June 1st, 2012 |
Classical approaches of computer science do not scale well for today’s large and complex software-intensive systems. Software systems cannot be considered in isolation, since they are connected among each other and interact massively. Instead they are to be designed as parts of a larger IT Ecosystem. In analogy to biological ecosystems, IT Ecosystems are based on the balance between individuals (autonomy) and sets of rules (control) defining equilibria within an IT Ecosystem. Maintaining and continuously evolving IT Ecosystems requires deep understanding of this balance.
In the IT Ecosystems subproject LocCom (local communities) we develop methods, concepts, and tools for decentralized IT Ecosystems. Our part is especially the resaerch of context detection, handling and pattern recognition.
One example is smart phones, that have become a powerful platform for wearable context recognition. We present a service/cloud-based recognition architecture which creates an evolving classification system using feedback from the user community. The approach utilizes classifiers based on fuzzy inference systems which use live annotation to personalize the classifier instance on the device. Our recognition system is designed for everyday use: it allows flexible placement of the device
(no assumed or fixed position), requires only minimal personalization effort from the user (1–3 minutes per activity) and is capable of detecting a high number of activities. The components of the service are shown in an evaluation scenario, in which recognition rates up to 97% can be achieved for ten activity classes.
Technische Universität Clausthal, Leibniz Universität Hannover
Context detection and handling, pattern recognition
[mendeley type=”groups” id=”8c337d2b-91fd-3f04-9db0-60b09db9bea4″ filter=”author=berchtold” sortby=”year” sortorder=”desc” groupby=”year”]