Is a kind of data structure to help perform some simple check in a large set of data, such as: find number of unique items, most frequent items or check whether some item exists in the data set.

They generally use a hash function to randomize and compactly represent a set of items.

The pros is, they use very little memory and have constant query time.