|
1
|
|
|
# -*- coding: utf-8 -*- |
|
2
|
|
|
|
|
3
|
|
|
# Copyright 2014-2018 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
|
|
|
"""abydos.clustering. |
|
20
|
|
|
|
|
21
|
|
|
The clustering module implements clustering algorithms such as: |
|
22
|
|
|
- mean pair-wise similarity |
|
23
|
|
|
""" |
|
24
|
|
|
|
|
25
|
|
|
from __future__ import division, unicode_literals |
|
26
|
|
|
|
|
27
|
|
|
from six.moves import range |
|
28
|
|
|
|
|
29
|
|
|
from .distance import sim |
|
30
|
|
|
from .stats import amean, hmean, std |
|
31
|
|
|
|
|
32
|
|
|
|
|
33
|
|
|
def mean_pairwise_similarity(collection, metric=sim, |
|
34
|
|
|
mean_func=hmean, symmetric=False): |
|
35
|
|
|
"""Calculate the mean pairwise similarity of a collection of strings. |
|
36
|
|
|
|
|
37
|
|
|
Takes the mean of the pairwise similarity between each member of a |
|
38
|
|
|
collection, optionally in both directions (for asymmetric similarity |
|
39
|
|
|
metrics. |
|
40
|
|
|
|
|
41
|
|
|
:param list collection: a collection of terms or a string that can be split |
|
42
|
|
|
:param function metric: a similarity metric function |
|
43
|
|
|
:param function mean_func: a mean function that takes a list of values and |
|
44
|
|
|
returns a float |
|
45
|
|
|
:param bool symmetric: set to True if all pairwise similarities should be |
|
46
|
|
|
calculated in both directions |
|
47
|
|
|
:returns: the mean pairwise similarity of a collection of strings |
|
48
|
|
|
:rtype: str |
|
49
|
|
|
|
|
50
|
|
|
>>> mean_pairwise_similarity(['Christopher', 'Kristof', 'Christobal']) |
|
51
|
|
|
0.51980198019801982 |
|
52
|
|
|
>>> mean_pairwise_similarity(['Niall', 'Neal', 'Neil']) |
|
53
|
|
|
0.54545454545454541 |
|
54
|
|
|
""" |
|
55
|
|
|
if not callable(mean_func): |
|
56
|
|
|
raise ValueError('mean_func must be a function') |
|
57
|
|
|
|
|
58
|
|
|
if hasattr(collection, 'split'): |
|
59
|
|
|
collection = collection.split() |
|
60
|
|
|
if not hasattr(collection, '__iter__'): |
|
61
|
|
|
raise ValueError('collection is neither a string nor iterable type') |
|
62
|
|
|
elif len(collection) < 2: |
|
63
|
|
|
raise ValueError('collection has fewer than two members') |
|
64
|
|
|
|
|
65
|
|
|
collection = list(collection) |
|
66
|
|
|
|
|
67
|
|
|
pairwise_values = [] |
|
68
|
|
|
|
|
69
|
|
|
for i in range(len(collection)): |
|
|
|
|
|
|
70
|
|
|
for j in range(i+1, len(collection)): |
|
71
|
|
|
pairwise_values.append(metric(collection[i], collection[j])) |
|
72
|
|
|
if symmetric: |
|
73
|
|
|
pairwise_values.append(metric(collection[j], collection[i])) |
|
74
|
|
|
|
|
75
|
|
|
return mean_func(pairwise_values) |
|
76
|
|
|
|
|
77
|
|
|
|
|
78
|
|
|
def pairwise_similarity_statistics(src_collection, tar_collection, metric=sim, |
|
79
|
|
|
mean_func=amean, symmetric=False): |
|
80
|
|
|
"""Calculate the mean pairwise similarity of a collection of strings. |
|
81
|
|
|
|
|
82
|
|
|
Takes the mean of the pairwise similarity between each member of a |
|
83
|
|
|
collection, optionally in both directions (for asymmetric similarity |
|
84
|
|
|
metrics. |
|
85
|
|
|
|
|
86
|
|
|
:param list src_collection: a collection of terms or a string that can be |
|
87
|
|
|
split |
|
88
|
|
|
:param list tar_collection: a collection of terms or a string that can be |
|
89
|
|
|
split |
|
90
|
|
|
:param function metric: a similarity metric function |
|
91
|
|
|
:param function mean_func: a mean function that takes a list of values and |
|
92
|
|
|
returns a float |
|
93
|
|
|
:param bool symmetric: set to True if all pairwise similarities should be |
|
94
|
|
|
calculated in both directions |
|
95
|
|
|
:returns: the max, min, mean, and standard deviation of similarities |
|
96
|
|
|
:rtype: str |
|
97
|
|
|
""" |
|
98
|
|
|
if not callable(mean_func): |
|
99
|
|
|
raise ValueError('mean_func must be a function') |
|
100
|
|
|
|
|
101
|
|
|
if hasattr(src_collection, 'split'): |
|
102
|
|
|
src_collection = src_collection.split() |
|
103
|
|
|
if not hasattr(src_collection, '__iter__'): |
|
104
|
|
|
raise ValueError('src_collection is neither a string nor iterable') |
|
105
|
|
|
|
|
106
|
|
|
if hasattr(tar_collection, 'split'): |
|
107
|
|
|
tar_collection = tar_collection.split() |
|
108
|
|
|
if not hasattr(tar_collection, '__iter__'): |
|
109
|
|
|
raise ValueError('tar_collection is neither a string nor iterable') |
|
110
|
|
|
|
|
111
|
|
|
src_collection = list(src_collection) |
|
112
|
|
|
tar_collection = list(tar_collection) |
|
113
|
|
|
|
|
114
|
|
|
pairwise_values = [] |
|
115
|
|
|
|
|
116
|
|
|
for src in src_collection: |
|
117
|
|
|
for tar in tar_collection: |
|
118
|
|
|
pairwise_values.append(metric(src, tar)) |
|
119
|
|
|
if symmetric: |
|
120
|
|
|
pairwise_values.append(metric(tar, src)) |
|
121
|
|
|
|
|
122
|
|
|
return (max(pairwise_values), min(pairwise_values), |
|
123
|
|
|
mean_func(pairwise_values), std(pairwise_values, mean_func, 0)) |
|
124
|
|
|
|
|
125
|
|
|
|
|
126
|
|
|
if __name__ == '__main__': |
|
127
|
|
|
import doctest |
|
128
|
|
|
doctest.testmod() |
|
129
|
|
|
|
chrislit /
abydos
Loading history...