#!/usr/local/bin/perl

# Copyright @ 1994 MIT 
# Written by Eric Brill 
#THIS SOFTWARE IS PROVIDED "AS IS", AND M.I.T. MAKES NO REPRESENTATIONS 
#OR WARRANTIES, EXPRESS OR IMPLIED.  By way of example, but not 
#limitation, M.I.T. MAKES NO REPRESENTATIONS OR WARRANTIES OF 
#MERCHANTABILITY OR FITNESS FOR ANY PARTICULAR PURPOSE OR THAT THE USE OF 
#THE LICENSED SOFTWARE OR DOCUMENTATION WILL NOT INFRINGE ANY THIRD PARTY 
#PATENTS, COPYRIGHTS, TRADEMARKS OR OTHER RIGHTS.  



# assumes lines with tagged text.
# extracts lexicon of form:
# word t1 t2 t3 .. tn
# where t1 is most likely tag for word, t2 ... tn are other seen
# taggings in no particular order.

while(<>) {
	s/^\s+//;
	if (! /^$/) {
	    @temp = split;
	    for ($count=0;$count<=$#temp;++$count){
		$theguy = $temp[$count];
		$theguy =~ s/^(.*)\/([^\/]+)$/$1 $2/;
		@temp2 = split(/\s+/,$theguy);
		++$seen{$temp2[0] . " " . $temp2[1]};
	    }
	}
    }
		
while(($key,$val) = each %seen) {
    @temp = split(/\s+/,$key);
    if ($val > $bestval{$temp[0]}) {
	$bestkey{$temp[0]} = $temp[1];
	$bestval{$temp[0]} = $val;
    }
}

while(($key,$val) = each %seen) {
    @temp = split(/\s+/,$key);
    if ($bestkey{$temp[0]} eq $temp[1]) {
	$toprint{$temp[0]} = $temp[1] . " " . $toprint{$temp[0]}; 
    }
    else {
	$toprint{$temp[0]} .= " " . $temp[1];
    }
}

while (($key,$val) = each %toprint) {
    $val =~ s/^\s+//;
    $val =~ s/\s+/ /g;
    print "$key $val\n"; }