KnoFuss represents a modular architecture for automated knowledge fusion of Semantic Web repositories. Knowledge fusion involves two main subtasks: coreference resolution and inconsistency resolution. Different datasets often contain information about the same entities but refer to them using different URIs. Identifying such individuals is the task of coreference resolution. Moreover, pieces of data defined in different knowledge bases can influence each other (e.g., they can be mutually contradictory). Inconsistency resolution is needed to process such cases. The KnoFuss architecture allows different methods for coreference resolution and inconsistency resolution to be combined in order to improve the quality of resulting data.
Nikolov, Andriy, d'Aquin, Mathieu, Motta, Enrico, Unsupervised learning of link discovery configuration
Nikolov, Andriy, Motta, Enrico, Capturing emerging relations between schema ontologies on the Web of Data
Nikolov, Andriy, Uren, Victoria, Motta, Enrico, Data linking: capturing and utilising implicit schema-level relations
Nikolov, Andriy, Uren, Victoria, Motta, Enrico, De Roeck, Anne, Overcoming schema heterogeneity between linked semantic repositories to improve coreference resolution
Nikolov, Andriy, Uren, Victoria, Motta, Enrico, De Roeck, Anne, Towards data fusion in a multi-ontology environment