Ontology engineering: From an art to a craft : The case of the data mining ontologies

Abstract

In this paper, we report on our experience and discuss the problems we encountered while designing, implementing and revising a set of ontologies describing the domain of data mining. After giving a short description of the ontologies we have developed, we focus on a set of key issues that we think are important and need to be addressed by the ontology engineering community. These include ontology evaluation, testing, versioning, the use of design patterns, the use of IT portal(s), re-usability, and compatibility. To illustrate the key issues we provide examples that originate from our work on the ontologies for data mining. We conclude the paper with a summary and some suggestions that we believe should be addressed by the ontology engineering research community.