Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/13462
Title: On Query Languages for Linear Queries Definable with Polynomial Constraints
Authors: Vandeurzen, Luc 
Gyssens, Marc 
Van Gucht, Dirk
Issue Date: 1996
Publisher: Springer
Source: Freuder, Eugene C. (Ed.). Principles and Practice of Constraint Programming - CP96, Second International Conference, CP96, Cambridge, MA, USA, August 1996. Proceedings, p. 468-481
Series/Report: LECTURE NOTES IN COMPUTER SCIENCE
Series/Report no.: 1118
Abstract: It has been argued that the linear database model, in which semi-linear sets are the only geometric objects, is very suitable for most spatial database applications. For querying linear databases, the language FO + linear has been proposed. We present both negative and positive results regarding the expressiveness of FO+linear. First, we show that the dimension query is definable in FO + linear, which allows us to solve several interesting queries. Next, we show the non-definability of a whole class of queries that are related to sets not definable in FO+linear. This result both sharpens and generalizes earlier results independently found by Afrati et al. and the present authors, and demonstrates the need for more expressive linear query languages if we want to sustain the desirability of the linear database model. In this paper, we show how FO + linear can be strictly extended within FO + poly in a safe way. Whether any of the proposed extensions is complete for the linear queries definable in FO + poly remains open. We do show, however, that it is undecidable whether an expression in FO + poly induces a linear query.
Document URI: http://hdl.handle.net/1942/13462
ISBN: 3-540-61551-2
DOI: 10.1007/3-540-61551-2_94
Rights: Copyright 1996 Springer-Verlag Berlin-Heidelberg
Type: Proceedings Paper
Appears in Collections:Research publications

Show full item record

SCOPUSTM   
Citations

12
checked on Sep 6, 2020

Page view(s)

42
checked on May 15, 2022

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.