Service discovery and prediction on Pervasive Information System

Abstract

International audienceRecent evolution of technology and its usages, such as BYOD (Bring Your Own Device) and IoT (Internet of Things), transformed the way we interact with Information Systems (IS), leading to a new generation of IS, called the Pervasive Information Systems (PIS). These systems have to face heterogeneous pervasive environments and hide the complexity of such environment end-user. In order to reach transparency and proactivity necessary for successful PIS, new discovery and prediction mechanisms are necessary. In this paper, we present a new user-centric approach for PIS and propose new service discovery and prediction based on both user's context and intentions. Intentions allow focusing on goals user wants to satisfy when requesting a service. Those intentions rise in a given context, which influence the service implementation. We propose a service discovery mechanism that observes user's context and intention in order to offer him/her the most appropriate service satisfying her/his intention on the current context. We also propose a prediction mechanism that tries to anticipate user's intentions considering the user's history and the observed context. We evaluate both mechanisms and discuss advanced features future PIS will have to deal with

Machine Learning for a Context Mining Facility

“…In such systems, mining context data may have multiple uses: adaptation of system behavior and applications, recommendation of actions, data prediction, anticipation of user's needs, and decision making (e.g. [8] [9]). With the availability of context data, we may expect a generalization of such "smart" behaviors, built upon mining solutions.…”