Relational methodology for data mining and knowledge discovery

摘要

Abstract. Knowledge discovery and data mining methods have been successful in many domains. However, their abilitiesnto build or discover a domain theory remain unclear. This is largely due to the fact that many fundamental KDD&DMnmethodological questions are still unexplored such as (1) the nature of the information contained in input data relative tonthe domain theory, and (2) the nature of the knowledge that these methods discover. The goal of this paper is to clarifynmethodological questions of KDD&DM methods. This is done by using the concept of Relational Data Mining (RDM),nrepresentative measurement theory, an ontology of a subject domain, a many-sorted empirical system (algebraic structure innthe first-order logic), and an ontology of a KDD&DM method. The paper concludes with a review of our RDM approach andn‘Discovery’ system built on this methodology that can analyze any hypotheses represented in the first-order logic and use anyninput by representing it in many-sorted empirical system.