[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
## Re: [igraph] betweenness centrality on weighted networks

**From**: |
Szabolcs Horvát |

**Subject**: |
Re: [igraph] betweenness centrality on weighted networks |

**Date**: |
Fri, 6 Apr 2018 19:37:06 +0200 |

Hello Joice,
I do not think the terminology you are using (cost, strength) is
standard, and it is not clear what you mean.
The concept of betweenness centrality is based on the idea of
(shortest) path length. In the unweighted case, the length of a path
is simply the number of edges it consists of. In the weighted case,
it is the sum of the weights of these edges. This is precisely how
igraph uses edge weights for betweenness calculations.
Betweenness is, roughly speaking, the number of shortest paths passing
through a vertex (or edge). High edge weights make paths longer, thus
high weight edges are less likely to appear in the paths that are the
shortest.
This has hopefully answered your question.
On 6 April 2018 at 18:06, joice de faria poloni <address@hidden> wrote:
>* Dear igraph users,*
>
>* I'm working with gene correlation networks, where each vertex represents a*
>* gene and each edge represents an expression correlation among genes.*
>* In this sense, I was planning to use the igraph package to calculate network*
>* centralities, specifically betweenness and edge betweenness.*
>* However, I have a question: When I use the command "edge_betweenness" or*
>* "betweenness", the weight of the edges will be considered as costs or*
>* strengths? If they are considered as costs, can I use a command to calculate*
>* betweenness as a strength?*
>
>* Thank you for your assistance regarding in this matter.*
>
>
>* _______________________________________________*
>* igraph-help mailing list*
>* address@hidden*
>* https://lists.nongnu.org/mailman/listinfo/igraph-help*
>