Andreas Behrend: A Uniform Fixpoint Approach to the Implementation of Inference Methods for Deductive Databases

Abstract:

Deductive databases allow for inferring new knowledge from stored base data using rules. In such rule-based systems three main database tasks have been deeply investigated, specifically query evaluation, update propagation and view updating. Over the last thirty years various inference mechanisms have been proposed for implementing these functionalities within a database system but these mechanisms are rarely used in commercial DB systems until now. One important reason for this is the lack of a uniform approach well-suited for implementation in an SQL-based system. In this presentation, we propose such a uniform approach in form of the soft consequence operator. In addition we introduce new transformation-based approaches to query evaluation, update propagation and view updating which commonly utilize this new inference method. The feasibility of our transformation-based approach to update propagation is shown by a number of successful research projects.