NAME

Text::Phonetic::Phonix - Phonix algorithm

WARNING

Since I have only found ambiguous sources about this algorithm this implementation doesn't behave fully according to the few test cases I was able to find (currently three out of 17 test cases fail). I'm looking for more and better specifications of this algorithm. Until then this module should be regarded as EXPERIMENTAL.

DESCRIPTION

Phonix is a phonetic algorithm similar to Soundex.

The algorithm always returns an array reference with two elements. The fist element represents the sound of the name without the ending sound, and the second element represents the ending sound. To get a full representation of the name you need to concat the two elements.

If you want to compare two names the following rules apply:

• If the ending sound values of an entered name and a retrieved name are the same, the retrieved name is a LIKELY candidate.

• If an entered name has an ending-sound value, and the retrieved name does not, then the retrieved name is a LEAST-LIKELY candidate.

• If the two ending-sound values are the same for the length of the shorter, and the difference in length between the two ending-sound is one digit only, then the retrieved name isa LESS-LIKELY candidate.

• All other cases result in LEAST-LIKELY candidates.

AUTHOR

Maroš Kollár
CPAN ID: MAROS
maros [at] k-1.com
http://www.k-1.com

COPYRIGHT

Text::Phonetic::Phonix is Copyright (c) 2006,2007 Maroš. Kollár. All rights reserved.

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

The full text of the license can be found in the LICENSE file included with this module.

SEE ALSO