Mining a natural language corpus for multi-relational association rules
Luc De Raedt, Luc De Raedt · Lirias (KU Leuven) · 1997
Association rules are generally recognized as a highly valuable type of regularities and various algorithms have been presented for efficiently mining them in large databases. To the best of our knowledge, the application of these algorithms is so far restricted to databases that consist of a single relation composed of a set of binary attributes. We describe how these restrictions can be overcome through the combination of the available algorithms with standard techniques from the field of inductive logic programming. We present the algorithm AprioriRel, which extends Apriori [ Agrawal et al., 1996 ] to mine association rules in multiple relations. Whereas in Apriori each example is described by means of a single tuple, in AprioriRel each example is viewed as a separate database with a selection, from multiple relations, of all tuples related to the example. Accordingly, the association rules discovered by AprioriRel may combine information from various relations to statements of th...