The network to analyze must be undirected, otherwise there are no specia= l constraints.

=20Basic analysis tool, not particular for special disciplines or problems.=

=20The algorithm requires two inputs, the file where the edges of the netwo=
rk are listed and the number of points for the binned distribution describe=
d below. A first read-in of the inputfile will set the values of the number=
of nodes and edges of the network. In the second read-in the degree of all=
nodes is calculated and the edges are stored in an array. Then the cluster=
ing coefficients for all nodes are calculated and listed in one of the thre=
e output files, together with the corresponding node index. The average of =
the clustering coefficients is determined and displayed in the NWB console.=

Next the distribution of the clustering coefficients is calculated. Th=
e program generates two other output files, corresponding to two different =
ways of partitioning the interval spanned by the values of the clustering c=
oefficient. In the first output, one divides the range of variation of the =
clustering coefficient in equal bins and determines the number of nodes wit=
h clustering coefficients inside each bin: the scores are then divided by t=
he number of nodes of the network, so to obtain the probability: the output=
displays such probability values together with the center points of the bi=
ns they refer to.

The second output gives the *binned* distribut=
ion, i.e. the interval spanned by the values of the clustering coefficient =
is divided into bins whose size grows while going to higher values of the v=
ariable. The size of each bin is obtained by multiplying by a fixed number =
the size of the previous bin. The program calculates the fraction of nodes =
whose clustering coefficient falls within each bin. Because of the differen=
t sizes of the bins, these fractions must be divided by the respective bin =
size, to have meaningful averages.

This technique is particularly suita=
ble to study skewed distributions: the fact that the size of the bins grows=
large for large values of the variable compensates for the fact that not m=
any nodes have high values, so it suppresses the fluctuations that one woul=
d observe by using bins of equal size. On a double logarithmic scale, which=
is very useful to determine the possible power law behavior of the distrib=
ution, the points of the latter will appear equally spaced on the x-axis.** The algorithm runs in a time , where is the number of nodes of the netwo=
rk and is the average degree squared.**

A simple application of this algorithm could be to calculate the cluster= ing coefficient and its distribution for networks created by the modeling a= lgorithms of the NWB. For instance, the inputfile can be created through th= e Barabasi-Albert model.

=20- =20
- Source Code =20

The algorithm was implemented and documented by S. Fortunato, integrated= by S. Fortunato and W. Huang. For the description we acknowledge Wikipedia= .

=20Watts, D.J., Strogatz, S.H.(1998) Collective d= ynamics of 'small-world' networks. Nature 393:440-442.

=20
The license could not be verified: License Certificate has expired!=20
Generate a Free license now.