Query reformulation with constraints

Abstract

Let Σ1, Σ2 be two schemas, which may overlap, C be a set of constraints on the joint schema Σ1 ∪ Σ2, and q1 be a Σ1-query. An (equivalent) reformulation of q1 in the presence of C is a Σ2-query, q2, such that q2 gives the same answers as q1 on any Σ1 ∪ Σ2-database instance that satisfies C. In general, there may exist multiple such reformulations and choosing among them may require, for example, a cost model. DisciplinesComputer Sciences Authors retain the right to post a pre-print version of the journal article on Internet web sites including electronic pre-print servers, and to retain indefinitely such version on such servers or sites. The right to post a revised personal version of the text of the final journal article (to reflect changes made in the peer review process) on the author's personal or institutional web site or server, incorporating the complete citation and with a link to the Digital Object Identifier (DOI) of the article.

A Calculus for Propagating Semantic Annotations Through Scientific Workflow Queries

“…In general, resolution methods, including specialized versions such as the Chase (see e.g. [12]), may not terminate, e.g., due to recursive rules and Skolem symbols. In contrast, our approach terminates, because we guide and limit the inference steps using the structure of the operator tree.…”

Schema Mapping Evolution Through Composition and Inversion

“…In other words, we are looking for a query q 1 to satisfy q(K) = q 1 (K), for every instance K over S ∪ S 1 such that K satisfies the union of the constraints in M 1 and M 1 . In turn this is an instance of the general problem of query reformulation under constraints (Deutsch et al, 2006), which can be solved by the chase and backchase method (Deutsch et al, 1999). The application of the chase and backchase method in this context consists of, first, applying the chase on q with the constraints in M 1 , and then on applying the (back) chase with the reverse constraints in M 1 , in order to find equivalent rewritings of q.…”

Interactive source registration in community-oriented information integration

“…The community-based integration architecture can support both types of evolution through techniques studied extensively in [27,29,14] as explained next. We will start by presenting the case of schema evolution within a single community and then we will show that coalescing of communities can be reduced to the former.…”

“…In particular, mappings are adapted to the new target schema using techniques presented in [27,29]. Similarly, the queries are rewritten against the new target schema by modeling the schema evolution as a mapping between the old and the new target schema and using solutions on rewriting queries under constraints (see [14]). A source owner can subsequently call RIDE as usual to adjust the contribution of the new mappings to the new application queries.…”