AI Techniques in a Context-Aware Ubiquitous Environment

Abstract

Nowadays, the mobile computing paradigm and the widespread diffusion ofmobile devices are quickly changing and replacing many commonassumptions about software architectures and interaction/communicationmodels. The environment, in particular, or more generally, the so-calleduser context is claiming a central role in everyday's use of cellularphones, PDAs, etc. This is due to the huge amount of data``suggested{''} by the surrounding environment that can be helpful inmany common tasks. For instance, the current context can help a searchengine to refine the set of results in a useful way, providing the userwith a more suitable and exploitable information. Moreover, we can takefull advantage of this new data source by ``pushing{''} active contentstowards mobile devices, empowering the latter with new features (e.g.,applications) that can allow the user to fruitfully interact with thecurrent context. Following this vision, mobile devices become dynamicself-adapting tools, according to the user needs and the possibilitiesoffered by the environment. The present work proposes MoBe: an approachfor providing a basic infrastructure for pervasive context-awareapplications on mobile devices, in which AI techniques (namely aprincipled combination of rule-based systems, Bayesian networks andontologies) are applied to context inference. The aim is to devise ageneral inferential framework to make easier the development ofcontext-aware applications by integrating the information coming fromphysical and logical sensors (e.g., position, agenda) and reasoningabout this information in order to infer new and more abstract contexts.