Simhash

What is Simhash?

Simhash is a method for creating a fixed-length "hash" or "fingerprint" of a variable-length input, like a text or document. Although it resembles a hash function and is a type of locally sensitive hashing, it is made to be more resilient to collision attacks, in which two separate inputs generate the same hash. Simhash creates a hash of each feature it divides into, which are referred to as "features," before combining them together to form the input. The final hash for the input is created by combining these hashes.