Try this. I project that it should complete your 410 billion comparisons in a little under 10 hours.

The main attempt at efficiency here is to invoke the regex once in global mode (/g) for each word, against a single large string containing all the words and have it return all the matches. It then filters just the matching ones for your specific exclusions.

#! perl -slw
use strict;

my @words = do{ local @ARGV = 'words.txt'; <> };
chomp @words;

my $all = join ' ', @words;

my $start = time;
my $n = 0;
for my $i ( @words ) {

    for my $j ( $all =~ m[ ([^ ]*$i[^ ]*) ]g ) {
        next
            if $j eq $i
            or $j eq "${i}s"
            or $j eq "${i}'s";
#        print "$j contains $i";
    }
}

printf STDERR "Took %d seconds for %d words\n",
    time() - $start, scalar @words;
[download]