Home // DATA ANALYTICS 2014, The Third International Conference on Data Analytics // View article
Authors:
Daniel B. Rajchwald
Thomas J. Klemas
Keywords: Sensemaking; adaptive clustering; spectral clustering; network theory; silhouette; k-means.
Abstract:
Static and evolutionary clustering approaches exist that enable dynamically adaptive cluster analysis of large networks. These techniques are typically based on any of the traditional techniques, such as k-means, spectral, Kerninghan-Lin, and other partitioning or clustering algorithms. In this paper, we utilize spectral clustering and k-means as the fundamental clustering mechanisms but combine adaptive and evolutionary clustering to capture problem dynamics. We apply our approach to analyze a complex, dynamic multiple edge set network that was used to model call data from the Ivory Coast compiled from France Telecom/Orange anonymized call records over a 5 month period. Our methods are used to identify important but non-evident structural groupings, resolve community clusters, develop insights based on the evolving structure and associated history, and to make sense of the raw data, the ultimate objective for Sensemaking technologies.
Pages: 100 to 104
Copyright: Copyright (c) IARIA, 2014
Publication date: August 24, 2014
Published in: conference
ISSN: 2308-4464
ISBN: 978-1-61208-358-2
Location: Rome, Italy
Dates: from August 24, 2014 to August 28, 2013