Homec4science
Diffusion ISFC clustering (master) Browse

ISFCClustering/2019_03_03_BCT/data_and_demos/demo_efficiency_measures.m
3209cc2b0b1dmaster

Last Change
Raw File

demo_efficiency_measures.m
View Options

% This script loads 7 unweighted, undirected adjacency matrices
% corresponding to a clique, chain, ring, 1D lattice, star, rich-club, and
% bi-modular toy networks (all graphs have 50 nodes). The following
% efficiency measures are computed for each graph:
%
% - prob_SPL: probability of one particle traveling through shortest-paths
% - Erout: efficiency of routing -> based on shortest-paths
% - Ediff: efficiency of diffusion -> based on mean-first-passage-times
% - Eres: efficiency of resources -> based on number of particles
% necessary so that at least one particle taking shortest-paths with
% certain probability (lambda).
%
% If you are using this efficiency package for your research, plase kindly
% cite the paper:
%
% "Exploring the Morphospace of Communication Efficiency in Complex
% Networks" Goñi J, Avena-Koenigsberger A, Velez de Mendizabal N, van den
% Heuvel M, Betzel RF and Sporns O. PLoS ONE. 2013
%
% These examples and results correspond to Table 1 in the paper.
%
% Joaquin Goñi and Andrea Avena-Koenigsberger, IU Bloomington, 2012
close all;
clear all;
clc;
load demo_efficiency_measures_data.mat; % 7 adjacency matrices corresponding to the examples shown in Table 1 are loaded.
lambda = 0.5; % this parameter is an input for the computation of Eres.
% run and display efficiency measures for the 7 graphs
disp([' prob_SPL ',' Erout ',' Ediff ',' Eres '])
fprintf('----- clique ----- \n')
adj = clique;
N = size(adj,1);
EYE = logical(eye(N,N));
SPL = distance_wei_floyd(adj);
Erout = rout_efficiency(adj);
Ediff = diffusion_efficiency(adj);
[Eres,prob_SPL] = resource_efficiency_bin(adj,lambda,SPL);
prob_SPL = mean(prob_SPL(~EYE));
Eres = mean(Eres(~EYE));
disp([prob_SPL,Erout,Ediff,Eres])
fprintf('----- chain ----- \n')
adj = chain;
N = size(adj,1);
EYE = logical(eye(N,N));
SPL = distance_wei_floyd(adj);
Erout = rout_efficiency(adj);
Ediff = diffusion_efficiency(adj);
[Eres,prob_SPL] = resource_efficiency_bin(adj,lambda,SPL);
prob_SPL = mean(prob_SPL(~EYE));
Eres = mean(Eres(~EYE));
disp([prob_SPL,Erout,Ediff,Eres])
fprintf('----- ring ----- \n')
adj = ring;
N = size(adj,1);
EYE = logical(eye(N,N));
SPL = distance_wei_floyd(adj);
Erout = rout_efficiency(adj);
Ediff = diffusion_efficiency(adj);
[Eres,prob_SPL] = resource_efficiency_bin(adj,lambda,SPL);
prob_SPL = mean(prob_SPL(~EYE));
Eres = mean(Eres(~EYE));
disp([prob_SPL,Erout,Ediff,Eres])
fprintf('----- lattice1D ----- \n')
adj = lattice1D;
N = size(adj,1);
EYE = logical(eye(N,N));
SPL = distance_wei_floyd(adj);
Erout = rout_efficiency(adj);
Ediff = diffusion_efficiency(adj);
[Eres,prob_SPL] = resource_efficiency_bin(adj,lambda,SPL);
prob_SPL = mean(prob_SPL(~EYE));
Eres = mean(Eres(~EYE));
disp([prob_SPL,Erout,Ediff,Eres])
fprintf('----- star ----- \n')
adj = star;
N = size(adj,1);
EYE = logical(eye(N,N));
SPL = distance_wei_floyd(adj);
Erout = rout_efficiency(adj);
Ediff = diffusion_efficiency(adj);
[Eres,prob_SPL] = resource_efficiency_bin(adj,lambda,SPL);
prob_SPL = mean(prob_SPL(~EYE));
Eres = mean(Eres(~EYE));
disp([prob_SPL,Erout,Ediff,Eres])
fprintf('----- rich-club ----- \n')
adj = rich_club;
N = size(adj,1);
EYE = logical(eye(N,N));
SPL = distance_wei_floyd(adj);
Erout = rout_efficiency(adj);
Ediff = diffusion_efficiency(adj);
[Eres,prob_SPL] = resource_efficiency_bin(adj,lambda,SPL);
prob_SPL = mean(prob_SPL(~EYE));
Eres = mean(Eres(~EYE));
disp([prob_SPL,Erout,Ediff,Eres])
fprintf('----- bi-modular ----- \n')
adj = bi_modular;
N = size(adj,1);
EYE = logical(eye(N,N));
SPL = distance_wei_floyd(adj);
Erout = rout_efficiency(adj);
Ediff = diffusion_efficiency(adj);
[Eres,prob_SPL] = resource_efficiency_bin(adj,lambda,SPL);
prob_SPL = mean(prob_SPL(~EYE));
Eres = mean(Eres(~EYE));
disp([prob_SPL,Erout,Ediff,Eres])
c4science · Help