On containment of conjunctive queries with arithmetic comparisons

Afrati, F; Li, C; Mitra, P

dc.contributor.author	Afrati, F	en
dc.contributor.author	Li, C	en
dc.contributor.author	Mitra, P	en
dc.date.accessioned	2014-03-01T01:21:09Z
dc.date.available	2014-03-01T01:21:09Z
dc.date.issued	2004	en
dc.identifier.issn	0302-9743	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/16098
dc.subject	Conjunctive Queries	en
dc.subject	Domain Knowledge	en
dc.subject	Problem Complexity	en
dc.subject.classification	Computer Science, Theory & Methods	en
dc.title	On containment of conjunctive queries with arithmetic comparisons	en
heal.type	journalArticle	en
heal.identifier.primary	10.1007/978-3-540-24741-8_27	en
heal.identifier.secondary	http://dx.doi.org/10.1007/978-3-540-24741-8_27	en
heal.language	English	en
heal.publicationDate	2004	en
heal.abstract	We study the following problem: how to test if Q2 is contained in Q1, where Q1 and Q2 are conjunctive queries with arithmetic comparisons? This problem is fundamental in a large variety of database applications. Existing algorithms first normalize the queries, then test a logical implication using multiple containment mappings from Q 1 to Q2. We are interested in cases where the containment can be tested more efficiently. This work aims to (a) reduce the problem complexity from IIP2-completeness to NP-completeness in these cases; (b) utilize the advantages of the homomorphism property (i.e., the containment test is based on a single containment mapping) in applications such.as those of answering queries using views; and (c) observing that many real queries have the homomorphism property. The following are our results. (1) We show several cases where the normalization step is not needed, thus reducing the size of the queries and the number of containment mappings. (2) We develop an algorithm for checking various syntactic conditions on queries, under which the homomorphism property holds. (3) We further reduce the conditions of these classes using practical domain knowledge that is easily obtainable. (4) We conducted experiments on real queries, and show that most of the queries pass this test. © Springer-Verlag 2004.	en
heal.publisher	SPRINGER-VERLAG BERLIN	en
heal.journalName	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)	en
heal.bookName	LECTURE NOTES IN COMPUTER SCIENCE	en
dc.identifier.doi	10.1007/978-3-540-24741-8_27	en
dc.identifier.isi	ISI:000189447000027	en
dc.identifier.volume	2992	en
dc.identifier.spage	459	en
dc.identifier.epage	476	en