Learning query rules for optimizing databases with update rules

Abstract

In this paper, we are interested in coupling deductive database approaches and inductive logic programming. We focus on a particular approach to updating Datalog^neg databases based on two kinds of rules, namely update rules and query rules. In this approach, every fact to be inserted or to be deleted is stored in the database, in order to handle updates over extensional or intensional predicates in a sound and deterministic way.

However, two important problems occur: first, the overhead incurred by the storage of inserted and deleted facts may be important, and, second, the intensional database (i.e., the query rules) may enable to derive contradictions with respect to the extensional database (i.e., the facts stored in the database together with the update rules).

In order to cope with these difficulties, we study storage optimization, and then, we propose to use Machine Learning techniques in order to compute new query rules, so that the semantics of the resulting database contains the semantics of the original one and satisfies our optimization criterion.