Ted Pedersen > Text-NSP > Text::NSP::Measures::2D::MI::ll

Download:
Text-NSP-1.27.tar.gz

Dependencies

Annotate this POD

CPAN RT

Open  0
View/Report Bugs
Module Version: 0.97   Source  

NAME ^

Text::NSP::Measures::2D::MI::ll - Perl module that implements Loglikelihood measure of association for bigrams.

SYNOPSIS ^

Basic Usage

  use 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 $errorCode." - ".getErrorMessage();
  }
  else
  {
    print getStatisticName."value for bigram is ".$ll_value;
  }

DESCRIPTION ^

The log-likelihood ratio measures the deviation between the observed data and what would be expected if <word1> and <word2> were independent. The higher the score, the less evidence there is in favor of concluding that the words are independent.

Assume that the frequency count data associated with a bigram <word1><word2> as shown by 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

Then the deviation between observed and expected values for each internal cell is computed to arrive at the log-likelihood value.

 Log-Likelihood = 2 * [n11 * log(n11/m11) + n12 * log(n12/m12) +
           n21 * log(n21/m21) + n22 * log(n22/m22)]

Methods

calculateStatistic() - This method calculates the ll value

INPUT PARAMS : $count_values .. Reference of an hash containing the count values computed by the count.pl program.

RETURN VALUES : $loglikelihood .. Loglikelihood 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: ll.pm,v 1.23 2008/03/26 17:20:27 tpederse Exp $

BUGS ^

SEE ALSO ^

  @article{Dunning93,
            author = {Dunning, T.},
            title = {Accurate Methods for the Statistics of
                    Surprise and Coincidence},
            journal = {Computational Linguistics},
            volume = {19},
            number = {1},
            year = {1993},
            pages = {61-74}
            url = L<http://www.comp.lancs.ac.uk/ucrel/papers/tedstats.pdf>}

  @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.

syntax highlighting: