In my application the user types in the (potentially faulty) name of an amenity type, for example "supermarked" for "supermarket" and I want to match this with my list of amenities, so I can present the user the highest match like
"you typed in "supermarked, do you mean "supermarket"?".
Which metric would be best suited for this task?
It should be case insensitive and be suited for a wide array of languages, german and english are priority though.
Jaro-Winkler sound like good bets. Hope it helps.
Ok thank you! I also noticed I don't need a case insensitive metric because I can just lowercase both strings before the comparison.