Modeling Classification Systems in SKOS: Some Challenges and Best-Practice Recommendations

Abstract

Representing classification systems on the web for publication and exchange within the SKOS framework continues to be a challenge. This paper focuses on the differences between classification schemes and other families of KOS that make it difficult to express classifications without sacrificing a large amount of their semantic richness. Issues resulting from the specific set of relationships between classes and topics that defines the basic nature of any classification system are discussed. Where possible, different solutions with SKOS and OWL are proposed and examined.