in reply to Identifying Files with Similar Names
There are a lot of other algorithms to check for similarity which might help you, e.g the Levenshtein-distance. A related implementation on CPAN is Text::Levenshtein.