Computing average path length

I'm using karate.gml dataset, from two visualization software gephi and SocNetV both gives me an average path length of 2.408. I want to compute the average path length in matlab using the same dataset, but mine gives me 1.76. My matlab procedure is below:
I add the distance {d} returned by the shortestpath function {[P,d] = shortestpath(G,c,nodesize);} to an array I've declared and afterwards find the mean. Not sure what I'm doing wrong.
size = nodesize;
x = [];
for c = 1:size
[P,d] = shortestpath(G,c,nodesize);
x = [x, d];
end
answer = mean(x)

1 Commento

I would tend to think that average path length would include alternative paths, not just the shortest path.

Accedi per commentare.

 Risposta accettata

Isaac Osei Agyemang
Isaac Osei Agyemang il 20 Set 2018
Modificato: Walter Roberson il 20 Set 2018
size = nodesize;
x = [];
for c = 1:size
for dest=1:size
if(c~=dest)
[P,d] = shortestpath(G,c,dest);
x = [x, d];
end
end
answer = mean(x)

Più risposte (1)

KSSV
KSSV il 20 Set 2018
Modificato: KSSV il 20 Set 2018
nodesize = numnodes(G);
thesize = nodesize;
x = zeros(thesize,1);
for c = 1:size
[P,d] = shortestpath(G,c,nodesize);
x(C) = d ;
end
answer = mean(x)

7 Commenti

@KSSV, Please it still gave me 1.7647
KSSV
KSSV il 20 Set 2018
What is value of thesize?According to the above code...x should be a vector. Show us your complete code.
nodesize = numnodes(G);
size = nodesize;
x = [];
for c = 1:size
[P,d] = shortestpath(G,c,nodesize);
x = [x, d];
end
answer = mean(x)
KSSV
KSSV il 20 Set 2018
YOu try the code, whic I gave. x will be an vector and answer will be a number as it is mean of x.
I've tried your code, it still gives me 1.76 which the other two visualization software gephi and SocNetV gives 2.4
KSSV
KSSV il 20 Set 2018
YOu mean mean(x)?
I've been able to work things out, thanks for your help.
size = nodesize;
x = [];
for c = 1:size
for dest=1:size
if(c~=dest)
[P,d] = shortestpath(G,c,dest);
x = [x, d];
end
end
answer = mean(x)

Accedi per commentare.

Categorie

Tag

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by