| line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
|
1
|
|
|
|
|
|
|
# WordNet::stem.pm version 2.04 |
|
2
|
|
|
|
|
|
|
# (Last updated $Id: stem.pm,v 1.1 2008/03/27 05:13:01 sidz1979 Exp $) |
|
3
|
|
|
|
|
|
|
# |
|
4
|
|
|
|
|
|
|
# Package used by WordNet::Similarity::lesk module that |
|
5
|
|
|
|
|
|
|
# computes semantic relatedness of word senses in WordNet |
|
6
|
|
|
|
|
|
|
# using gloss overlaps. |
|
7
|
|
|
|
|
|
|
# |
|
8
|
|
|
|
|
|
|
# Copyright (c) 2005, |
|
9
|
|
|
|
|
|
|
# |
|
10
|
|
|
|
|
|
|
# Ted Pedersen, University of Minnesota Duluth |
|
11
|
|
|
|
|
|
|
# tpederse at d.umn.edu |
|
12
|
|
|
|
|
|
|
# |
|
13
|
|
|
|
|
|
|
# Satanjeev Banerjee, Carnegie Mellon University, Pittsburgh |
|
14
|
|
|
|
|
|
|
# banerjee+ at cs.cmu.edu |
|
15
|
|
|
|
|
|
|
# |
|
16
|
|
|
|
|
|
|
# This program is free software; you can redistribute it and/or |
|
17
|
|
|
|
|
|
|
# modify it under the terms of the GNU General Public License |
|
18
|
|
|
|
|
|
|
# as published by the Free Software Foundation; either version 2 |
|
19
|
|
|
|
|
|
|
# of the License, or (at your option) any later version. |
|
20
|
|
|
|
|
|
|
# |
|
21
|
|
|
|
|
|
|
# This program is distributed in the hope that it will be useful, |
|
22
|
|
|
|
|
|
|
# but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
23
|
|
|
|
|
|
|
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
24
|
|
|
|
|
|
|
# GNU General Public License for more details. |
|
25
|
|
|
|
|
|
|
# |
|
26
|
|
|
|
|
|
|
# You should have received a copy of the GNU General Public License |
|
27
|
|
|
|
|
|
|
# along with this program; if not, write to |
|
28
|
|
|
|
|
|
|
# |
|
29
|
|
|
|
|
|
|
# The Free Software Foundation, Inc., |
|
30
|
|
|
|
|
|
|
# 59 Temple Place - Suite 330, |
|
31
|
|
|
|
|
|
|
# Boston, MA 02111-1307, USA. |
|
32
|
|
|
|
|
|
|
# |
|
33
|
|
|
|
|
|
|
# ------------------------------------------------------------------ |
|
34
|
|
|
|
|
|
|
|
|
35
|
|
|
|
|
|
|
package WordNet::stem; |
|
36
|
|
|
|
|
|
|
|
|
37
|
|
|
|
|
|
|
=head1 NAME |
|
38
|
|
|
|
|
|
|
|
|
39
|
|
|
|
|
|
|
WordNet::stem - Module that find the stem of a word or the stems of a |
|
40
|
|
|
|
|
|
|
string of words, using WordNet. |
|
41
|
|
|
|
|
|
|
|
|
42
|
|
|
|
|
|
|
=head1 SYNOPSIS |
|
43
|
|
|
|
|
|
|
|
|
44
|
|
|
|
|
|
|
use WordNet::stem; |
|
45
|
|
|
|
|
|
|
|
|
46
|
|
|
|
|
|
|
my $wn = WordNet::QueryData->new(); |
|
47
|
|
|
|
|
|
|
|
|
48
|
|
|
|
|
|
|
my $stemmer = WordNet::stem->new($wn) |
|
49
|
|
|
|
|
|
|
|
|
50
|
|
|
|
|
|
|
my @stems = $stemmer->stemWord($word); |
|
51
|
|
|
|
|
|
|
|
|
52
|
|
|
|
|
|
|
my $string = $stemmer->stemString($inString, $cache); |
|
53
|
|
|
|
|
|
|
|
|
54
|
|
|
|
|
|
|
=head1 DESCRIPTION |
|
55
|
|
|
|
|
|
|
|
|
56
|
|
|
|
|
|
|
This module uses the internal stemming algorithm of WordNet to |
|
57
|
|
|
|
|
|
|
stem words and strings of words. This module is used by the |
|
58
|
|
|
|
|
|
|
lesk measure of the WordNet::Similarity package. |
|
59
|
|
|
|
|
|
|
|
|
60
|
|
|
|
|
|
|
=head2 Methods |
|
61
|
|
|
|
|
|
|
|
|
62
|
|
|
|
|
|
|
=over |
|
63
|
|
|
|
|
|
|
|
|
64
|
|
|
|
|
|
|
=cut |
|
65
|
|
|
|
|
|
|
|
|
66
|
1
|
|
|
1
|
|
4
|
use strict; |
|
|
1
|
|
|
|
|
2
|
|
|
|
1
|
|
|
|
|
22
|
|
|
67
|
1
|
|
|
1
|
|
4
|
use Exporter; |
|
|
1
|
|
|
|
|
3
|
|
|
|
1
|
|
|
|
|
35
|
|
|
68
|
1
|
|
|
1
|
|
4
|
use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS); |
|
|
1
|
|
|
|
|
1
|
|
|
|
1
|
|
|
|
|
545
|
|
|
69
|
|
|
|
|
|
|
|
|
70
|
|
|
|
|
|
|
@ISA = qw(Exporter); |
|
71
|
|
|
|
|
|
|
|
|
72
|
|
|
|
|
|
|
%EXPORT_TAGS = (); |
|
73
|
|
|
|
|
|
|
|
|
74
|
|
|
|
|
|
|
@EXPORT_OK = (); |
|
75
|
|
|
|
|
|
|
|
|
76
|
|
|
|
|
|
|
@EXPORT = (); |
|
77
|
|
|
|
|
|
|
|
|
78
|
|
|
|
|
|
|
$VERSION = '2.04'; |
|
79
|
|
|
|
|
|
|
|
|
80
|
|
|
|
|
|
|
=item new |
|
81
|
|
|
|
|
|
|
|
|
82
|
|
|
|
|
|
|
Creates a new stemmer object and initilizes it with a |
|
83
|
|
|
|
|
|
|
WordNet::QueryData object. |
|
84
|
|
|
|
|
|
|
|
|
85
|
|
|
|
|
|
|
Parameters: $wn |
|
86
|
|
|
|
|
|
|
|
|
87
|
|
|
|
|
|
|
Returns: $stemmer |
|
88
|
|
|
|
|
|
|
|
|
89
|
|
|
|
|
|
|
=cut |
|
90
|
|
|
|
|
|
|
|
|
91
|
|
|
|
|
|
|
# function to create the stemmer object |
|
92
|
|
|
|
|
|
|
sub new |
|
93
|
|
|
|
|
|
|
{ |
|
94
|
0
|
|
|
0
|
1
|
|
my $className = shift; |
|
95
|
0
|
|
|
|
|
|
my $wn = shift; |
|
96
|
0
|
|
|
|
|
|
my $self = {}; |
|
97
|
|
|
|
|
|
|
|
|
98
|
0
|
|
|
|
|
|
$self->{wn} = $wn; |
|
99
|
0
|
|
|
|
|
|
$self->{wordStemHash} = (); |
|
100
|
0
|
|
|
|
|
|
$self->{stringStemHash} = (); |
|
101
|
0
|
|
|
|
|
|
bless($self, $className); |
|
102
|
|
|
|
|
|
|
|
|
103
|
0
|
|
|
|
|
|
return $self; |
|
104
|
|
|
|
|
|
|
} |
|
105
|
|
|
|
|
|
|
|
|
106
|
|
|
|
|
|
|
=item stemString |
|
107
|
|
|
|
|
|
|
|
|
108
|
|
|
|
|
|
|
Takes a string of words as input and returns a string of stemmed words. |
|
109
|
|
|
|
|
|
|
|
|
110
|
|
|
|
|
|
|
Parameters: $inString |
|
111
|
|
|
|
|
|
|
|
|
112
|
|
|
|
|
|
|
Returns: $retString |
|
113
|
|
|
|
|
|
|
|
|
114
|
|
|
|
|
|
|
=cut |
|
115
|
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
# Function to take a string, and process it in such a way that all the |
|
117
|
|
|
|
|
|
|
# words in it get stemmed. Note that if a single word has two or more |
|
118
|
|
|
|
|
|
|
# possible stems, we return the original surface form since there is |
|
119
|
|
|
|
|
|
|
# no way to select from the competing stems. The stem of the string |
|
120
|
|
|
|
|
|
|
# can be cached if requested. Useful if the calling function knows |
|
121
|
|
|
|
|
|
|
# which strings it will have to stem over and over again. Strings that |
|
122
|
|
|
|
|
|
|
# will be only stemmed ones need not be cached - thereby saving space. |
|
123
|
|
|
|
|
|
|
sub stemString |
|
124
|
|
|
|
|
|
|
{ |
|
125
|
0
|
|
|
0
|
1
|
|
my $self = shift; |
|
126
|
0
|
|
|
|
|
|
my $inputString = shift; |
|
127
|
0
|
|
|
|
|
|
my $cache = shift; |
|
128
|
|
|
|
|
|
|
|
|
129
|
|
|
|
|
|
|
# whether or not this string has been requested for cacheing, |
|
130
|
|
|
|
|
|
|
# check in the cache |
|
131
|
0
|
0
|
|
|
|
|
return $self->{'stringStemHash'}->{$inputString} if (defined $self->{'stringStemHash'}->{$inputString}); |
|
132
|
|
|
|
|
|
|
|
|
133
|
|
|
|
|
|
|
# Not in cache. Stem. |
|
134
|
|
|
|
|
|
|
|
|
135
|
|
|
|
|
|
|
# for each word in the input get the stem and put in the output string |
|
136
|
0
|
|
|
|
|
|
my $outputString = ""; |
|
137
|
0
|
|
|
|
|
|
while ($inputString =~ /(\w+)/g) |
|
138
|
|
|
|
|
|
|
{ |
|
139
|
0
|
|
|
|
|
|
my $word = $1; |
|
140
|
0
|
|
|
|
|
|
my @stems = $self->stemWord($word); |
|
141
|
|
|
|
|
|
|
|
|
142
|
|
|
|
|
|
|
# if multiple or no stems, use surface form. |
|
143
|
0
|
0
|
|
|
|
|
$outputString .= ($#stems != 0) ? "$word " : "$stems[0] "; |
|
144
|
|
|
|
|
|
|
} |
|
145
|
|
|
|
|
|
|
|
|
146
|
|
|
|
|
|
|
# if cache required, do so |
|
147
|
0
|
0
|
|
|
|
|
$self->{'stringStemHash'}->{$inputString} = $outputString if (defined($cache)); |
|
148
|
|
|
|
|
|
|
|
|
149
|
|
|
|
|
|
|
# return the string |
|
150
|
0
|
|
|
|
|
|
return($outputString); |
|
151
|
|
|
|
|
|
|
} |
|
152
|
|
|
|
|
|
|
|
|
153
|
|
|
|
|
|
|
=item stemWord |
|
154
|
|
|
|
|
|
|
|
|
155
|
|
|
|
|
|
|
Takes a word as input and returns its stems. A word may have more than |
|
156
|
|
|
|
|
|
|
one stem. All are returned. |
|
157
|
|
|
|
|
|
|
|
|
158
|
|
|
|
|
|
|
Parameters: $word |
|
159
|
|
|
|
|
|
|
|
|
160
|
|
|
|
|
|
|
Returns: @stems |
|
161
|
|
|
|
|
|
|
|
|
162
|
|
|
|
|
|
|
=back |
|
163
|
|
|
|
|
|
|
|
|
164
|
|
|
|
|
|
|
=cut |
|
165
|
|
|
|
|
|
|
|
|
166
|
|
|
|
|
|
|
# stem the word passed to this function and return an array of words |
|
167
|
|
|
|
|
|
|
# that contain all the possible stems of this word. All possible stems |
|
168
|
|
|
|
|
|
|
# of the word may include the surface form too if its a valid WordNet |
|
169
|
|
|
|
|
|
|
# lemma. |
|
170
|
|
|
|
|
|
|
sub stemWord |
|
171
|
|
|
|
|
|
|
{ |
|
172
|
0
|
|
|
0
|
1
|
|
my $self = shift; |
|
173
|
0
|
|
|
|
|
|
my $word = shift; |
|
174
|
0
|
|
|
|
|
|
my $wn = $self->{wn}; |
|
175
|
0
|
|
|
|
|
|
my @stems = (); |
|
176
|
|
|
|
|
|
|
|
|
177
|
|
|
|
|
|
|
# if not in the cache, create and put in cache |
|
178
|
0
|
0
|
|
|
|
|
if (!defined $self->{wordStemHash}->{$word}) |
|
179
|
|
|
|
|
|
|
{ |
|
180
|
|
|
|
|
|
|
# So not in the hash. gotta check for all possible parts of speech. |
|
181
|
0
|
|
|
|
|
|
my %stems = (); |
|
182
|
0
|
|
|
|
|
|
my $possiblePartsOfSpeech = "nvar"; |
|
183
|
|
|
|
|
|
|
|
|
184
|
0
|
|
|
|
|
|
my $pos; |
|
185
|
0
|
|
|
|
|
|
while ("nvar" =~ /(.)/g) |
|
186
|
|
|
|
|
|
|
{ |
|
187
|
0
|
|
|
|
|
|
foreach ($wn->validForms("$word\#$1")) |
|
188
|
|
|
|
|
|
|
{ |
|
189
|
|
|
|
|
|
|
# put underscore for space |
|
190
|
0
|
|
|
|
|
|
$_ =~ s/ /_/g; |
|
191
|
|
|
|
|
|
|
|
|
192
|
|
|
|
|
|
|
# remove part of speech if any |
|
193
|
0
|
|
|
|
|
|
$_ =~ s/\#\w$//; |
|
194
|
|
|
|
|
|
|
|
|
195
|
|
|
|
|
|
|
# put in stems hash (the hash allows us to not worry about |
|
196
|
|
|
|
|
|
|
# multiple copies of the same stem!) |
|
197
|
0
|
|
|
|
|
|
$stems{$_} = 1; |
|
198
|
|
|
|
|
|
|
} |
|
199
|
|
|
|
|
|
|
} |
|
200
|
|
|
|
|
|
|
|
|
201
|
|
|
|
|
|
|
# put in the cache |
|
202
|
0
|
|
|
|
|
|
$self->{wordStemHash}->{$word} = join(" ", (keys %stems)); |
|
203
|
|
|
|
|
|
|
} |
|
204
|
|
|
|
|
|
|
|
|
205
|
|
|
|
|
|
|
# return the stems |
|
206
|
0
|
|
|
|
|
|
return (split / /, $self->{wordStemHash}->{$word}); |
|
207
|
|
|
|
|
|
|
} |
|
208
|
|
|
|
|
|
|
|
|
209
|
|
|
|
|
|
|
1; |
|
210
|
|
|
|
|
|
|
|
|
211
|
|
|
|
|
|
|
__END__ |