MetadataShow full item record
An ontology represents knowledge in the form of concepts and concept taxonomies, as well as relationships existing among them. Domain-specific ontologies serve a very important purpose, not only as an integral component of different semantics-driven applications, but also as an important source of knowledge in those domains. Ontological approach to data integration challenge allows for creating a unified resource for a specific domain, which precisely represents the domain knowledge and enables hypothesis generation based on integrative analyses of existing data in one place. A high quality domain ontology should accurately define the domain knowledge in terms of classes and relationships, which are populated either directly by a community of researchers or using instances obtained from well-curated data sources. Although the data sources containing the relevant data to fill these concepts and relationships are usually available, the data of interest is frequently “buried” among large amounts of other irrelevant data and hence, it is challenging for researchers to find all data about a domain of interest and to assemble it into a useful block of knowledge. We present mOntage, a framework for creating domain-specific ontologies built from existing data sources. The framework approaches the data integration challenge by montaging the subsets of relevant data from different data sources and creating a unified Ontology in one place. The framework encompasses Creation and Population, Verification, Assisted Graphical Querying and Updating and Versioning of the created ontology.