User Tools

Site Tools


algorithms:closeness

Closeness Centrality

Measures the mean distance from a node to all remaining nodes. It is a non-normalized quantity. A high value of closeness centrality indicates that the node is a central node of the network, i.e. it can be reached easily from most of the other nodes. Parallel edges are allowed and may be used in the geodesic distance calculation. A link property, considered as distance can be taken into account in the distances calculation.

Input parameter Type Default Description
Distance text None Link property that represents distances. The property values should be positive numbers.


The next plot shows closeness centrality for a little network, where small values are colored in blue and large values in yellow/red. We can easily identify the most central node, which would be the best-connected person in the network. For example, marketing strategies that start on him will spread very fast compared with other starting nodes.



References:

  • M.E.J. Newman, Networks, an introduction, Oxford University Press, 2010, ISBN: 978-0-19-920665-0, sec. 7.6.
algorithms/closeness.txt · Last modified: 2018/12/20 14:19 by systems