Query simplification by using knowledge about data semantics

In this paper the problem of query trasformation semantic properties of data is addressed. In particular a trasformation rules is defined, and then theorems stating correctness of the rules are given. The theorems ensure that the query obtained by applying the trasformations to a given query is equivalent to the original one, that is the two queries define the same set of tuples. Since join elimination clustering index introduction, empty query test are allowed by such trasformations, they become useful in the process of query optimization.