tnet (version 3.0.16)

clustering_local_w: Barrat et al. (2004) generalised local clusering coefficient

Description

This function calculates Barrat et al. (2004) generalised local clusering coefficient. See http://toreopsahl.com/2009/01/23/weighted-local-clustering-coefficient/ for a detailed description. By default it defines the triplet value as the average of the two tie weights; however it can also define it differently. See the blog post. Note: If there are very large tie weights in a network, the geometric method in R fails. However, this can be fixed by transforming the values. net[,"w"] <- (net[,"w"]/min(net[,"w"])) This step is not required unless you receive warnings when running the function.

Usage

clustering_local_w(net, measure = "am")

Arguments

net

A weighted edgelist

measure

The measure-switch control the method used to calculate the value of the triplets. am implies the arithmetic mean method (default) gm implies the geometric mean method mi implies the minimum method ma implies the maximum method bi implies the binary measures This can be c("am", "gm", "mi", "ma", "bi") to calculate all.

Value

Returns a data.frame with at least two columns: the first column contains the nodes' ids, and the remaining columns contain the corresponding clustering scores.

References

Barrat, A., Barthelemy, M., Pastor-Satorras, R., Vespignani, A., 2004. The architecture of complex weighted networks. Proceedings of the National Academy of Sciences 101 (11), 3747-3752. arXiv:cond-mat/0311416 http://toreopsahl.com/2009/01/23/weighted-local-clustering-coefficient/

Examples

# NOT RUN {
## Generate a random graph
#density: 300/(100*99)=0.03030303; 
#this should be average from random samples
rg <- rg_w(nodes=100,arcs=300,weights=1:10,directed=FALSE)

## Run clustering function
clustering_local_w(rg)

# }