Dr. Christian M. Meyer

OntoWiktionary

Constructing an Ontology from the Collaborative Online Dictionary Wiktionary

Abstract. The semi-automatic development of ontologies is an important field of research, since existing ontologies often suffer from their small size, unaffordable construction cost, or limited quality of ontology learning systems. The main objective of this article is to introduce Wiktionary, which is a collaborative online dictionary encoding in­for­ma­tion about words, word senses, and relations between them, as a resource for ontology construction. We find that a Wiktionary-based ontology can exceed the size of, for example, OpenCyc and OntoWordNet. One particular advantage of Wiktionary is its multilingual nature, which allows the construction of ontologies for different languages. Additionally, its collaborative construction approach means that novel concepts and domain-specific knowledge are quick to appear in the dictionary.

For constructing our ontology OntoWiktionary, we present a two-step approach that involves (i) harvesting structured knowledge from Wiktionary and (ii) ontologizing this knowledge (i.e., the formation of ontological concepts and relationships from the harvested knowledge). We evaluate our approach based on human judgments and find our new ontology to be of overall good quality. To encourage further research in this field, we make the final OntoWiktionary publicly available and suggest integrating this novel resource with the linked data cloud as well as other existing ontology projects.

Eingereicht: 25.03.2011 | Veröffentlicht: 29.02.2012

OntoWiktionary’s user interface.

OntoWiktionary’s user interface.