| line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
|
1
|
|
|
|
|
|
|
=head1 NAME |
|
2
|
|
|
|
|
|
|
|
|
3
|
|
|
|
|
|
|
Text::NSP::Measures::2D::Fisher::right - Perl module implementation of the right sided |
|
4
|
|
|
|
|
|
|
Fisher's exact test. |
|
5
|
|
|
|
|
|
|
|
|
6
|
|
|
|
|
|
|
=head1 SYNOPSIS |
|
7
|
|
|
|
|
|
|
|
|
8
|
|
|
|
|
|
|
=head3 Basic Usage |
|
9
|
|
|
|
|
|
|
|
|
10
|
|
|
|
|
|
|
use Text::NSP::Measures::2D::Fisher::right; |
|
11
|
|
|
|
|
|
|
|
|
12
|
|
|
|
|
|
|
my $npp = 60; my $n1p = 20; my $np1 = 20; my $n11 = 10; |
|
13
|
|
|
|
|
|
|
|
|
14
|
|
|
|
|
|
|
$right_value = calculateStatistic( n11=>$n11, |
|
15
|
|
|
|
|
|
|
n1p=>$n1p, |
|
16
|
|
|
|
|
|
|
np1=>$np1, |
|
17
|
|
|
|
|
|
|
npp=>$npp); |
|
18
|
|
|
|
|
|
|
|
|
19
|
|
|
|
|
|
|
if( ($errorCode = getErrorCode())) |
|
20
|
|
|
|
|
|
|
{ |
|
21
|
|
|
|
|
|
|
print STDERR $errorCode." - ".getErrorMessage(); |
|
22
|
|
|
|
|
|
|
} |
|
23
|
|
|
|
|
|
|
else |
|
24
|
|
|
|
|
|
|
{ |
|
25
|
|
|
|
|
|
|
print getStatisticName."value for bigram is ".$right_value; |
|
26
|
|
|
|
|
|
|
} |
|
27
|
|
|
|
|
|
|
|
|
28
|
|
|
|
|
|
|
|
|
29
|
|
|
|
|
|
|
=head1 DESCRIPTION |
|
30
|
|
|
|
|
|
|
|
|
31
|
|
|
|
|
|
|
Assume that the frequency count data associated with a bigram |
|
32
|
|
|
|
|
|
|
is stored in a 2x2 contingency table: |
|
33
|
|
|
|
|
|
|
|
|
34
|
|
|
|
|
|
|
word2 ~word2 |
|
35
|
|
|
|
|
|
|
word1 n11 n12 | n1p |
|
36
|
|
|
|
|
|
|
~word1 n21 n22 | n2p |
|
37
|
|
|
|
|
|
|
-------------- |
|
38
|
|
|
|
|
|
|
np1 np2 npp |
|
39
|
|
|
|
|
|
|
|
|
40
|
|
|
|
|
|
|
where n11 is the number of times occur together, and |
|
41
|
|
|
|
|
|
|
n12 is the number of times occurs with some word other than |
|
42
|
|
|
|
|
|
|
word2, and n1p is the number of times in total that word1 occurs as |
|
43
|
|
|
|
|
|
|
the first word in a bigram. |
|
44
|
|
|
|
|
|
|
|
|
45
|
|
|
|
|
|
|
The fishers exact tests are calculated by fixing the marginal totals |
|
46
|
|
|
|
|
|
|
and computing the hypergeometric probabilities for all the possible |
|
47
|
|
|
|
|
|
|
contingency tables, |
|
48
|
|
|
|
|
|
|
|
|
49
|
|
|
|
|
|
|
A right sided test is calculated by adding the probabilities of all |
|
50
|
|
|
|
|
|
|
the possible two by two contingency tables formed by fixing the |
|
51
|
|
|
|
|
|
|
marginal totals and changing the value of n11 to greater than or |
|
52
|
|
|
|
|
|
|
equal to the given value. A right sided Fisher's Exact Test tells us |
|
53
|
|
|
|
|
|
|
how likely it is to randomly sample a table where n11 is greater |
|
54
|
|
|
|
|
|
|
than observed. In other words, it tells us how likely it is to sample |
|
55
|
|
|
|
|
|
|
an observation where the two words are more dependent than currently |
|
56
|
|
|
|
|
|
|
observed. |
|
57
|
|
|
|
|
|
|
|
|
58
|
|
|
|
|
|
|
=head2 Methods |
|
59
|
|
|
|
|
|
|
|
|
60
|
|
|
|
|
|
|
=over |
|
61
|
|
|
|
|
|
|
|
|
62
|
|
|
|
|
|
|
=cut |
|
63
|
|
|
|
|
|
|
|
|
64
|
|
|
|
|
|
|
package Text::NSP::Measures::2D::Fisher::right; |
|
65
|
|
|
|
|
|
|
|
|
66
|
|
|
|
|
|
|
|
|
67
|
1
|
|
|
1
|
|
1387
|
use Text::NSP::Measures::2D::Fisher; |
|
|
1
|
|
|
|
|
2
|
|
|
|
1
|
|
|
|
|
184
|
|
|
68
|
1
|
|
|
1
|
|
4
|
use strict; |
|
|
1
|
|
|
|
|
2
|
|
|
|
1
|
|
|
|
|
16
|
|
|
69
|
1
|
|
|
1
|
|
4
|
use Carp; |
|
|
1
|
|
|
|
|
1
|
|
|
|
1
|
|
|
|
|
42
|
|
|
70
|
1
|
|
|
1
|
|
4
|
use warnings; |
|
|
1
|
|
|
|
|
1
|
|
|
|
1
|
|
|
|
|
35
|
|
|
71
|
1
|
|
|
1
|
|
4
|
no warnings 'redefine'; |
|
|
1
|
|
|
|
|
2
|
|
|
|
1
|
|
|
|
|
419
|
|
|
72
|
|
|
|
|
|
|
require Exporter; |
|
73
|
|
|
|
|
|
|
|
|
74
|
|
|
|
|
|
|
our ($VERSION, @EXPORT, @ISA); |
|
75
|
|
|
|
|
|
|
|
|
76
|
|
|
|
|
|
|
@ISA = qw(Exporter); |
|
77
|
|
|
|
|
|
|
|
|
78
|
|
|
|
|
|
|
@EXPORT = qw(initializeStatistic calculateStatistic |
|
79
|
|
|
|
|
|
|
getErrorCode getErrorMessage getStatisticName); |
|
80
|
|
|
|
|
|
|
|
|
81
|
|
|
|
|
|
|
$VERSION = '0.97'; |
|
82
|
|
|
|
|
|
|
|
|
83
|
|
|
|
|
|
|
|
|
84
|
|
|
|
|
|
|
=item calculateStatistic() - This method calculates the right Fisher value |
|
85
|
|
|
|
|
|
|
|
|
86
|
|
|
|
|
|
|
INPUT PARAMS : $count_values .. Reference of an hash containing |
|
87
|
|
|
|
|
|
|
the count values computed by the |
|
88
|
|
|
|
|
|
|
count.pl program. |
|
89
|
|
|
|
|
|
|
|
|
90
|
|
|
|
|
|
|
RETURN VALUES : $right .. Right Fisher value. |
|
91
|
|
|
|
|
|
|
|
|
92
|
|
|
|
|
|
|
=cut |
|
93
|
|
|
|
|
|
|
|
|
94
|
|
|
|
|
|
|
sub calculateStatistic |
|
95
|
|
|
|
|
|
|
{ |
|
96
|
12
|
|
|
12
|
|
167
|
my %values = @_; |
|
97
|
|
|
|
|
|
|
|
|
98
|
12
|
|
|
|
|
12
|
my $probabilities; |
|
99
|
12
|
|
|
|
|
14
|
my $left_flag = 0; |
|
100
|
|
|
|
|
|
|
|
|
101
|
|
|
|
|
|
|
# computes and returns the observed and marginal values from |
|
102
|
|
|
|
|
|
|
# the frequency combination values. returns 0 if there is an |
|
103
|
|
|
|
|
|
|
# error in the computation or the values are inconsistent. |
|
104
|
12
|
100
|
|
|
|
29
|
if( !(Text::NSP::Measures::2D::Fisher::getValues(\%values)) ) |
|
105
|
|
|
|
|
|
|
{ |
|
106
|
10
|
|
|
|
|
21
|
return; |
|
107
|
|
|
|
|
|
|
} |
|
108
|
|
|
|
|
|
|
|
|
109
|
2
|
50
|
|
|
|
5
|
my $final_limit = ($n1p < $np1) ? $n1p : $np1; |
|
110
|
2
|
|
|
|
|
3
|
my $n11_org = $n11; |
|
111
|
|
|
|
|
|
|
|
|
112
|
2
|
|
|
|
|
4
|
my $n11_start = $n1p + $np1 - $npp; |
|
113
|
2
|
100
|
|
|
|
5
|
if($n11_start < $n11) |
|
114
|
|
|
|
|
|
|
{ |
|
115
|
1
|
|
|
|
|
2
|
$n11_start = $n11; |
|
116
|
|
|
|
|
|
|
} |
|
117
|
|
|
|
|
|
|
|
|
118
|
|
|
|
|
|
|
|
|
119
|
|
|
|
|
|
|
# to make the computations faster, we check which would require less computations |
|
120
|
|
|
|
|
|
|
# computing the leftfisher value and subtracting it from 1 or directly computing |
|
121
|
|
|
|
|
|
|
# the right fisher value. We do this since, generally for bigrams n11 is quite small |
|
122
|
|
|
|
|
|
|
# so its much faster to compute the left Fisher value. |
|
123
|
2
|
|
|
|
|
3
|
my $left_final_limit = $n11-1; |
|
124
|
2
|
|
|
|
|
4
|
my $left_n11 = $n1p + $np1 - $npp; |
|
125
|
2
|
100
|
|
|
|
8
|
if($left_n11<0) |
|
126
|
|
|
|
|
|
|
{ |
|
127
|
1
|
|
|
|
|
2
|
$left_n11 = 0; |
|
128
|
|
|
|
|
|
|
} |
|
129
|
|
|
|
|
|
|
|
|
130
|
|
|
|
|
|
|
# if computing the left fisher values first will take lesser amount of time them |
|
131
|
|
|
|
|
|
|
# we set a flag for later reference and then compute the leftfisher score for |
|
132
|
|
|
|
|
|
|
# n11-1 and then subtract the total score from one to get the right fisher value. |
|
133
|
2
|
50
|
|
|
|
6
|
if(($left_final_limit - $left_n11) < ($final_limit - $n11_start)) |
|
134
|
|
|
|
|
|
|
{ |
|
135
|
2
|
|
|
|
|
2
|
$left_flag = 1; |
|
136
|
2
|
50
|
|
|
|
8
|
if( !($probabilities = Text::NSP::Measures::2D::Fisher::computeDistribution($left_n11, $left_final_limit))) |
|
137
|
|
|
|
|
|
|
{ |
|
138
|
0
|
|
|
|
|
0
|
return; |
|
139
|
|
|
|
|
|
|
} |
|
140
|
|
|
|
|
|
|
} |
|
141
|
|
|
|
|
|
|
|
|
142
|
|
|
|
|
|
|
#else we compute the value normally and simply sum to get the rightfisher value. |
|
143
|
|
|
|
|
|
|
else |
|
144
|
|
|
|
|
|
|
{ |
|
145
|
0
|
0
|
|
|
|
0
|
if( !($probabilities = Text::NSP::Measures::2D::Fisher::computeDistribution($n11_start, $final_limit))) |
|
146
|
|
|
|
|
|
|
{ |
|
147
|
0
|
|
|
|
|
0
|
return; |
|
148
|
|
|
|
|
|
|
} |
|
149
|
|
|
|
|
|
|
} |
|
150
|
|
|
|
|
|
|
|
|
151
|
2
|
|
|
|
|
2
|
my $key_n11; |
|
152
|
|
|
|
|
|
|
|
|
153
|
2
|
|
|
|
|
3
|
my $rightfisher=0; |
|
154
|
|
|
|
|
|
|
|
|
155
|
2
|
|
|
|
|
7
|
foreach $key_n11 (sort { $b <=> $a } keys %$probabilities) |
|
|
24
|
|
|
|
|
24
|
|
|
156
|
|
|
|
|
|
|
{ |
|
157
|
11
|
50
|
|
|
|
18
|
if($left_flag) |
|
158
|
|
|
|
|
|
|
{ |
|
159
|
11
|
100
|
|
|
|
19
|
if($key_n11 >= $n11_org) |
|
160
|
|
|
|
|
|
|
{ |
|
161
|
1
|
|
|
|
|
2
|
last; |
|
162
|
|
|
|
|
|
|
} |
|
163
|
|
|
|
|
|
|
} |
|
164
|
|
|
|
|
|
|
else |
|
165
|
|
|
|
|
|
|
{ |
|
166
|
0
|
0
|
|
|
|
0
|
if($key_n11 < $n11_org) |
|
167
|
|
|
|
|
|
|
{ |
|
168
|
0
|
|
|
|
|
0
|
last; |
|
169
|
|
|
|
|
|
|
} |
|
170
|
|
|
|
|
|
|
} |
|
171
|
10
|
|
|
|
|
32
|
$rightfisher += exp($probabilities->{$key_n11}); |
|
172
|
|
|
|
|
|
|
} |
|
173
|
|
|
|
|
|
|
|
|
174
|
|
|
|
|
|
|
# if we computed the leftfisher value to get the right fisher value, we subtract |
|
175
|
|
|
|
|
|
|
# the sum of the probabilities for the tables from one to get the right fisher score. |
|
176
|
2
|
50
|
|
|
|
11
|
if($left_flag) |
|
177
|
|
|
|
|
|
|
{ |
|
178
|
2
|
50
|
|
|
|
4
|
if ($rightfisher > 1) |
|
179
|
|
|
|
|
|
|
{ |
|
180
|
0
|
|
|
|
|
0
|
$rightfisher = 0; |
|
181
|
|
|
|
|
|
|
} |
|
182
|
|
|
|
|
|
|
else |
|
183
|
|
|
|
|
|
|
{ |
|
184
|
2
|
|
|
|
|
3
|
$rightfisher = 1 - $rightfisher; |
|
185
|
|
|
|
|
|
|
} |
|
186
|
|
|
|
|
|
|
} |
|
187
|
|
|
|
|
|
|
|
|
188
|
2
|
|
|
|
|
12
|
return $rightfisher; |
|
189
|
|
|
|
|
|
|
} |
|
190
|
|
|
|
|
|
|
|
|
191
|
|
|
|
|
|
|
|
|
192
|
|
|
|
|
|
|
=item getStatisticName() - Returns the name of this statistic |
|
193
|
|
|
|
|
|
|
|
|
194
|
|
|
|
|
|
|
INPUT PARAMS : none |
|
195
|
|
|
|
|
|
|
|
|
196
|
|
|
|
|
|
|
RETURN VALUES : $name .. Name of the measure. |
|
197
|
|
|
|
|
|
|
|
|
198
|
|
|
|
|
|
|
=cut |
|
199
|
|
|
|
|
|
|
|
|
200
|
|
|
|
|
|
|
sub getStatisticName |
|
201
|
|
|
|
|
|
|
{ |
|
202
|
0
|
|
|
0
|
|
|
return "Right Fisher"; |
|
203
|
|
|
|
|
|
|
} |
|
204
|
|
|
|
|
|
|
|
|
205
|
|
|
|
|
|
|
|
|
206
|
|
|
|
|
|
|
|
|
207
|
|
|
|
|
|
|
1; |
|
208
|
|
|
|
|
|
|
__END__ |