...
  my $docsProcessed;
  my $size = scalar (@{$arrayDocs});    
  for (my $i = 0; $i < ($size - 1); $i++) {
    # next if already processed
    next if (defined $docsProcessed->{$arrayDocs->[$i]});

    for (my $j = $i + 1; $j < ($size - 1); $j++) {
      next if (defined $docsProcessed->{$arrayDocs->[$j]});

      $similarity = similarity($arrayDocs->[$i], $arrayDocs->[$j], $self->{THRESHOLD});

      if ($similarity >= $self->{THRESHOLD}) {
	      # Add the processed document into the cluster
	      push (@{$clusters->{$arrayDocs->[$i]}}, $arrayDocs->[$j]);
				
				# Add the document to the processed hash
	      $docsProcessed->{$arrayDocs->[$j]} = 1;
      }
    }
  }
...