Streamlining Vocabulary Conversion to SKOS: A YAML-Based Approach to Facilitate Participation in the Semantic Web.

Controlled vocabularies, such as classification schemes, glossaries, taxonomies, or thesauri, play an important role in many Web services. One of the main areas of application of controlled vocabularies is the domain of information retrieval systems, as they can be used to improve the findability of resources. For instance, concepts described in a vocabulary may be used to uniquely classify resources, to tag them with relevant keywords, or to annotate them with domain-specific attributes. The Simple Knowledge Organization System (SKOS) is an established data model of the Semantic Web domain that can be used to describe vocabularies in a semantically structured format. However, modelling a vocabulary is oftentimes highly time demanding, labor-intensive, and requires both familiarity with basic Semantic Web technologies and expertise in the application domain. This complicates both the development of new vocabularies and the conversion of existing vocabularies into the RDF data model. We propose an intermediate, YAML-based format to express concepts and their relationships hierarchically. The intermediate format can be converted automatically into a SKOS vocabulary using a command-line conversion program. To demonstrate the feasibility of our approach, we selected 26 vocabularies of highly diverse formats, expressed them in the proposed intermediate format, which was subsequently converted in an automated manner into the corresponding SKOS vocabulary using our yaml2skos program. Our approach enables users with little to no familiarity with the Semantic Web to develop SKOS vocabularies, thereby lowering the barrier to participation in the Semantic Web landscape.