An efficient computation of frequent queries in a star schema

Abstract

Although the problem of computing frequent queries in relational databases is known to be intractable, it has been argued in our previous work that using functional and inclusion dependencies, computing frequent conjunctive queries becomes feasible for databases operating over a star schema. However, the implementation considered in this previous work showed severe limitations for large fact tables. The main contribution of this paper is to overcome these limitations using appropriate auxiliary tables. We thus introduce a novel algorithm, called Frequent Query Finder (FQF), and we report on experiments showing that our algorithm allows for an effective and efficient computation of frequent queries. © 2010 Springer-Verlag.