A Tool Chain for a Lightweight, Robust and Uncertainty-based Context Classification System (CCS)

Conference: ARCS 2010 - 23th International Conference on Architecture of Computing Systems
02/22/2010 - 02/23/2010 at Hannover, Germany

Proceedings: ARCS 2010

Pages: 9Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Günther, Henning; Simrany, Firas El; Berchtold, Martin; Beigl, Michael (Institut für Betriebssysteme und Rechnerverbund, Technische Universität Braunschweig)

In this paper we present a tool chain developed to support a Context Classification System (CCS). The CCS is especially designed to run even on lightweight commodity phones with a high detection rate and low calculation effort. The main design aspect considered while building the tool chain was the support for most common users and not only developers. Moreover, the CCS we are using proves to be a robust context recognition method, which supports the gain of both context classes and a fuzzy uncertainty value describing the confidence of their classification.