Abstract | ||
---|---|---|
This paper presents a novel approach for the zone design problem that is based on techniques from the field of complex networks research: community detection by betweenness centrality and label propagation. A new algorithm called Spatial Graph based Clustering by Label Propagation (SGCLAP) is introduced. It can deal with very large spatial clustering problems with time complexity O (n logn ). Besides, we use a parallel version of a betweenness-based community detection algorithm that outputs the graph partitioning that maximizes the so-called modularity metric. Both these methods are put at the centre of an effort to build an open source interactive high performance computing platform to assist researchers working with population data. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1007/978-3-642-05258-3_16 | MICAI |
Keywords | Field | DocType |
betweenness-based community detection algorithm,complex networks research,complex networks approach,label propagation,large spatial,graph partitioning,interactive high performance computing,spatial graph,new algorithm,community detection,demographic zonification,betweenness centrality,time complexity,complex network | Network science,Population,Data mining,Computer science,Theoretical computer science,Betweenness centrality,Artificial intelligence,Complex network,Time complexity,Graph partition,Cluster analysis,Machine learning,Modularity | Conference |
Volume | ISSN | Citations |
5845 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 7 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Alberto Ochoa | 1 | 24 | 6.95 |
Beatriz Loranca | 2 | 0 | 0.34 |
Omar Ochoa | 3 | 10 | 5.19 |