A Relational View of Pattern Discovery
Résumé
The elegant integration of pattern mining techniques into database remains an open issue. In particular, no language is able to manipulate data and patterns without introducing opaque operators or loop-like statement. In this paper, we cope with this problem using relational algebra to formulate pattern mining queries. We introduce several operators based on the notion of cover allowing to express a wide range of queries like the mining of frequent patterns. Beyond modeling aspects, we show how to reason on queries for characterizing and rewriting them for optimization purpose. Thus, we algebraically reformulate the principle of the levelwise algorithm.