When Real-World Database Constraints meet Para-consistent Logics

A/Prof. Sebastian Link

School of Information Management, Centre for Logic, Language and Computation, The Victoria University of Wellington, New Zealand

Date and time: 11.30am - 12.30pm, Friday 22th January, 2010

Venue: Floor 8, Building 14 (IS Common Area)

Abstract:

Constraints can restrict the possible instances of a database system to those considered meaningful for the underlying application domain. Therefore, they are fundamental to database design, efficient update processing and semantic query optimization. While research on this topic has been extensive, currently existing theories of database constraints apply to only a small number of instances that occur in real database systems.

In this talk we will develop a theory that i) applies to real-world instances as permitted by SQL (the data definition and query language standard), ii) subsumes several existing orthogonal theories, and iii) results in efficient tools for effective constraint and database management. Specifically, we axiomatize the implication of a large class of database constraints, and establish almost linear-time algorithms that decide whether a database constraint is implied by a given set of constraints.  Furthermore, we show that the implication of several classes of constraints corresponds exactly to the logical implication of a number of para-consistent logic fragments, including Graham Priest's well-known Logic of Paradox. These correspondences provide database administrators with an effective mechanism to balance the expressiveness and tractability of consequence relations, and to control the degree by which the existing classical theory of database constraints can be soundly approximated in practice.

No previous knowledge of databases or logics is assumed.


About the speaker:

From Sebastian's web page:

Sebastian is Associate Professor of e-Commerce at SIM. Before joining VUW he was a member of the Department of Information Systems and the Information Science Research Center at Massey University. Sebastian's research is considered to be on the leading edge of international research in database theory. He was a recipient in the 2008 round of Marsden funding for his project Cardinality Constraints for XML: Challenging the Trade-off between Expressiveness and Tractability, 2009-2011.


 


Seminar Organisation

Seminars are free and open to the general public. No booking is necessary. If you are interested in giving a presentation in this seminar series, or to make suggestions for speakers, please contact Sebastian Sardina, the seminar co-ordinator.