Knowledge discovery in complex objects

Learning concepts and rules from structured (complex) objects is a quite ch allenging but very relevant problem in the area of machine learning and kno wledge discovery. In order to take into account and exploit the semantic re lationships that hold between atomic components of structured objects, we p ropose a knowledge discovery process, which starts from a set of complex ob jects to produce a set of related atomic objects (called contexts). The sec ond step of the process makes use of the concatenation product to get a glo bal context in which binary relations of individual contexts coexist with r elations produced by the application of some operators to individual contex ts. The last step permits the discovery of concepts and implication rules u sing the concept lattice as a framework in order to discover and interpret nontrivial concepts and rules that may relate different components of compl ex objects. This paper focuses on two main steps of the knowledge discovery process, namely data mining and interpretation.