Data Exchange in the Presence of Queries with Arithmetic Comparisons

In this paper we study the problem of data exchange and query answering in the presence of arithmetic comparisons (ACs). The contributions of this work are the following: a) we extend the work in [FKMP03] by assuming source-to-target constraints, target constraints and queries to include ACs, b) we give an algebraic specification to compute a universal solution using a variant of the chase procedure (this in fact states by itself a contribution of this work since it is the first chase procedure in the literature that deals with arithmetic comparisons), c) we study how to compute the certain answers of a conjunctive query with ACs (CQAC) by using a universal solution and d) we study the computational complexity of computing the certain answers in this setting, e) we investigate how to use typed information to reduce the complexity of query answering, and f) we examine special cases where computing the certain answers can be done in polynomial time. Of independent interest is a test for checking the containment between two unions of CQACs.