Query Rewriting on Aggregate Queries over Uncertain Database

Abstract

Although integrity constraints effectively maintain certainty of data, and some situation may be satisfied with integrity constraints. This paper proposes uncertain databases and candidate databases. To uncertainty of tuple values in uncertain databases and their aggregation queries with different results for every candidate database, the work adjusts query semantics to employ query writing for computing the maximum and the minimum of aggregation attributes based on certain range semantics, the method can execute effectively SQL query in databases. The experiments show that the overloads of rewritten queries are longer than original queries, but the overloads should be accepted.