Towards Context Awareness Using Symbol Clustering Map

J. Himberg, J. A. Flanagan, and J. Mäntyjärvi. Towards context awareness using symbol clustering map. In Proceedings of the Workshop for Self-Organizing Maps 2003 (WSOM2003), pages 249–254, Kitakyushu, Japan, 2003. [url]

———————

This paper present the Symbol Clustering Map methods that is used to recognizing the context of use of a mobile phone. The authors present an approach for context recognition that is based in a vectorial fusion of sensory information. Two sorts of fusions are presented: a multi-source fusion and a temporal fusion of events. The Symbol Clustering Map (SCM) is based on a lattice structure, generally two-dimensional, with a total of M x M lattice points. For each node k of the lattice, a symbol set is associated, combined with a weight vector.

Tags: ,

Leave a Reply

Your email address will not be published. Required fields are marked *