Roth, JörgHegering, Heinz-GerdLehmann, AxelOhlbach, Hans JürgenScheideler, Christian2019-04-032019-04-032008978-3-88579-228-4https://dl.gi.de/handle/20.500.12116/21288In this paper we present an approach for reasoning about continuous context variables. We introduce a probabilistic mechanism based on efficient geo-metric structures that avoid typical restrictions of existing approaches. Especially, we can model non-Gaussian distributions, negated statements and we can deal with external knowledge that is only accessible on demand.enText/Conference Paper1617-5468