If I'm understanding you correctly, it seems that you are attempting to use named capture groups and, if there's a match, retrieve the name of the matching group. If so, consider the following:

use strict;
use warnings;

my $string = 'word_one word_one word_two 2000 word_three 3000';

while ( $string =~ m/(?<Word>\w+)\s+(?<Digit>\w+)/g ) {
    my %groupNames;
    push @{ $groupNames{ $+{$_} } }, $_ for keys %+;
    print "Match : $1 GroupName(s): @{ $groupNames{$1} }\n";
}
[download]

Output:

Match : word_one GroupName(s): Word Digit
Match : word_two GroupName(s): Word
Match : word_three GroupName(s): Word
[download]

Group names and their associated values are contained in the hash %+. The above creates a hash of arrays (HoA) keyed on the (captured) values of %+, since two different keys may have identical values. The @{ $groupNames{$1} } notation uses the captured match as a key, and its associated value is the list of capture group names.

Hope this helps!