forked from cfiltnlp/pyiwn
-
Notifications
You must be signed in to change notification settings - Fork 0
/
iwn.py
282 lines (234 loc) · 8.79 KB
/
iwn.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
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
from enum import Enum, unique
import re
import logging
import glob
import ntpath
import os
import pandas as pd
import pyiwn.constants as constants
logging.basicConfig(format='[%(filename)s:%(lineno)d] %(message)s',
datefmt='%Y-%m-%d:%H:%M:%S',
level=logging.INFO)
logger = logging.getLogger(__name__)
@unique
class Language(Enum):
ASSAMESE = 'assamese'
BENGALI = 'bengali'
BODO = 'bodo'
GUJARATI = 'gujarati'
HINDI = 'hindi'
KANNADA = 'kannada'
KASHMIRI = 'kashmiri'
KONKANI = 'konkani'
MALAYALAM = 'malayalam'
MARATHI = 'marathi'
MEITEI = 'meitei'
NEPALI = 'nepali'
ORIYA = 'oriya'
PUNJABI = 'punjabi'
SANSKRIT = 'sanskrit'
TAMIL = 'tamil'
TELUGU = 'telugu'
URDU = 'urdu'
class IndoWordNet:
def __init__(self, lang=Language.HINDI):
logger.info(f'Loading {lang.value} language synsets...')
self._synset_idx_map = {}
self._synset_df = self._load_synset_file(lang.value)
self._synset_relations_dict = self._load_synset_relations()
def _load_synset_file(self, lang):
filename = os.path.join(
*[constants.IWN_DATA_PATH, 'synsets', 'all.{}'.format(lang)])
f = open(filename, encoding='utf8')
synsets = list(
map(lambda line: self._load_synset(line), f.readlines()))
synset_df = pd.DataFrame(
synsets, columns=['synset_id', 'synsets', 'pos'])
synset_df = synset_df.dropna()
synset_df = synset_df.set_index('synset_id')
return synset_df
def _load_synset_relations(self):
relations_dict = {}
for file_path, relation_name in self._relation_list():
relations_dict[relation_name] = []
d = {}
for line in open(file_path):
line_parts = line.split('\t')
synset_id, synset_ids = line_parts
synset_id = int(synset_id)
synset_ids = list(map(int, synset_ids.split(',')))
synset_ids = list(
filter(lambda x: True if x in self._synset_df.index else False, synset_ids))
if synset_id in d:
d[synset_id].extend(synset_ids)
else:
if synset_ids:
d[synset_id] = synset_ids
relations_dict[relation_name] = d
return relations_dict
def _relation_list(self, type='synset_relations'):
relations = []
path_parts = '{},{},*'.format(constants.IWN_DATA_PATH, type).split(',')
for file_path in glob.glob(os.path.join(*path_parts)):
file_name = ntpath.basename(file_path)
file_name_parts = file_name.split('.')
if len(file_name_parts) != 2:
continue
relation_name, pos_tag = file_name_parts
relations.append((file_path, relation_name))
return relations
def _update_synset_idx_map(self, synset):
synset_id = synset.synset_id()
for word in synset.lemma_names():
if word in self._synset_idx_map:
self._synset_idx_map[word].append(synset_id)
else:
self._synset_idx_map[word] = [synset_id]
return True
def _load_synset(self, synset_string):
if 'null' in synset_string:
return None, None, None
synset_string = synset_string.replace('\n', '').strip()
synset_pattern = '([0-9]+)\t(.+)\t(.+)\t([a-zA-Z]+)'
try:
matches = re.findall(synset_pattern, synset_string)
synset_id, synset_words, gloss_examples, pos = matches[0]
except Exception as e:
return None, None, None
synset_id = int(synset_id)
synset_words = list(filter(lambda x: False if x ==
'' else True, synset_words.split(',')))
if not synset_words:
return None, None, None
head_word = synset_words[0]
if gloss_examples != '':
if ':"' in gloss_examples:
ge_list = gloss_examples.split(':')
gloss = ge_list[0]
if len(ge_list) > 1:
examples = ''.join(ge_list[1:])
examples = re.sub('["]', '', examples)
examples = examples.split(' / ')
else:
examples = []
else:
gloss = gloss_examples
examples = []
else:
return None, None, None
synset = Synset(synset_id, head_word,
synset_words, pos, gloss, examples)
self._update_synset_idx_map(synset)
return synset_id, synset, pos
def all_synsets(self, pos=None):
if pos is None:
result = self._synset_df
else:
mask = (self._synset_df.pos == pos.value)
result = self._synset_df[mask]
return list(result['synsets'].values)
def synsets(self, word, pos=None):
try:
synset_id_list = self._synset_idx_map[word]
except:
return []
synsets = []
if pos is not None:
for synset_id in synset_id_list:
synset = self._synset_df.loc[[synset_id]]['synsets'].values[0]
if synset.pos() == pos.value:
synsets.append(synset)
else:
for synset_id in synset_id_list:
synset = self._synset_df.loc[[synset_id]]['synsets'].values[0]
synsets.append(synset)
return synsets
def all_words(self, pos=None):
if pos is None:
return list(self._synset_idx_map.keys())
words = set()
mask = (self._synset_df.pos == pos.value)
for synset in self._synset_df[mask]['synsets'].values:
for word in synset.lemma_names():
words.add(word)
words = list(words)
return words
def synset_relation(self, synset, relation):
return list(self._synset_df[self._synset_df.index.isin(self._synset_relations_dict[relation.value].get(synset.synset_id(), []))]['synsets'])
class Synset:
def __init__(self, synset_id, head_word, lemma_names, pos, gloss, examples):
self._synset_id = synset_id
self._head_word = head_word
self._lemma_names = lemma_names
self._pos = pos
self._gloss = gloss
self._examples = examples
def __repr__(self):
return 'Synset(\'{}.{}.{}\')'.format(self._head_word, self._pos, self._synset_id)
def synset_id(self):
return self._synset_id
def head_word(self):
return self._head_word
def lemma_names(self):
return self._lemma_names
def lemmas(self):
return [Lemma(self, lemma) for lemma in self._lemma_names]
def pos(self):
return self._pos
def gloss(self):
return self._gloss
def examples(self):
return self._examples
def ontology_nodes(self):
raise NotImplementedError("This method will be implemented soon.")
class Lemma:
def __init__(self, synset, name):
self._synset = synset
self._name = name
def __repr__(self):
return 'Lemma(\'{}.{}.{}.{}\')'.format(self._synset.head_word(), self._synset.pos(), self._synset.synset_id(), self._name)
def name(self):
return self._name
def synset(self):
return self._synset
def gradation(self):
raise NotImplementedError("This method will be implemented soon.")
def antonym(self):
raise NotImplementedError("This method will be implemented soon.")
@unique
class PosTag(Enum):
NOUN = 'noun'
VERB = 'verb'
ADVERB = 'adverb'
ADJECTIVE = 'adjective'
class IndoWordNetError(Exception):
""" An exception class for IndoWordNet-related errors. """
@unique
class SynsetRelations(Enum):
MERO_MEMBER_COLLECTION = 'mero_member_collection'
ABILITY_VERB = 'ability_verb'
CAUSATIVE = 'causative'
CAPABILITY_VERB = 'capability_verb'
MERO_COMPONENT_OBJECT = 'mero_component_object'
HOLO_PORTION_MASS = 'holo_portion_mass'
FUNCTION_VERB = 'function_verb'
HOLO_COMPONENT_OBJECT = 'holo_component_object'
HYPERNYMY = 'hypernymy'
ENTAILMENT = 'entailment'
ALSO_SEE = 'also_see'
MERO_FEATURE_ACTIVITY = 'mero_feature_activity'
HOLO_PLACE_AREA = 'holo_place_area'
MODIFIES_VERB = 'modifies_verb'
ATTRIBUTES = 'attributes'
MERO_PORTION_MASS = 'mero_portion_mass'
MODIFIES_NOUN = 'modifies_noun'
HOLO_FEATURE_ACTIVITY = 'holo_feature_activity'
MERO_STUFF_OBJECT = 'mero_stuff_object'
TROPONYMY = 'troponymy'
MERO_PLACE_AREA = 'mero_place_area'
HOLO_MEMBER_COLLECTION = 'holo_member_collection'
HYPONYMY = 'hyponymy'
SIMILAR = 'similar'
MERO_POSITION_AREA = 'mero_position_area'
HOLO_POSITION_AREA = 'holo_position_area'
HOLO_STUFF_OBJECT = 'holo_stuff_object'