use strict; use Set::Similarity::Cosine; my $cosine = Set::Similarity::Cosine->new; open(DATA,"File.txt") or die " File not "; my @data; my $linha; my $lineIni; my $lineEnd; while () { $linha = $_."\n"; push @data, {map { $_ => 1} split }; } for my $i (0 .. $#data-1) { for my $j ($i + 1 .. $#data) { my $similar = $cosine->Similarity($data[$i],$data[$j]); $similar =~s/0\.([0-9]{2}).+/\1/i; if ($similar > 79) { #print "A linha $lineIni Tem $i similar com a linha $j\n"; print "line $i is $similar line $j\n"; open (DATA, ">>RESULT.txt"); print DATA "$j\n"; close(DATA); } } }