Dice's coefficient

• S = ( 2 * Nt ) / ( Nx + Ny )

#Obtain the number of Bigrams in the intersection between Word1 and Word2
function Intersection(BigramsWord1,BigramsWord2) {
    Nt=0;
    for (Bigram in BigramsWord1) {
      if (Bigram in BigramsWord2) {
          if (BigramsWord1[Bigram] <= BigramsWord2[Bigram]) 
               Nt = Nt + BigramsWord1[Bigram]
          else
               Nt = Nt + BigramsWord2[Bigram];
      }
    }
    return Nt; 
}

#Obtain the Bigrams and the number of Bigrams per Word 
function ObtainBigrams(Word,LettersWord,BigramsWord) { 
 Bigram = ""; 
    Cardinality = 0;  
 NumberBigrams = 0;
    WordLength=length(Word);
    for (i=1; i<=WordLength; i++) {         
     Bigram = Bigram""LettersWord[i]; 
     Cardinality++;
        if (Cardinality == 2) {
            Cardinality = 1;
            NumberBigrams++; 
         BigramsWord[Bigram]++; 
         Bigram=LettersWord[i];
        }
    }     
 return NumberBigrams; 
}
#Obtain the DICE's coefficient between two words
function DICE (Word1,Word2) {     
 split(Word1, LettersWord1,"");
    split(Word2, LettersWord2,"");
    Nx = ObtainBigrams(Word1,LettersWord1,BigramsWord1); 
 Ny = ObtainBigrams(Word2,LettersWord2,BigramsWord2);
    Nt = Intersection(BigramsWord1,BigramsWord2);
    if ((Nx+Ny) > 0)
       return (2*Nt)/(Nx+Ny)
    else
       return 0; 
}
BEGIN{
                  FS=",";
}
{
 Word1=$1;
    Word2=$2;
    print DICE(Word1,Word2); 
}