-
Notifications
You must be signed in to change notification settings - Fork 185
/
post_repair.py
263 lines (257 loc) · 9.96 KB
/
post_repair.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
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
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
#! -*- coding: utf-8 -*-
from __future__ import print_function
import json
import uniout
import re
f = open('test_pred.json')
F = open('test_pred2.json', 'w')
orders = ['subject', 'predicate', 'object', 'object_type', 'subject_type']
company_names = [
u'集团', u'公司', u'有限公司', u'电视台', u'集团公司', u'文化传媒有限公司', u'传媒有限公司', u'股份有限公司',
u'有限责任公司', u'工作室', u'株式会社', u'中心', u'影视传媒', u'传媒股份有限公司', u'保险有限公司'
]
n = 0
for t in f:
t = json.loads(t)
spos = []
for spo in t['spo_list']:
if spo['predicate'] == u'目' and spo['subject'][-1] in [u'科', u'属']:
# print t, spo
n += 1
continue
elif spo['predicate'] == u'国籍' and spo[
'object'] == u'中国' and u'中国共产党' in t['text'] and t[
'text'].count(u'中国') == 1:
# print t, spo
n += 1
continue
elif spo['predicate'] == u'作者' and (spo[
'object'][-2:] in [u'公司', u'学会'] or spo['object'][-3:] in [u'出版社', u'委员会']):
# print t, spo
n += 1
continue
elif spo['object'] == spo['subject'] and spo['predicate'] in [
u'父亲', u'母亲', u'丈夫', u'妻子']:
# print t, spo
n += 1
continue
elif spo['predicate'] == u'气候' and len(
spo['object']) >= 2 and t['text'][t['text'].find(spo['object'])
- 1] in [
u'南', u'北', u'亚', u'中'
]:
# print t, spo
n += 1
spo['object'] = t['text'][t['text'].find(spo['object']) -
1] + spo['object']
spos.append(spo)
elif spo['predicate'] == u'祖籍' and spo['object'][-1] == u'人':
# print t, spo
n += 1
spo['object'] = spo['object'][:-1]
spos.append(spo)
elif spo['predicate'] == u'出版社' and spo['object'][
-3:] != u'出版社' and spo['object'] + u'出版社' in t['text']:
# print t, spo
n += 1
spo['object'] = spo['object'] + u'出版社'
spos.append(spo)
elif spo['predicate'] in [
u'成立日期', u'注册资本', u'董事长', u'创始人', u'总部地点', u'占地面积', u'简称'
]:
whocare = 0
for com in company_names:
if spo['subject'][-len(com):] != com and spo[
'subject'] + com in t['text']:
# print t, spo
n += 1
spo['subject'] = spo['subject'] + com
spos.append(spo)
whocare = 1
if whocare == 0:
spos.append(spo)
elif spo['predicate'] == u'出品公司':
whocare = 0
for com in company_names:
if spo['object'][-len(com):] != com and spo[
'object'] + com in t['text']:
# print t, spo
n += 1
spo['object'] = spo['object'] + com
spos.append(spo)
whocare = 1
if whocare == 0:
spos.append(spo)
elif re.findall(u'[、，]', spo['object']):
# print t, spo
n += 1
for o in re.split(u'[、，]', spo['object']):
_ = spo.copy()
_['object'] = o
spos.append(_)
# print _
else:
spos.append(spo)
if not spos:
for spo in t['spo_list']:
if spo['predicate'] == u'目' and spo['subject'][-1] == u'属':
spos.append(spo)
break
if not spos:
for spo in t['spo_list']:
if spo['predicate'] == u'目' and spo['subject'][-1] == u'科':
spos.append(spo)
break
R_dic = {}
for spo in spos:
spo = (spo['subject'], spo['predicate'], spo['object'])
if spo[:2] not in R_dic:
R_dic[spo[:2]] = set()
R_dic[spo[:2]].add(spo[2])
R = set()
for k,v in R_dic.items():
v = list(v)
if len(v) == 2 and v[0].encode('utf-8').isalpha() and v[1].encode('utf-8').isalpha():
if '%s %s' % (v[0], v[1]) in t['text'].lower():
# print t
n += 1
R.add((k[0], k[1], '%s %s' % (v[0], v[1])))
elif '%s %s' % (v[1], v[0]) in t['text'].lower():
# print t
n += 1
R.add((k[0], k[1], '%s %s' % (v[1], v[0])))
else:
for o in v:
R.add((k[0], k[1], o))
elif len(v) == 2 and not re.findall(u'[^\u4e00-\u9fa5]', v[0]) and not re.findall(u'[^\u4e00-\u9fa5]', v[1]):
whocare = False
if v[1] in v[0]:
a, b = v[1], v[0]
i = b.find(a)
if i == 0:
if len(b) > i + len(a):
if b[i + len(a)] not in [u'和', u'的']:
whocare = True
else:
whocare = True
elif i > 0 and b[i-1] not in [u'和', u'的']:
if len(b) > i + len(a):
if b[i + len(a)] not in [u'和', u'的']:
whocare = True
else:
whocare = True
elif v[0] in v[1]:
a, b = v[0], v[1]
i = b.find(a)
if i == 0:
if len(b) > i + len(a):
if b[i + len(a)] not in [u'和', u'的']:
whocare = True
else:
whocare = True
elif i > 0 and b[i-1] not in [u'和', u'的']:
if len(b) > i + len(a):
if b[i + len(a)] not in [u'和', u'的']:
whocare = True
else:
whocare = True
if whocare:
# print t
# print k, a, b
n += 1
R.add((k[0], k[1], b))
else:
for o in v:
R.add((k[0], k[1], o))
else:
for o in v:
R.add((k[0], k[1], o))
R_dic = {}
for spo in R:
spo = spo[::-1]
if spo[:2] not in R_dic:
R_dic[spo[:2]] = set()
R_dic[spo[:2]].add(spo[2])
R = set()
for k,v in R_dic.items():
v = list(v)
if len(v) == 2 and not re.findall(u'[^\u4e00-\u9fa5]', v[0]) and not re.findall(u'[^\u4e00-\u9fa5]', v[1]):
whocare = False
if v[1] in v[0]:
a, b = v[1], v[0]
i = b.find(a)
if i == 0:
if len(b) > i + len(a):
if b[i + len(a)] not in [u'和', u'的']:
whocare = True
else:
whocare = True
elif i > 0 and b[i-1] not in [u'和', u'的']:
if len(b) > i + len(a):
if b[i + len(a)] not in [u'和', u'的']:
whocare = True
else:
whocare = True
elif v[0] in v[1]:
a, b = v[0], v[1]
i = b.find(a)
if i == 0:
if len(b) > i + len(a):
if b[i + len(a)] not in [u'和', u'的']:
whocare = True
else:
whocare = True
elif i > 0 and b[i-1] not in [u'和', u'的']:
if len(b) > i + len(a):
if b[i + len(a)] not in [u'和', u'的']:
whocare = True
else:
whocare = True
if whocare:
# print t
# print k, a, b
n += 1
R.add((k[0], k[1], b))
else:
for o in v:
R.add((k[0], k[1], o))
else:
for o in v:
R.add((k[0], k[1], o))
R = [spo[::-1] for spo in R]
R_dic = {}
for spo in R:
if spo[0] not in R_dic:
R_dic[spo[0]] = {}
R_dic[spo[0]][spo[1]] = spo[2]
for k, v in R_dic.items():
if set(v.keys()) == set([u'父亲', u'母亲']):
a, b = v[u'父亲'], v[u'母亲']
if (a, u'妻子', b) not in R:
# print t
# print (a, u'妻子', b)
R.append((a, u'妻子', b))
n += 1
if (b, u'丈夫', a) not in R:
# print t
# print (b, u'丈夫', a)
R.append((b, u'丈夫', a))
n += 1
if u'父亲' in v and u'母亲' not in v:
if u'妻子' in R_dic.get(v[u'父亲'], {}):
if k != R_dic[v[u'父亲']][u'妻子']:
R.append((k, u'母亲', R_dic[v[u'父亲']][u'妻子']))
# print t
# print (k, u'母亲', R_dic[v[u'父亲']][u'妻子'])
n += 1
elif u'母亲' in v and u'父亲' not in v:
if u'丈夫' in R_dic.get(v[u'母亲'], {}):
if k != R_dic[v[u'母亲']][u'丈夫']:
R.append((k, u'父亲', R_dic[v[u'母亲']][u'丈夫']))
# print t
# print (k, u'父亲', R_dic[v[u'母亲']][u'丈夫'])
n += 1
t['spo_list'] = [dict(zip(orders, spo + ('', ''))) for spo in R]
s = json.dumps(t, ensure_ascii=False)
F.write(s.encode('utf-8') + '\n')
print(n)