Thus, the similarity is 1 if the words are identical and 0 if they are totally different.
Now consider the similarity value  for a specific potential cognate pair  ,  . (Now these are two words with a same meaning!) By itself, this value is not very telling. What we want to estimate, is how likely it is for a random pair of words from the two languages to have the same (or higher) similarity value. We estimate this probability,  , as the number of pairs with the similarity greater or equal to  , divided by the overall number of pairs.
Share with your friends: |
