line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
# WordNet::Similarity::random.pm version 2.04 |
2
|
|
|
|
|
|
|
# (Last updated $Id: random.pm,v 1.15 2008/03/27 06:21:17 sidz1979 Exp $) |
3
|
|
|
|
|
|
|
# |
4
|
|
|
|
|
|
|
# Random semantic distance generator module. |
5
|
|
|
|
|
|
|
# |
6
|
|
|
|
|
|
|
# Copyright (c) 2005, |
7
|
|
|
|
|
|
|
# |
8
|
|
|
|
|
|
|
# Ted Pedersen, University of Minnesota Duluth |
9
|
|
|
|
|
|
|
# tpederse at d.umn.edu |
10
|
|
|
|
|
|
|
# |
11
|
|
|
|
|
|
|
# Siddharth Patwardhan, University of Utah, Salt Lake City |
12
|
|
|
|
|
|
|
# sidd at cs.utah.edu |
13
|
|
|
|
|
|
|
# |
14
|
|
|
|
|
|
|
# This program is free software; you can redistribute it and/or |
15
|
|
|
|
|
|
|
# modify it under the terms of the GNU General Public License |
16
|
|
|
|
|
|
|
# as published by the Free Software Foundation; either version 2 |
17
|
|
|
|
|
|
|
# of the License, or (at your option) any later version. |
18
|
|
|
|
|
|
|
# |
19
|
|
|
|
|
|
|
# This program is distributed in the hope that it will be useful, |
20
|
|
|
|
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
21
|
|
|
|
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
22
|
|
|
|
|
|
|
# GNU General Public License for more details. |
23
|
|
|
|
|
|
|
# |
24
|
|
|
|
|
|
|
# You should have received a copy of the GNU General Public License |
25
|
|
|
|
|
|
|
# along with this program; if not, write to |
26
|
|
|
|
|
|
|
# |
27
|
|
|
|
|
|
|
# The Free Software Foundation, Inc., |
28
|
|
|
|
|
|
|
# 59 Temple Place - Suite 330, |
29
|
|
|
|
|
|
|
# Boston, MA 02111-1307, USA. |
30
|
|
|
|
|
|
|
# |
31
|
|
|
|
|
|
|
# ------------------------------------------------------------------ |
32
|
|
|
|
|
|
|
|
33
|
|
|
|
|
|
|
package WordNet::Similarity::random; |
34
|
|
|
|
|
|
|
|
35
|
|
|
|
|
|
|
=head1 NAME |
36
|
|
|
|
|
|
|
|
37
|
|
|
|
|
|
|
WordNet::Similarity::random - Perl module for computing semantic relatedness |
38
|
|
|
|
|
|
|
of word senses using a random measure. |
39
|
|
|
|
|
|
|
|
40
|
|
|
|
|
|
|
=head1 SYNOPSIS |
41
|
|
|
|
|
|
|
|
42
|
|
|
|
|
|
|
use WordNet::Similarity::random; |
43
|
|
|
|
|
|
|
|
44
|
|
|
|
|
|
|
use WordNet::QueryData; |
45
|
|
|
|
|
|
|
|
46
|
|
|
|
|
|
|
my $wn = WordNet::QueryData->new(); |
47
|
|
|
|
|
|
|
|
48
|
|
|
|
|
|
|
my $random = WordNet::Similarity::random->new($wn); |
49
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
my $value = $random->getRelatedness("car#n#1", "bus#n#2"); |
51
|
|
|
|
|
|
|
|
52
|
|
|
|
|
|
|
($error, $errorString) = $random->getError(); |
53
|
|
|
|
|
|
|
|
54
|
|
|
|
|
|
|
die "$errorString\n" if($error); |
55
|
|
|
|
|
|
|
|
56
|
|
|
|
|
|
|
print "car (sense 1) <-> bus (sense 2) = $value\n"; |
57
|
|
|
|
|
|
|
|
58
|
|
|
|
|
|
|
=head1 DESCRIPTION |
59
|
|
|
|
|
|
|
|
60
|
|
|
|
|
|
|
This module generates random numbers as a measure of semantic relatedness |
61
|
|
|
|
|
|
|
of word senses. It is possible to assign a random value for a word sense |
62
|
|
|
|
|
|
|
pair and return the same value if the same word sense pair is passed as |
63
|
|
|
|
|
|
|
input. It is also possible to generate a new random value for the same |
64
|
|
|
|
|
|
|
word sense pair every time. |
65
|
|
|
|
|
|
|
|
66
|
|
|
|
|
|
|
=head2 Methods |
67
|
|
|
|
|
|
|
|
68
|
|
|
|
|
|
|
=over |
69
|
|
|
|
|
|
|
|
70
|
|
|
|
|
|
|
=cut |
71
|
|
|
|
|
|
|
|
72
|
1
|
|
|
1
|
|
4624
|
use strict; |
|
1
|
|
|
|
|
4
|
|
|
1
|
|
|
|
|
63
|
|
73
|
|
|
|
|
|
|
|
74
|
1
|
|
|
1
|
|
57
|
use WordNet::Similarity; |
|
0
|
|
|
|
|
|
|
|
0
|
|
|
|
|
|
|
75
|
|
|
|
|
|
|
|
76
|
|
|
|
|
|
|
use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS); |
77
|
|
|
|
|
|
|
|
78
|
|
|
|
|
|
|
@ISA = qw(WordNet::Similarity); |
79
|
|
|
|
|
|
|
|
80
|
|
|
|
|
|
|
%EXPORT_TAGS = (); |
81
|
|
|
|
|
|
|
|
82
|
|
|
|
|
|
|
@EXPORT_OK = (); |
83
|
|
|
|
|
|
|
|
84
|
|
|
|
|
|
|
@EXPORT = (); |
85
|
|
|
|
|
|
|
|
86
|
|
|
|
|
|
|
$VERSION = '2.04'; |
87
|
|
|
|
|
|
|
|
88
|
|
|
|
|
|
|
WordNet::Similarity::addConfigOption ('maxrand', 1, 'f', 1.0); |
89
|
|
|
|
|
|
|
|
90
|
|
|
|
|
|
|
=item $random->setPosList() |
91
|
|
|
|
|
|
|
|
92
|
|
|
|
|
|
|
This method is internally called to determine the parts of speech |
93
|
|
|
|
|
|
|
this measure is capable of dealing with. |
94
|
|
|
|
|
|
|
|
95
|
|
|
|
|
|
|
Parameters: none. |
96
|
|
|
|
|
|
|
|
97
|
|
|
|
|
|
|
Returns: none. |
98
|
|
|
|
|
|
|
|
99
|
|
|
|
|
|
|
=cut |
100
|
|
|
|
|
|
|
|
101
|
|
|
|
|
|
|
sub setPosList |
102
|
|
|
|
|
|
|
{ |
103
|
|
|
|
|
|
|
my $self = shift; |
104
|
|
|
|
|
|
|
$self->{n} = 1; |
105
|
|
|
|
|
|
|
$self->{v} = 1; |
106
|
|
|
|
|
|
|
$self->{a} = 1; |
107
|
|
|
|
|
|
|
$self->{r} = 1; |
108
|
|
|
|
|
|
|
} |
109
|
|
|
|
|
|
|
|
110
|
|
|
|
|
|
|
=item $random->initialize($file) |
111
|
|
|
|
|
|
|
|
112
|
|
|
|
|
|
|
Overrides the initialize method in the parent class. This method |
113
|
|
|
|
|
|
|
essentially initializes the measure for use. |
114
|
|
|
|
|
|
|
|
115
|
|
|
|
|
|
|
Parameters: $file -- configuration file. |
116
|
|
|
|
|
|
|
|
117
|
|
|
|
|
|
|
Returns: none. |
118
|
|
|
|
|
|
|
|
119
|
|
|
|
|
|
|
=cut |
120
|
|
|
|
|
|
|
|
121
|
|
|
|
|
|
|
# Initialization of the WordNet::Similarity::random object... parses the config file and sets up |
122
|
|
|
|
|
|
|
# global variables, or sets them to default values. |
123
|
|
|
|
|
|
|
# INPUT PARAMS : $paramFile .. File containing the module specific params. |
124
|
|
|
|
|
|
|
# RETURN VALUES : (none) |
125
|
|
|
|
|
|
|
sub initialize |
126
|
|
|
|
|
|
|
{ |
127
|
|
|
|
|
|
|
my $self = shift; |
128
|
|
|
|
|
|
|
$self->SUPER::initialize (@_); |
129
|
|
|
|
|
|
|
$self->{maxCacheSize} = WordNet::Similarity::UNLIMITED_CACHE; |
130
|
|
|
|
|
|
|
} |
131
|
|
|
|
|
|
|
|
132
|
|
|
|
|
|
|
=item $random->getRelatedness ($synset1, $synset2) |
133
|
|
|
|
|
|
|
|
134
|
|
|
|
|
|
|
Returns a value for the relatedness of the two synsets. This value is |
135
|
|
|
|
|
|
|
a random number greater-than or equal-to zero and less-than 'maxrand'. |
136
|
|
|
|
|
|
|
|
137
|
|
|
|
|
|
|
If the synsets are not properly formed word#pos#sense strings, or if they |
138
|
|
|
|
|
|
|
are not found in WordNet, then the error level will be set to non-zero and |
139
|
|
|
|
|
|
|
an error string will be generated. |
140
|
|
|
|
|
|
|
|
141
|
|
|
|
|
|
|
=cut |
142
|
|
|
|
|
|
|
|
143
|
|
|
|
|
|
|
sub getRelatedness |
144
|
|
|
|
|
|
|
{ |
145
|
|
|
|
|
|
|
my $self = shift; |
146
|
|
|
|
|
|
|
my $wps1 = shift; |
147
|
|
|
|
|
|
|
my $wps2 = shift; |
148
|
|
|
|
|
|
|
my $wn = $self->{wn}; |
149
|
|
|
|
|
|
|
my $class = ref $self || $self; |
150
|
|
|
|
|
|
|
|
151
|
|
|
|
|
|
|
# check if the synsets are well-formed and are found in WordNet |
152
|
|
|
|
|
|
|
my $ref = $self->parseWps ($wps1, $wps2); |
153
|
|
|
|
|
|
|
ref $ref or return $ref; |
154
|
|
|
|
|
|
|
|
155
|
|
|
|
|
|
|
# Initialize traces. |
156
|
|
|
|
|
|
|
$self->{traceString} = ""; |
157
|
|
|
|
|
|
|
|
158
|
|
|
|
|
|
|
# Now check if the similarity value for these two synsets is in |
159
|
|
|
|
|
|
|
# fact in the cache... if so return the cached value. |
160
|
|
|
|
|
|
|
my $relatedness = |
161
|
|
|
|
|
|
|
$self->{doCache} ? $self->fetchFromCache ($wps1, $wps2) : undef; |
162
|
|
|
|
|
|
|
defined $relatedness and return $relatedness; |
163
|
|
|
|
|
|
|
|
164
|
|
|
|
|
|
|
# Now get down to really finding the relatedness of these two. |
165
|
|
|
|
|
|
|
|
166
|
|
|
|
|
|
|
my $score = rand ($self->{maxrand}); |
167
|
|
|
|
|
|
|
$score = sprintf ("%.3f", $score); |
168
|
|
|
|
|
|
|
|
169
|
|
|
|
|
|
|
$self->{doCache} and $self->storeToCache ($wps1, $wps2, $score); |
170
|
|
|
|
|
|
|
return $score; |
171
|
|
|
|
|
|
|
} |
172
|
|
|
|
|
|
|
|
173
|
|
|
|
|
|
|
=item $random->traceOptions() |
174
|
|
|
|
|
|
|
|
175
|
|
|
|
|
|
|
This method is internally called to determine the extra options |
176
|
|
|
|
|
|
|
specified by this measure (apart from the default options specified |
177
|
|
|
|
|
|
|
in the WordNet::Similarity base class). |
178
|
|
|
|
|
|
|
|
179
|
|
|
|
|
|
|
Parameters: none. |
180
|
|
|
|
|
|
|
|
181
|
|
|
|
|
|
|
Returns: none. |
182
|
|
|
|
|
|
|
|
183
|
|
|
|
|
|
|
=cut |
184
|
|
|
|
|
|
|
|
185
|
|
|
|
|
|
|
# 12/5/03 JM (#1) |
186
|
|
|
|
|
|
|
# show all config options specific to this module |
187
|
|
|
|
|
|
|
sub traceOptions { |
188
|
|
|
|
|
|
|
my $self = shift; |
189
|
|
|
|
|
|
|
$self->{traceString} .= "maxrand :: $self->{maxrand}\n"; |
190
|
|
|
|
|
|
|
$self->SUPER::traceOptions(); |
191
|
|
|
|
|
|
|
} |
192
|
|
|
|
|
|
|
|
193
|
|
|
|
|
|
|
1; |
194
|
|
|
|
|
|
|
|
195
|
|
|
|
|
|
|
__END__ |