Show simple item record

dc.contributor.advisorHandschuh, Siegfried
dc.contributor.advisorCunningham, Hamish
dc.contributor.authorDavis, Brian Patrick
dc.description.abstractCreating formal data is a high initial barrier for small organisations and individuals wishing to create ontologies and thus benefit from semantic technologies. Part of the solution comes from ontology authoring, but this often requires specialist skills in ontology engineering. Defining a Controlled Natural Language (CNL) for formal data description can enable naive users to develop ontologies using a subset of natural language. How- ever despite the benefits of CNLs, users are still required to learn the correct syntactic structures in order to use the Controlled Language properly. This can be time consuming, annoying and in certain cases may prevent uptake of the tool. The reversal of the CNL authoring process involves generation of the controlled language from an existing ontology using Natural Language Generation (NLG) techniques, which results in a round trip ontology authoring environment: one can start with an existing imported ontology (re)produce the CNL using NLG, modify or edit the text as required and subsequently parse the text back into the ontology using the CNL authoring environment. By introducing language generation into the authoring process, the learning curve associated with the CNL can be reduced. While the creation of ontologies is critical for the Se- mantic Web, without a critical mass of richly interlinked metadata, this vision cannot become a reality. Manual semantic annotation is a labor-intensive task requiring training in formal ontological descriptions for the otherwise non-expert user. Although automatic annotation tools attempt to ease this knowledge acquisition barrier, their development often requires access to specialists in Natural Language Processing (NLP). This challenges researchers to develop user-friendly annotation environments. While CNLs have been applied to ontology authoring, little research has focused on their application to semantic annotation. In summary, this research applies CNL techniques to both ontology authoring and semantic annotation, and provides solid empirical evidence that for certain scenarios applying CNLs to both tasks can be more user friendly than standard ontology authoring and manual semantic annotation tools respectively.en_US
dc.rightsAttribution-NonCommercial-NoDerivs 3.0 Ireland
dc.subjectControlled natural languagesen_US
dc.subjectOntology authoringen_US
dc.subjectSemantic annotationen_US
dc.subjectNatural language generationen_US
dc.subjectNatural language processingen_US
dc.titleOn Applying Controlled Natural Languages for Ontology Authoring and Semantic Annotationen_US
dc.contributor.funderScience Foundation Ireland under Grant No. SFI/02/CE1/I131en_US
dc.contributor.funderEuropean project NEPOMUK under Grant No. FP6-027705.en_US
dc.local.noteThis thesis describes software, which allows non-expert users to author formal knowledge using a restricted form of machine understandable English, called Controlled Natural Language (CNL). It also describes novel software tools for associating machine understandable annotations to snippets of free text within a document using CNL. These semantic annotations can be leveraged for more efficient document search on the Web. Finally, the research presents user studies, which compare the tools described above with their present day equivalent. The results show that the CNL tools are more user-friendly than conventional tools.en_US

Files in this item


This item appears in the following Collection(s)

Show simple item record

Attribution-NonCommercial-NoDerivs 3.0 Ireland
Except where otherwise noted, this item's license is described as Attribution-NonCommercial-NoDerivs 3.0 Ireland