Ontology evaluation with Protégé using OWLET

Amalgamation of formalised knowledge and real-world datasets is a pivotal challenge in the realm of information and communication technologies. Semi-automated classification of datasets can be performed by utilisation of ontologies. The detection process of image objects in Very High Resolution Satellite Imagery (VHRSI) gives a prominent example. The process of refinement of formalised expert knowledge within the related ontology still remains a challenging and time-consuming task. In this paper, the JSON2OWL Converter (OWLET) extension for Protégé is presented which supports experts during this refinement phase. The extension offers an integrated approach to transfer real-world dataset objects into the ontology modelling software for semi-automated classification. This transfer is achieved by combining open standard formats from both domains, the (Geo) Web domain (GeoJSON) and the Web ontology domain (OWL2). Thereby OWLET supports the process of accuracy analysis and accuracy fostering. By utilising the OWLET extension, experts can not only speed up their classification procedure considerably, but they can also refine their formalised knowledge by using the results of the classification process in conjunction with the outcomes of the accuracy analysis.