|
1
|
|
|
# -*- coding: utf-8 -*- |
|
2
|
|
|
|
|
3
|
|
|
# Copyright 2019 by Christopher C. Little. |
|
4
|
|
|
# This file is part of Abydos. |
|
5
|
|
|
# |
|
6
|
|
|
# Abydos is free software: you can redistribute it and/or modify |
|
7
|
|
|
# it under the terms of the GNU General Public License as published by |
|
8
|
|
|
# the Free Software Foundation, either version 3 of the License, or |
|
9
|
|
|
# (at your option) any later version. |
|
10
|
|
|
# |
|
11
|
|
|
# Abydos is distributed in the hope that it will be useful, |
|
12
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
13
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
14
|
|
|
# GNU General Public License for more details. |
|
15
|
|
|
# |
|
16
|
|
|
# You should have received a copy of the GNU General Public License |
|
17
|
|
|
# along with Abydos. If not, see <http://www.gnu.org/licenses/>. |
|
18
|
|
|
|
|
19
|
1 |
|
"""abydos.distance._chao_jaccard. |
|
20
|
|
|
|
|
21
|
|
|
Chao's Jaccard similarity |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
1 |
|
from __future__ import ( |
|
25
|
|
|
absolute_import, |
|
26
|
|
|
division, |
|
27
|
|
|
print_function, |
|
28
|
|
|
unicode_literals, |
|
29
|
|
|
) |
|
30
|
|
|
|
|
31
|
1 |
|
from collections import Counter |
|
32
|
|
|
|
|
33
|
1 |
|
try: |
|
34
|
1 |
|
from random import choices |
|
35
|
|
|
except ImportError: # pragma: no cover |
|
36
|
|
|
from random import choice |
|
37
|
|
|
|
|
38
|
|
|
def choices(population, k=1): |
|
39
|
|
|
"""Quick implementation of choices for Python < 3.6.""" |
|
40
|
|
|
return [choice(population) for _ in range(k)] |
|
|
|
|
|
|
41
|
|
|
|
|
42
|
|
|
|
|
43
|
1 |
|
from ._token_distance import _TokenDistance |
|
44
|
|
|
|
|
45
|
1 |
|
__all__ = ['ChaoJaccard'] |
|
46
|
|
|
|
|
47
|
|
|
|
|
48
|
1 |
|
class ChaoJaccard(_TokenDistance): |
|
49
|
|
|
r"""Chao's Jaccard similarity. |
|
50
|
|
|
|
|
51
|
|
|
Chao's Jaccard similarity :cite:`Chao:2004` |
|
52
|
|
|
|
|
53
|
|
|
.. versionadded:: 0.4.1 |
|
54
|
|
|
""" |
|
55
|
|
|
|
|
56
|
1 |
|
def __init__(self, **kwargs): |
|
57
|
|
|
"""Initialize ChaoJaccard instance. |
|
58
|
|
|
|
|
59
|
|
|
Parameters |
|
60
|
|
|
---------- |
|
61
|
|
|
**kwargs |
|
62
|
|
|
Arbitrary keyword arguments |
|
63
|
|
|
|
|
64
|
|
|
|
|
65
|
|
|
.. versionadded:: 0.4.1 |
|
66
|
|
|
|
|
67
|
|
|
""" |
|
68
|
1 |
|
super(ChaoJaccard, self).__init__(**kwargs) |
|
69
|
|
|
|
|
70
|
1 |
|
def sim(self, src, tar): |
|
71
|
|
|
"""Return normalized Chao's Jaccard similarity of two strings. |
|
72
|
|
|
|
|
73
|
|
|
Parameters |
|
74
|
|
|
---------- |
|
75
|
|
|
src : str |
|
76
|
|
|
Source string for comparison |
|
77
|
|
|
tar : str |
|
78
|
|
|
Target string for comparison |
|
79
|
|
|
|
|
80
|
|
|
Returns |
|
81
|
|
|
------- |
|
82
|
|
|
float |
|
83
|
|
|
Normalized Chao's Jaccard similarity |
|
84
|
|
|
|
|
85
|
|
|
Examples |
|
86
|
|
|
-------- |
|
87
|
|
|
>>> import random |
|
88
|
|
|
>>> random.seed(0) |
|
89
|
|
|
>>> cmp = ChaoJaccard() |
|
90
|
|
|
>>> cmp.sim('cat', 'hat') |
|
91
|
|
|
0.22448979591836735 |
|
92
|
|
|
>>> cmp.sim('Niall', 'Neil') |
|
93
|
|
|
0.1619047619047619 |
|
94
|
|
|
>>> cmp.sim('aluminum', 'Catalan') |
|
95
|
|
|
0.0 |
|
96
|
|
|
>>> cmp.sim('ATCG', 'TAGC') |
|
97
|
|
|
0.0 |
|
98
|
|
|
|
|
99
|
|
|
|
|
100
|
|
|
.. versionadded:: 0.4.1 |
|
101
|
|
|
|
|
102
|
|
|
""" |
|
103
|
1 |
|
return max(0.0, min(1.0, self.sim_score(src, tar))) |
|
104
|
|
|
|
|
105
|
1 |
|
def sim_score(self, src, tar): |
|
106
|
|
|
"""Return Chao's Jaccard similarity of two strings. |
|
107
|
|
|
|
|
108
|
|
|
Parameters |
|
109
|
|
|
---------- |
|
110
|
|
|
src : str |
|
111
|
|
|
Source string for comparison |
|
112
|
|
|
tar : str |
|
113
|
|
|
Target string for comparison |
|
114
|
|
|
|
|
115
|
|
|
Returns |
|
116
|
|
|
------- |
|
117
|
|
|
float |
|
118
|
|
|
Chao's Jaccard similarity |
|
119
|
|
|
|
|
120
|
|
|
Examples |
|
121
|
|
|
-------- |
|
122
|
|
|
>>> import random |
|
123
|
|
|
>>> random.seed(0) |
|
124
|
|
|
>>> cmp = ChaoJaccard() |
|
125
|
|
|
>>> cmp.sim_score('cat', 'hat') |
|
126
|
|
|
0.22448979591836735 |
|
127
|
|
|
>>> cmp.sim_score('Niall', 'Neil') |
|
128
|
|
|
0.1619047619047619 |
|
129
|
|
|
>>> cmp.sim_score('aluminum', 'Catalan') |
|
130
|
|
|
0.0 |
|
131
|
|
|
>>> cmp.sim_score('ATCG', 'TAGC') |
|
132
|
|
|
0.0 |
|
133
|
|
|
|
|
134
|
|
|
|
|
135
|
|
|
.. versionadded:: 0.4.1 |
|
136
|
|
|
|
|
137
|
|
|
""" |
|
138
|
1 |
|
self._tokenize(src, tar) |
|
139
|
1 |
|
self._intersection() |
|
140
|
|
|
|
|
141
|
1 |
|
if self._intersection_card() == 0: |
|
142
|
1 |
|
return 0.0 |
|
143
|
|
|
|
|
144
|
1 |
|
u_hat, v_hat = self._get_estimates(src, tar) |
|
145
|
|
|
|
|
146
|
1 |
|
num = u_hat * v_hat |
|
147
|
1 |
|
if num: |
|
148
|
1 |
|
return num / (u_hat + v_hat - u_hat * v_hat) |
|
149
|
1 |
|
return 0.0 |
|
150
|
|
|
|
|
151
|
1 |
|
def _get_estimates(self, src, tar): |
|
152
|
|
|
"""Get the estimates U-hat & V-hat used for Chao's measures. |
|
153
|
|
|
|
|
154
|
|
|
Parameters |
|
155
|
|
|
---------- |
|
156
|
|
|
src : str |
|
157
|
|
|
Source string for comparison |
|
158
|
|
|
tar : str |
|
159
|
|
|
Target string for comparison |
|
160
|
|
|
|
|
161
|
|
|
Returns |
|
162
|
|
|
------- |
|
163
|
|
|
tuple(float, float) |
|
164
|
|
|
The estimates U-hat & V-hat |
|
165
|
|
|
|
|
166
|
|
|
.. versionadded:: 0.4.1 |
|
167
|
|
|
|
|
168
|
|
|
""" |
|
169
|
1 |
|
src_card = self._src_card() # n |
|
170
|
1 |
|
tar_card = self._tar_card() # m |
|
171
|
|
|
|
|
172
|
1 |
|
src_token_list = self.params['tokenizer'].tokenize(src).get_list() |
|
173
|
1 |
|
tar_token_list = self.params['tokenizer'].tokenize(tar).get_list() |
|
174
|
|
|
|
|
175
|
1 |
|
src_sampled = Counter(choices(src_token_list, k=src_card)) |
|
176
|
1 |
|
tar_sampled = Counter(choices(tar_token_list, k=tar_card)) |
|
177
|
1 |
|
sample_intersection = src_sampled & tar_sampled |
|
178
|
|
|
|
|
179
|
1 |
|
f_1_plus = sum( |
|
180
|
|
|
1 if src_sampled[tok] == 1 and tar_sampled[tok] >= 1 else 0 |
|
181
|
|
|
for tok in sample_intersection |
|
182
|
|
|
) |
|
183
|
1 |
|
f_2_plus = sum( |
|
184
|
|
|
1 if src_sampled[tok] == 2 and tar_sampled[tok] >= 1 else 0 |
|
185
|
|
|
for tok in sample_intersection |
|
186
|
|
|
) |
|
187
|
1 |
|
if not f_2_plus: |
|
188
|
1 |
|
f_2_plus = 1 |
|
189
|
|
|
|
|
190
|
1 |
|
f_plus_1 = sum( |
|
191
|
|
|
1 if src_sampled[tok] >= 1 and tar_sampled[tok] == 1 else 0 |
|
192
|
|
|
for tok in sample_intersection |
|
193
|
|
|
) |
|
194
|
1 |
|
f_plus_2 = sum( |
|
195
|
|
|
1 if src_sampled[tok] >= 1 and tar_sampled[tok] == 2 else 0 |
|
196
|
|
|
for tok in sample_intersection |
|
197
|
|
|
) |
|
198
|
1 |
|
if not f_plus_2: |
|
199
|
1 |
|
f_plus_2 = 1 |
|
200
|
|
|
|
|
201
|
1 |
|
u_hat = 0 |
|
202
|
1 |
|
if src_card: |
|
203
|
1 |
|
u_hat += sum( |
|
204
|
|
|
src_sampled[tok] / src_card |
|
205
|
|
|
for tok in sample_intersection.keys() |
|
206
|
|
|
) |
|
207
|
1 |
|
if tar_card: |
|
208
|
1 |
|
u_hat += ( |
|
209
|
|
|
(tar_card - 1) |
|
210
|
|
|
/ tar_card |
|
211
|
|
|
* f_plus_1 |
|
212
|
|
|
/ (2 * f_plus_2) |
|
213
|
|
|
* sum( |
|
214
|
|
|
src_sampled[tok] / src_card * (tar_sampled[tok] == 1) |
|
215
|
|
|
for tok in sample_intersection.keys() |
|
216
|
|
|
) |
|
217
|
|
|
) |
|
218
|
|
|
|
|
219
|
1 |
|
v_hat = 0 |
|
220
|
1 |
|
if tar_card: |
|
221
|
1 |
|
v_hat += sum( |
|
222
|
|
|
tar_sampled[tok] / tar_card |
|
223
|
|
|
for tok in sample_intersection.keys() |
|
224
|
|
|
) |
|
225
|
1 |
|
if src_card: |
|
226
|
1 |
|
v_hat += ( |
|
227
|
|
|
(src_card - 1) |
|
228
|
|
|
/ src_card |
|
229
|
|
|
* f_1_plus |
|
230
|
|
|
/ (2 * f_2_plus) |
|
231
|
|
|
* sum( |
|
232
|
|
|
tar_sampled[tok] / tar_card * (src_sampled[tok] == 1) |
|
233
|
|
|
for tok in sample_intersection.keys() |
|
234
|
|
|
) |
|
235
|
|
|
) |
|
236
|
|
|
|
|
237
|
1 |
|
return u_hat, v_hat |
|
238
|
|
|
|
|
239
|
|
|
|
|
240
|
|
|
if __name__ == '__main__': |
|
241
|
|
|
import doctest |
|
242
|
|
|
|
|
243
|
|
|
doctest.testmod() |
|
244
|
|
|
|
chrislit /
abydos
Loading history...