An alternative approach is to use HTML::TokeParser to extract the 'linked' text from anchors:

#!/usr/bin/perl -w

use strict;
use HTML::TokeParser;

my $p = HTML::TokeParser->new('/tmp/ml_test.html');

while ( my $t = $p->get_tag('a')) {
  print $p->get_trimmed_text('/a'), "\n";
}
[download]

--sacked