conllup') or filename.endswith('.conllu'):
src = os.path.join(root, filename)
year = get_year(src)
year_path = os.path.join(out_corpora_directory, year)
os.makedirs(year_path, exist_ok=True)
dst = os.path.join(year_path, filename)
add_nes(src, dst)
def get_year(filepath):
with open(filepath, 'r') as conllup_file:
for line in conllup_file:
line = line.strip()
if is_segment(line):
continue
elif is_metadata(line):
name, value = get_metadata(line)
if name == 'PublicationDate':
return value.split('-')[0]
return '0'
def is_segment(line):
if line and line[0].isdigit():
return True
return False
def is_metadata(line):
if line.startswith('#'):
return True
return False
def get_metadata(line):
name_value_pair = line.split('=', 1)
name = name_value_pair[0].lstrip('#').strip()
value = name_value_pair[1].strip()
return name, value
def add_nes(src, dst):
cleaned_lines = []
paragraph = []
sentence = []
with open(src, 'r') as conllup_file:
for line in conllup_file:
line = line.strip()
if is_segment(line):
sentence.append(line)
elif is_metadata(line):
name, value = get_metadata(line)
if name == 'newpar id':
if sentence:
add_additional_nes_to_sentence(sentence)
sentence.append('')
paragraph.extend(sentence)
sentence = []
if len(paragraph) > 1:
cleaned_lines.extend(paragraph)
paragraph = [line]
elif name == 'sent_id':
if sentence:
add_additional_nes_to_sentence(sentence)
sentence.append('')
paragraph.extend(sentence)
sentence = [line]
elif name == 'text':
sentence.append(line)
else:
cleaned_lines.append(line)
if sentence:
add_additional_nes_to_sentence(sentence)
sentence.append('')
paragraph.extend(sentence)
if len(paragraph) > 1:
cleaned_lines.extend(paragraph)
if not cleaned_lines[-1]:
cleaned_lines.append('')
if not cleaned_lines[-1].strip():
cleaned_lines.pop()
with open(dst, 'w') as dst_file:
dst_file.write('\n'.join(cleaned_lines))
def add_additional_nes_to_sentence(sentence):
for i in range(2, len(sentence)+1):
for j in reversed(range(i + 1, len(sentence)+1)):
tokens = sentence[i:j]
if not contains_ne(tokens):
text = tokens_to_orth(tokens)
ne_name = check_for_ne(text)
if ne_name:
sentence[i:j] = mark_ne(tokens, ne_name)
break
def contains_ne(tokens):
for tok in tokens:
columns = tok.split('\t')
ne = columns[10]
if ne != 'O':
return True
return False
def tokens_to_orth(tokens):
text = ''
for tok in tokens:
columns = tok.split('\t')
orth = columns[1]
nsa = columns[9] == 'SpaceAfter=No'
if nsa:
text += orth
else:
text += f'{orth} '
return text.strip()
def check_for_ne(text):
for name, expression in REGEXPS.items():
if re.search(expression, text):
return name
return ''
def mark_ne(tokens, ne_name):
first = True
marked_tokens = []
for it, tok in enumerate(tokens):
columns = tok.split('\t')
if it == len(tokens)-1 and columns[3] == 'PUNCT' and columns[2] not in ['/', ')']:
print(ne_name, 'removed col', columns)
marked_tokens.append('\t'.join(columns))
else:
if first:
columns[10] = f'B-{ne_name}'
else:
columns[10] = f'I-{ne_name}'
marked_tokens.append('\t'.join(columns))
first = False
return marked_tokens
if __name__ == '__main__':
main()