|
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._scott_pi. |
|
20
|
|
|
|
|
21
|
|
|
Scott's Pi correlation |
|
22
|
|
|
""" |
|
23
|
|
|
|
|
24
|
1 |
|
from __future__ import ( |
|
25
|
|
|
absolute_import, |
|
26
|
|
|
division, |
|
27
|
|
|
print_function, |
|
28
|
|
|
unicode_literals, |
|
29
|
|
|
) |
|
30
|
|
|
|
|
31
|
1 |
|
from ._token_distance import _TokenDistance |
|
32
|
|
|
|
|
33
|
1 |
|
__all__ = ['ScottPi'] |
|
34
|
|
|
|
|
35
|
|
|
|
|
36
|
1 |
|
class ScottPi(_TokenDistance): |
|
37
|
|
|
r"""Scott's Pi correlation. |
|
38
|
|
|
|
|
39
|
|
|
For two sets X and Y and a population N, Scott's :math:`\pi` correlation |
|
40
|
|
|
:cite:`Scott:1955` is |
|
41
|
|
|
|
|
42
|
|
|
.. math:: |
|
43
|
|
|
|
|
44
|
|
|
corr_{Scott_\pi}(X, Y) = \pi = |
|
45
|
|
|
\frac{p_o - p_e^\pi}{1 - p_e^\pi} |
|
46
|
|
|
|
|
47
|
|
|
where |
|
48
|
|
|
|
|
49
|
|
|
.. math:: |
|
50
|
|
|
|
|
51
|
|
|
\begin{array}{ll} |
|
52
|
|
|
p_o &= \frac{|X \cap Y| + |(N \setminus X) \setminus Y|}{|N|} |
|
53
|
|
|
|
|
54
|
|
|
p_e^\pi &= \Big(\frac{|X| + |Y|}{2 \cdot |N|}\Big)^2 + |
|
55
|
|
|
\Big(\frac{|N \setminus X| + |N \setminus Y|}{2 \cdot |N|}\Big)^2 |
|
56
|
|
|
\end{array} |
|
57
|
|
|
|
|
58
|
|
|
|
|
59
|
|
|
In :ref:`2x2 confusion table terms <confusion_table>`, where a+b+c+d=n, |
|
60
|
|
|
this is |
|
61
|
|
|
|
|
62
|
|
|
.. math:: |
|
63
|
|
|
|
|
64
|
|
|
\begin{array}{ll} |
|
65
|
|
|
p_o &= \frac{a+d}{n} |
|
66
|
|
|
|
|
67
|
|
|
p_e^\pi &= \Big(\frac{2a+b+c}{2n}\Big)^2 + |
|
68
|
|
|
\Big(\frac{2d+b+c}{2n}\Big)^2 |
|
69
|
|
|
\end{array} |
|
70
|
|
|
|
|
71
|
|
|
|
|
72
|
|
|
.. versionadded:: 0.4.0 |
|
73
|
|
|
""" |
|
74
|
|
|
|
|
75
|
1 |
|
def __init__( |
|
76
|
|
|
self, |
|
77
|
|
|
alphabet=None, |
|
78
|
|
|
tokenizer=None, |
|
79
|
|
|
intersection_type='crisp', |
|
80
|
|
|
**kwargs |
|
81
|
|
|
): |
|
82
|
|
|
"""Initialize ScottPi instance. |
|
83
|
|
|
|
|
84
|
|
|
Parameters |
|
85
|
|
|
---------- |
|
86
|
|
|
alphabet : Counter, collection, int, or None |
|
87
|
|
|
This represents the alphabet of possible tokens. |
|
88
|
|
|
See :ref:`alphabet <alphabet>` description in |
|
89
|
|
|
:py:class:`_TokenDistance` for details. |
|
90
|
|
|
tokenizer : _Tokenizer |
|
91
|
|
|
A tokenizer instance from the :py:mod:`abydos.tokenizer` package |
|
92
|
|
|
intersection_type : str |
|
93
|
|
|
Specifies the intersection type, and set type as a result: |
|
94
|
|
|
See :ref:`intersection_type <intersection_type>` description in |
|
95
|
|
|
:py:class:`_TokenDistance` for details. |
|
96
|
|
|
**kwargs |
|
97
|
|
|
Arbitrary keyword arguments |
|
98
|
|
|
|
|
99
|
|
|
Other Parameters |
|
100
|
|
|
---------------- |
|
101
|
|
|
qval : int |
|
102
|
|
|
The length of each q-gram. Using this parameter and tokenizer=None |
|
103
|
|
|
will cause the instance to use the QGram tokenizer with this |
|
104
|
|
|
q value. |
|
105
|
|
|
metric : _Distance |
|
106
|
|
|
A string distance measure class for use in the ``soft`` and |
|
107
|
|
|
``fuzzy`` variants. |
|
108
|
|
|
threshold : float |
|
109
|
|
|
A threshold value, similarities above which are counted as |
|
110
|
|
|
members of the intersection for the ``fuzzy`` variant. |
|
111
|
|
|
|
|
112
|
|
|
|
|
113
|
|
|
.. versionadded:: 0.4.0 |
|
114
|
|
|
|
|
115
|
|
|
""" |
|
116
|
1 |
|
super(ScottPi, self).__init__( |
|
117
|
|
|
alphabet=alphabet, |
|
118
|
|
|
tokenizer=tokenizer, |
|
119
|
|
|
intersection_type=intersection_type, |
|
120
|
|
|
**kwargs |
|
121
|
|
|
) |
|
122
|
|
|
|
|
123
|
1 |
|
def corr(self, src, tar): |
|
124
|
|
|
"""Return the Scott's Pi correlation of two strings. |
|
125
|
|
|
|
|
126
|
|
|
Parameters |
|
127
|
|
|
---------- |
|
128
|
|
|
src : str |
|
129
|
|
|
Source string (or QGrams/Counter objects) for comparison |
|
130
|
|
|
tar : str |
|
131
|
|
|
Target string (or QGrams/Counter objects) for comparison |
|
132
|
|
|
|
|
133
|
|
|
Returns |
|
134
|
|
|
------- |
|
135
|
|
|
float |
|
136
|
|
|
Scott's Pi correlation |
|
137
|
|
|
|
|
138
|
|
|
Examples |
|
139
|
|
|
-------- |
|
140
|
|
|
>>> cmp = ScottPi() |
|
141
|
|
|
>>> cmp.corr('cat', 'hat') |
|
142
|
|
|
0.49743589743589733 |
|
143
|
|
|
>>> cmp.corr('Niall', 'Neil') |
|
144
|
|
|
0.35914053833129245 |
|
145
|
|
|
>>> cmp.corr('aluminum', 'Catalan') |
|
146
|
|
|
0.10798833377524023 |
|
147
|
|
|
>>> cmp.corr('ATCG', 'TAGC') |
|
148
|
|
|
-0.006418485237489689 |
|
149
|
|
|
|
|
150
|
|
|
|
|
151
|
|
|
.. versionadded:: 0.4.0 |
|
152
|
|
|
|
|
153
|
|
|
""" |
|
154
|
1 |
|
if src == tar: |
|
155
|
1 |
|
return 1.0 |
|
156
|
|
|
|
|
157
|
1 |
|
self._tokenize(src, tar) |
|
158
|
|
|
|
|
159
|
1 |
|
a = self._intersection_card() |
|
160
|
1 |
|
b = self._src_only_card() |
|
161
|
1 |
|
c = self._tar_only_card() |
|
162
|
1 |
|
d = self._total_complement_card() |
|
163
|
1 |
|
n = a + b + c + d |
|
164
|
|
|
|
|
165
|
1 |
|
po = (a + d) / n |
|
166
|
1 |
|
pe = ((2 * a + b + c) / (2 * n)) ** 2 + ( |
|
167
|
|
|
(2 * d + b + c) / (2 * n) |
|
168
|
|
|
) ** 2 |
|
169
|
|
|
|
|
170
|
1 |
|
if po != pe: |
|
171
|
1 |
|
return (po - pe) / (1 - pe) |
|
172
|
|
|
return 0.0 |
|
173
|
|
|
|
|
174
|
1 |
|
def sim(self, src, tar): |
|
175
|
|
|
"""Return the Scott's Pi similarity of two strings. |
|
176
|
|
|
|
|
177
|
|
|
Parameters |
|
178
|
|
|
---------- |
|
179
|
|
|
src : str |
|
180
|
|
|
Source string (or QGrams/Counter objects) for comparison |
|
181
|
|
|
tar : str |
|
182
|
|
|
Target string (or QGrams/Counter objects) for comparison |
|
183
|
|
|
|
|
184
|
|
|
Returns |
|
185
|
|
|
------- |
|
186
|
|
|
float |
|
187
|
|
|
Scott's Pi similarity |
|
188
|
|
|
|
|
189
|
|
|
Examples |
|
190
|
|
|
-------- |
|
191
|
|
|
>>> cmp = ScottPi() |
|
192
|
|
|
>>> cmp.sim('cat', 'hat') |
|
193
|
|
|
0.7487179487179487 |
|
194
|
|
|
>>> cmp.sim('Niall', 'Neil') |
|
195
|
|
|
0.6795702691656462 |
|
196
|
|
|
>>> cmp.sim('aluminum', 'Catalan') |
|
197
|
|
|
0.5539941668876202 |
|
198
|
|
|
>>> cmp.sim('ATCG', 'TAGC') |
|
199
|
|
|
0.49679075738125517 |
|
200
|
|
|
|
|
201
|
|
|
|
|
202
|
|
|
.. versionadded:: 0.4.0 |
|
203
|
|
|
|
|
204
|
|
|
""" |
|
205
|
1 |
|
return (1.0 + self.corr(src, tar)) / 2.0 |
|
206
|
|
|
|
|
207
|
|
|
|
|
208
|
|
|
if __name__ == '__main__': |
|
209
|
|
|
import doctest |
|
210
|
|
|
|
|
211
|
|
|
doctest.testmod() |
|
212
|
|
|
|
chrislit /
abydos
