NAME

AI::ConfusionMatrix - make a confusion matrix

SYNOPSIS

my %matrix;

Loop over your tests

---

$matrix{$expected}{$predicted} += 1;

---

makeConfusionMatrix(\%matrix, 'output.csv');

DESCRIPTION

This module prints a confusion matrix from a hash reference. This module tries to be generic enough to be used within a lot of machine learning projects.

Function

makeConfusionMatrix($hash_ref, $filename)

This function makes a confusion matrix from $hash_ref and writes it to $filename.

Example:

makeConfusionMatrix(\%matrix, 'output.csv');

The hash reference must look like this :

$VAR1 = {


          'value_expected1' => {
                      'value_predicted1' => value
                    },
          'value_expected2' => {
                      'value_predicted1' => value,
                      'value_predicted2' => value
                    },
          'value_expected3' => {
                      'value_predicted3' => value
                    }

        };

The output will be in CSV. Here is an example:

,1997,1998,2001,2003,2005,2008,2012,2015,TOTAL,TP,FP,FN,ACC
1997,1,,,,,,,1,2,1,0,1,50.00%
1998,,1,,,,,,,1,1,0,0,100.00%
2001,,,1,,,,,,1,1,0,0,100.00%
2003,,,,5,,,,2,7,5,0,2,71.43%
2005,,,,,5,,,4,9,5,0,4,55.56%
2008,,,,,,3,,,3,3,0,0,100.00%
2012,,,,,,,5,,5,5,0,0,100.00%
2015,,,,,,,,2,2,2,7,0,100.00%
TOTAL,1,1,1,5,5,3,5,9,30,23,7,7,84.62%

Prettified:

| | 1997 | 1998 | 2001 | 2003 | 2005 | 2008 | 2012 | 2015 | TOTAL | TP | FP | FN | ACC | |-------|------|------|------|------|------|------|------|------|-------|----|----|----|---------| | 1997 | 1 | | | | | | | 1 | 2 | 1 | 0 | 1 | 50.00% | | 1998 | | 1 | | | | | | | 1 | 1 | 0 | 0 | 100.00% | | 2001 | | | 1 | | | | | | 1 | 1 | 0 | 0 | 100.00% | | 2003 | | | | 5 | | | | 2 | 7 | 5 | 0 | 2 | 71.43% | | 2005 | | | | | 5 | | | 4 | 9 | 5 | 0 | 4 | 55.56% | | 2008 | | | | | | 3 | | | 3 | 3 | 0 | 0 | 100.00% | | 2012 | | | | | | | 5 | | 5 | 5 | 0 | 0 | 100.00% | | 2015 | | | | | | | | 2 | 2 | 2 | 7 | 0 | 100.00% | | TOTAL | 1 | 1 | 1 | 5 | 5 | 3 | 5 | 9 | 30 | 23 | 7 | 7 | 84.62% |

TP:

True Positive

FP:

False Positive

FN:

False Negative

ACC:

Accuracy

AUTHOR

Vincent Lequertier <sky@riseup.net>

LICENSE

This library is free software; you can redistribute it and/or modify it under the same terms as Perl itself.