Издательство Cambridge University Press, 2010, -360 pp.
The mapping between knowledge representation and natural language is fast becoming a focal point of both knowledge engineering (KE) and computational linguistics (CL). Ontologies have a special role to play in this interface. They are essential stepping stones (a) from natural language to knowledge representation and manipulation and (b) from formal theories of knowledge to their application in (natural language) processing. Moreover, the emergence of the Semantic Web initiative presents a unique opportunity to bring research results in this area to real-world applications, at the leading edge of human-language technology. An essential and perhaps foundational aspect of the mapping between knowledge representation and natural language is the interface between ontologies and lexical resources. On the one hand, their integration includes, but is not restricted to, the use of ontologies (a) as language-independent structures of multilingual computational lexicons and (b) as powerful tools for improving the performance of existing lexical resources on various natural language processing (NLP) tasks such as word-sense disambiguation. On the other hand, lexical resources constitute a formidable source of information for generating ontological knowledge both at foundational and domain levels.
This current volume aims to be an essential general reference book on the interface between ontology and lexical resources. Given the fast developments in this new research direction, we introduce a general framework with a terminology to accommodate both ontological and lexical perspectives. However, to show its relevance and its direction of development, the focus of the book is on the current applications of such a framework. The chapters present an up-todate overview of contributions to the field, from both CL and KE communities. Hence it constitutes a self-contained reference from introduction to the domain to the latest applications and tools developed.
We hope this book presents a productive dialogue among the following communities: those from the CL research community who are interested in ontology and ontologies, those from Semantic Web and KE communities who pay special attention to natural language, and those from the linguistic community who want to know how lexical knowledge can be formalized and computed. As suggested by the title, this book not only aims to reduce the gaps among these research communities, but does so on a solid ground provided by a precise framework and definitions on all sides. The book should be accessible to Master and PhD students and has been used as material for such a course.
Part I Fundamental aspectsOntology and the lexicon: a multidisciplinary perspective
Formal ontology as interlingua: the SUMO and WordNet linking project and global WordNet
Interfacing WordNet with DOLCE: towards OntoWordNet
Reasoning over natural language text by means of FrameNet and ontologies
Synergizing ontologies and the lexicon: a roadmap
Part II Discovery and representation of conceptual systemsExperiments of ontology construction with Formal Concept Analysis
Ontology, lexicon, and fact repository as leveraged to interpret events of change
Hantology: conceptual system discovery based on orthographic convention
What’s in a schema?
Part III Interfacing ontologies and lexical resourcesInterfacing ontologies and lexical resources
Sinica BOW (Bilingual Ontological WordNet): integration of bilingual WordNet and SUMO
Ontology-based semantic lexicons: mapping between terms and object descriptions
Merging global and specialized linguistic ontologies
Part IV Learning and using ontological knowledgeThe life cycle of knowledge
The Omega ontology
Automatic acquisition of lexico-semantic knowledge for question answering
Agricultural ontology construction and maintenance in Thai