Abstract. We present a method for rapid development of benchmarks for Semantic Web knowledge base systems. At the core, we have a synthetic data generation approach for OWL that is...
Pervasive environments can be hard to configure and interact with using handheld computing devices, due to the mismatch between physical and digital worlds. Usually, smart resourc...
Today's knowledge workers have to spend a large amount of time and manual effort on creating, analyzing, and modifying textual content. While more advanced semantically-orient...
This paper describes Seeker, a platform for large-scale text analytics, and SemTag, an application written on the platform to perform automated semantic tagging of large corpora. ...
Stephen Dill, Nadav Eiron, David Gibson, Daniel Gr...
Background: Neuroscientists often need to access a wide range of data sets distributed over the Internet. These data sets, however, are typically neither integrated nor interopera...
Hugo Y. K. Lam, Luis N. Marenco, Tim Clark, Yong G...