Here is how I would probably do it, assuming your keyword matches are literal (not patterns).

use strict;
use warnings;

open my $keyword_fh, '<', 'keywords.txt' or die $!
my %keywords;
while ( <$keyword_fh> ) {
    chomp;
    $keywords{ $_ } = '';
}
close $keyword_fh;

open my $data_fh, '<', 'data.txt' or die $!;
while( my $line = <$data_fh> ) {
    chomp $line ;
    my @found = grep { exists $keywords{ $_ } }
        $line =~ m/([^\W\d_]+)/g;
    print "Line $.: @found\n";
}
close $data_fh;
[download]

This assumes 'words' are entirely alphabetical (no embedded characters such as '-' or " ' " (apostrophe)). It could be modified to deal with those too.