Information Gathering in a Dynamic World
Thomas Hornung, Kai Simon, Georg Lausen
Abstract:Web resources with constantly fluctuating content, such as virtual market places, are becoming more and more relevant as information resources. Classic search engines, unfortunately, crawl and index the Web in sporadic intervals and therefore rely on outdated information. In this paper we present OntoGather, a framework based on ontology-driven inferences on dynamically gathered annotated instances from the Web, which consists of two main components: the Web data-extraction and annotation system ViPER and the deductive object-oriented database system Florid.