Crowdsourcing Ontology Lexicons

Crowdsourcing Ontology Lexicons Bettina Lanser, Christina Unger, Philipp Cimiano Semantic Computing Group, CITEC, Bielefeld University Inspiration 1/ Zehlendorfer Damm 201, 33619 Bielefeld, Germany {blanser, cunger, cimiano}@cit-ec.uni-bielefeld.de Abstract In order to make the growing amount of conceptual knowledge available through ontologies and datasets accessible to humans, NLP applications need access to information on how this knowledge can be verbalized in natural language. One way to provide this kind of information are ontology lexicons, which apart from the actual verbalizations in a given target language can provide further, rich linguistic information about them. Compiling such lexicons manually is a very time-consuming task and requires expertise both in Semantic Web technologies and lexicon engineering, as well as a very good knowledge of the target language at hand. In this paper we present an alternative approach to generating ontology lexicons by means of crowdsourcing: We use CrowdFlower to generate a small Japanese ontology lexicon for ten exemplary ontology elements from the DBpedia ontology according to a two-stage workflow, the main underlying idea of which is to turn the task of generating lexicon entries into a translation task; the starting point of this translation task is a manually created English lexicon for DBpedia. Comparison of the results to a manually created Japanese lexicon shows that the presented workflow is a viable option if an English seed lexicon is already available. Keywords: Ontology lexicalization, crowdsourcing, DBpedia 1. Introduction 1.1. Motivation As the amount of formalized conceptual knowledge avail- able through ontologies and datasets such as DBpedia (Bizer et al., 2009) grows, there is an increasing need to make this knowledge accessible to humans in an easy and intuitive way. One way to accomplish this is by means of language technology, e.g. in the form of question answer- ing systems, that allows users to access and query reposito- ries of conceptual knowledge through natural language. Of course, systems of this kind do not rely solely on concep- tual knowledge, but also need access to lexical information about how the elements described in such repositories may be verbalized in a given language. Ontology languages support the inclusion of such informa- tion to a certain extent, e.g. by means of rdfs:label or SKOS properties. However, often the amount and depth of linguistic information offered this way is very limited. In case of DBpedia, for instance, the majority of individ- uals do not have any language label at all, and while most other ontology elements are assigned an English label, cov- erage for other languages is very restricted; for example, only around ten percent of DBpedia’s classes and proper- ties have a Japanese label, as shown in Table 1. Moreover, even when a label in the language of interest is provided, no further linguistic information is given, such as part-of- speech information, inflectional forms or subcategorization frames. In addition, labels only capture one canonical way of verbalizing an ontology element, but they do not pro- vide lexical variants. For example, for the DBpedia prop- erty spouse the English label spouse …

Content truncated to 500 words