Institute for Web Science and Technologies · Universität Koblenz - Landau

Improving a Computer Languages Ontology Based on Bad Smells

[go to overview]
Marcel Heinz

As the currently biggest online available encyclopedia, Wikipedia is a target of ontology extraction approaches. We assemble a suite of bad smells to perform a criteria based evaluation on an extracted ontology that aims at giving a classification taxonomy. We pay attention to Wikipedia's peculiarities that are described in guidelines. Next, we propose a suite of transformations (prunings and refactorings) that can be used for ontology improvement through a domain expert. At last, we derive a bad smell topology, which offers a clear improvement process. We apply the approach on an extracted computer languages ontology and explain our evaluation results.

31.05.16 - 10:00
B 016