Rechercher une page de manuel
Text::NSP::Measures::2D::MI.3pm
Langue: en
Version: 2006-06-21 (mandriva - 01/05/08)
Section: 3 (Bibliothèques de fonctions)
NAME
Text::NSP::Measures::2D::MI - Perl module that provides error checksfor Loglieklihood, Total Mutual
Information, Pointwise Mutual Information
and Poisson-Stirling Measure.
SYNOPSIS
Basic Usageuse Text::NSP::Measures::2D::MI::ll;
my $npp = 60; my $n1p = 20; my $np1 = 20; my $n11 = 10;
$ll_value = calculateStatistic( n11=>$n11, n1p=>$n1p, np1=>$np1, npp=>$npp);
if( ($errorCode = getErrorCode())) { print STDERR $errorCode." - ".getErrorMessage()."\n""; } else { print getStatisticName."value for bigram is ".$ll_value."\n""; }
DESCRIPTION
This module is the base class for the Loglikelihood, Total Mutual Information and the Pointwise Mutual Information measures. All these measure are similar. This module provides error checks specific for these measures, it also implements the computations that are common to these measures.- Log-Likelihood measure is computed as
- Log-Likelihood = 2 * [n11 * log(n11/m11) + n12 * log(n12/m12) +
n21 * log(n21/m21) + n22 * log(n22/m22)] - Total Mutual Information
- TMI = (1/npp)*[n11 * log(n11/m11)/log 2 + n12 * log(n12/m12)/log 2 +
n21 * log(n21/m21)/log 2 + n22 * log(n22/m22)/log 2] - Pointwise Mutual Information
- PMI = log (n11/m11)/log 2
- Poisson Stirling Measures
- PS = n11*(log (n11/m11)-1)
All these methods use the ratio of the observed values to expected values, for computations, and thus have common error checks, sothey have been grouped togrther.
Methods
- getValues() - This method calls the computeMarginalTotals(), computeObservedValues() and the computeExpectedValues() methods to compute the observed and expected values. It checks these values for any errors that might cause the Loglikelihood, TMI & PMI measures to fail.
- INPUT PARAMS : $count_values .. Reference of an hash containing
the count values computed by the
count.pl program.RETURN VALUES : 1/undef ..returns '1' to indicate success
and an undefined(NULL) value to indicate
faliure. - computePMI() - Computes the pmi of a given observed and expected value pair.
- INPUT PARAMS : $n ..Observed value
$m ..Expected valueRETURN VALUES : log(n/m) ..the log of the ratio of
observed value to expected
value.
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: MI.pm,v 1.26 2006/09/15 23:27:03 saiyam_kohli Exp $BUGS
SEE ALSO
<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 KohliThis 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.
Contenus ©2006-2024 Benjamin Poulain
Design ©2006-2024 Maxime Vantorre