Mobile devices capabilities have increased tremendously over the last few years, enabling mobile users to run resource-heavy applications, such as route planners, web browsers and games, at any time and any place. Much work has been done to facilitate interaction with mobile applications, in mobile settings and on devices with relatively small screens and cumbersome input features. For instance, multi-touch, gesture and voice recognition are deployed to easily execute certain actions, while obtrusiveness of mobile interactions is adapted to suit the user’s current situation (e.g., in a meeting). However, mobile interactions still have inherent limitations, mainly due to the fact they are mobile in the first place; mobile users often do not have the time, or the comfortable desktop-setting, to interact with mobile applications. In order to cope with such limitations, the mobile user’s context can further be leveraged. Context is defined as any piece of information related to application interaction, including information on the user’s surroundings as well as the user and device. By automatically presenting information and services suiting the user’s current context, mobile interactions can be enhanced. For example, considering the mobile user’s current surroundings and preferences, he can be notified of nearby shops selling items on his shopping list, or nearby public transportation stations leading back to his hotel.

In this dissertation, we present a client-side framework to provision context in mobile settings, which exploits recent evolutions in mobile device technology and the World Wide Web. By leveraging increased mobile processing power and memory capacity, computationally intensive tasks, such as context interpretation, integration and dissemination, are performed locally on the mobile device itself. Furthermore, the machine-readable Semantic Web, the next step in the evolution of the Web, is utilized as an online platform for retrieving context data. Much useful context information, describing people, places and things in the user’s vicinity, is already captured in small, machine-readable online web sources; including websites (e.g., shops, monuments) and online RDF files (e.g., person profiles). As websites are being increasingly semantically annotated, making the meaning of their content explicit, many of them have become fully-fledged semantic data sources as well. In order to achieve transparent, integrated query access to such small online semantic sources, this dissertation further presents a mobile, client-side query service. This query service comprises indexing and caching components to enable querying such an online dataset on mobile devices.