by 'word class' i meant 'a set of words that are considered equivalent by whatever your equivalency relation is'

similarly for 'character class'

cf. https://en.wikipedia.org/wiki/Equivalence_class

what i thought the linked program did was that it counted how many of those there were

now on looking at it further i can see that it doesn't seem to be doing that but i don't have any idea what it does do

however, it definitely doesn't take into account the information you would need to learn anything about which candidate equivalency relation is better, which is something you'd need to examine at at least a word level, considering examples like größte, Σίσυφος, and the notoriously fatal sıkışınca/sikişinca pair