|
Data Structures and Algorithms
with Object-Oriented Design Patterns in C# |
Ideally, given a set of 
distinct keys,
,
the set of hash values 
contains no duplicates.
In practice, unless we know something about the keys chosen,
we cannot guarantee that there will not be collisions.
However, in certain applications we have some specific knowledge
about the keys that we can exploit to reduce the likelihood of a collision.
For example,
if the keys in our application are telephone numbers,
and we know that the telephone numbers are all likely to be from
the same geographic area,
then it makes little sense to consider the area codes in the hash function,
the area codes are all likely to be the same.
Copyright © 2001 by Bruno R. Preiss, P.Eng. All rights reserved.