NAME
Text::NSP::Measures::2D::MI::tmi - Perl implementation for True Mutual Information.
SYNOPSIS
Basic Usage
use Text::NSP::Measures::2D::MI::tmi;
my $tmi = Text::NSP::Measures::2D::MI::tmi->new();
my $npp = 60; my $n1p = 20; my $np1 = 20; my $n11 = 10;
$tmi_value = $tmi->calculateStatistic( n11=>$n11,
n1p=>$n1p,
np1=>$np1,
npp=>$npp);
if( ($errorCode = $tmi->getErrorCode()))
{
print STDERR $erroCode." - ".$tmi->getErrorMessage();
}
else
{
print $tmi->getStatisticName."value for bigram is ".$tmi_value;
}
DESCRIPTION
Assume that the frequency count data associated with a bigram <word1><word2> is stored in a 2x2 contingency table:
word2 ~word2
word1 n11 n12 | n1p
~word1 n21 n22 | n2p
--------------
np1 np2 npp
where n11 is the number of times <word1><word2> occur together, and n12 is the number of times <word1> occurs with some word other than word2, and n1p is the number of times in total that word1 occurs as the first word in a bigram.
The expected values for the internal cells are calculated by taking the product of their associated marginals and dividing by the sample size, for example:
np1 * n1p
m11= ---------
npp
True Mutual Information (tmi) is defined as the weighted average of the pointwise mutual informations for all the observed and expected value pairs.
tmi = [n11/npp * log(n11/m11) + n12/npp * log(n12/m12) +
n21/npp * log(n21/m21) + n22/npp * log(n22/m22)]
PMI = log (n11/m11)
Methods
- calculateStatistic() - This method calculates the tmi value
-
INPUT PARAMS : $count_values .. Reference of an hash containing the count values computed by the count.pl program.
RETURN VALUES : $tmi .. TMI value for this bigram.
- getStatisticName() - Returns the name of this statistic
-
INPUT PARAMS : none
RETURN VALUES : $name .. Name of the measure.
AUTHOR
Ted Pedersen, University of Minnesota Duluth <tpederse@d.umn.edu>
Satanjeev Banerjee, Carnegie Mellon University <satanjeev@cmu.edu>
Amruta Purandare, University of Pittsburgh <amruta@cs.pitt.edu>
Bridget Thomson-McInnes, University of Minnesota Twin Cities <bthompson@d.umn.edu>
Saiyam Kohli, University of Minnesota Duluth <kohli003@d.umn.edu>
HISTORY
Last updated: $Id: tmi.pm,v 1.19 2006/06/15 16:53:04 saiyam_kohli Exp $
BUGS
SEE ALSO
@inproceedings{moore:2004:EMNLP,
author = {Moore, Robert C.},
title = {On Log-Likelihood-Ratios and the Significance of Rare
Events },
booktitle = {Proceedings of EMNLP 2004},
editor = {Dekang Lin and Dekai Wu},
year = 2004,
month = {July},
address = {Barcelona, Spain},
publisher = {Association for Computational Linguistics},
pages = {333--340}
url = L<http://acl.ldc.upenn.edu/acl2004/emnlp/pdf/Moore.pdf>}
http://groups.yahoo.com/group/ngram/
http://www.d.umn.edu/~tpederse/nsp.html
COPYRIGHT
Copyright (C) 2000-2006, Ted Pedersen, Satanjeev Banerjee, Amruta Purandare, Bridget Thomson-McInnes and Saiyam Kohli
This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License along with this program; if not, write to
The Free Software Foundation, Inc.,
59 Temple Place - Suite 330,
Boston, MA 02111-1307, USA.
Note: a copy of the GNU General Public License is available on the web at http://www.gnu.org/licenses/gpl.txt and is included in this distribution as GPL.txt.