Centrality#

A population can be interpreted as a graph in which the individual agents make up the nodes of the graph, and the edge weights are determined by the distance between their belief systems. Agents can inhibit the centre of a graph or be far removed at the fringes. The normalised closeness centrality indicates this position for a single agent in a population relative to a distance measure. In the graph below, the top position has a distance of 1 to each of the others, and the others have a higher distance toward each other. The top agent thus is the most central node.

../_images/ncc-illustration.svg../_images/ncc-illustration-dark.svg

Hint

There can be multiple, far removed agents with maximum NCC in a population. In other words, a population graph can have multiple graph centres.

taupy.analysis.agreement.ncc(population, *, agent, measure=<function hamming_distance>)[source]#

Returns the normalised closedness centrality (NCC) of an agent in a population relative to a measure (see [Betz2013], Section 2.4).

Parameters:

  • population – Iterable containing agents’ belief systems.

  • agent – A single belief system.

  • measure – A distance measure between belief systems.