Controlled Natural Languages and the Semantic Web
Paul Richard Smart · ePrints Soton (University of Southampton) · 2008
The Semantic Web grew out of research into formal logics, and many of the representational and technological commitments of the Semantic Web reflect this heritage. Notwithstanding the obvious advantages for machine processability and reasoning, the logical underpinnings of the Semantic Web present a number of usability challenges for human end-users, especially when it comes to the communication and exploitation of domain-relevant knowledge/information. In this report, we review one approach to the resolution of these usability challenges – an approach that is based on the use of natural language user interfaces. Natural language interfaces exploit the medium of natural language in order to support end-users with respect to a range of capabilities; for example, the authoring of knowledge content, the retrieval of information from semantic repositories, and the generation of natural language texts from formal ontologies. The current report reviews the state-of-the-art with respect to natural language interfaces in all these capability areas. It also attempts to explore issues associated with the use of controlled natural languages as the representational basis for knowledge content on the Semantic Web. The idea that controlled natural languages could serve as a replacement for conventional Semantic Web ontologies is explored in some detail, as is the notion that natural language interfaces could contribute to the usability of the Semantic Web without requiring the wholesale replacement of pre-existing approaches. The report concludes with a discussion about the potential contribution of natural language interfaces to the emergence of truly hybrid (human/machine) intelligent systems. This idea sees natural language interfaces to the Semantic Web as an essential component of a future extended cognitive system – one that co-opts elements of human cognition with the computational and representational resources of a globally-extensive and semantically-scaffolded information environment.