#betweennesscentrality — Public Fediverse posts

Calculating different #centrality measures for a street #network takes longer than expected.

#DegreeCentrality is calculated in a few milliseconds. But oh boy. #ClosenessCentrality and #BetweennessCentrality are proper whoppers. For a network of 65 000 nodes, we're talking about 2+ hour calculation times for the closeness centrality, not to mention the betweenness.

Apparently switching to #igraph would provide a speed boost over #networkx but the convenience of #osmnx has won me over.

Calculating different #centrality measures for a street #network takes longer than expected.

#DegreeCentrality is calculated in a few milliseconds. But oh boy. #ClosenessCentrality and #BetweennessCentrality are proper whoppers. For a network of 65 000 nodes, we're talking about 2+ hour calculation times for the closeness centrality, not to mention the betweenness.

Apparently switching to #igraph would provide a speed boost over #networkx but the convenience of #osmnx has won me over.

Calculating different #centrality measures for a street #network takes longer than expected.

#DegreeCentrality is calculated in a few milliseconds. But oh boy. #ClosenessCentrality and #BetweennessCentrality are proper whoppers. For a network of 65 000 nodes, we're talking about 2+ hour calculation times for the closeness centrality, not to mention the betweenness.

Apparently switching to #igraph would provide a speed boost over #networkx but the convenience of #osmnx has won me over.

Calculating different #centrality measures for a street #network takes longer than expected.

#DegreeCentrality is calculated in a few milliseconds. But oh boy. #ClosenessCentrality and #BetweennessCentrality are proper whoppers. For a network of 65 000 nodes, we're talking about 2+ hour calculation times for the closeness centrality, not to mention the betweenness.

Apparently switching to #igraph would provide a speed boost over #networkx but the convenience of #osmnx has won me over.

Calculating different #centrality measures for a street #network takes longer than expected.

#DegreeCentrality is calculated in a few milliseconds. But oh boy. #ClosenessCentrality and #BetweennessCentrality are proper whoppers. For a network of 65 000 nodes, we're talking about 2+ hour calculation times for the closeness centrality, not to mention the betweenness.

Apparently switching to #igraph would provide a speed boost over #networkx but the convenience of #osmnx has won me over.

@Telias @Downes @cogdog The point there being: maybe there’s a nonlinear function between the number of people involved and the visibility of the group. Classic measures from #SocialNetworkAnalysis might help, including #ClusteringCoefficient and #BetweennessCentrality.
In other words… Is there a #SocialButterflyEffect in #LearningNetworks?

@Telias @Downes @cogdog The point there being: maybe there’s a nonlinear function between the number of people involved and the visibility of the group. Classic measures from #SocialNetworkAnalysis might help, including #ClusteringCoefficient and #BetweennessCentrality.
In other words… Is there a #SocialButterflyEffect in #LearningNetworks?

@Telias @Downes @cogdog The point there being: maybe there’s a nonlinear function between the number of people involved and the visibility of the group. Classic measures from #SocialNetworkAnalysis might help, including #ClusteringCoefficient and #BetweennessCentrality.
In other words… Is there a #SocialButterflyEffect in #LearningNetworks?