centralityInformation() is slow when network N>100

[Dimitris Kalamaras]

the calculation of IC is slow when the graph order is large (say N > 100) because the centralityInformation() method still uses the Naive Gauss elimination algorithm to find the inverse of the pseudo-adjacency B
B is a matrix derived from 1-A where the diagonal is substituted with 1 plus each actor's degree.

Solution: We need to use LU-decomposition instead of Gauss elimination.