Hello all, I have a fuzzy cognitive model that is represented as a 27x27 matrix. How can I use openturns to perform a sensitivity analysis of modifying specific individual variables? I have read that openturns can perform a SOBOL analysis, but I have not seen where the function is called in openturns and how I could perform the task… I am a total newb here and need some serious help.
Hello,
You might want to check the sensitivity analysis examples about Sobol:
http://openturns.github.io/openturns/latest/auto_reliability_sensitivity/index.html#sensitivity-analysis
These examples assume fonctions take vectors as inputs, so your matrix might have to be flattened in one input vector of size 729 with ot.PythonFunction.
#Schueller, Thanks very much for the suggestion. I reviewed the examples, but I struggled to understand how they would apply to my case… My matrix refers to specific units of cognition that frame how a group of subject matter experts make decisions. Rows refer to causes, and Columns refer to effects. This relationship is characterized as an “Edge.” There is also a weight (edge weight) that refers to the degree to which an individual row is able to cause an individual column. This relationship can range from -1 → 0 → +1. I would like to simulate what happens when probability functions represent the edge weights. I would also like to be able to modify edge weights within the matrix to understand what happens when specific inputs are made. I thought I could do that easily with Python and NetworkX. However, I am unable to load the matrix…
import NetworkX as nx
adjacency_matrix = [[ ,b,c,d,e,f,g,h,l,j,k,l,m,n,o,p,q,r,s,t,u,v,w,x,y],
[b,0,0.52,0.5,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0.7,0,0,0,0],
[c,0,0,0,0.76,0,0,0,0,0,0.5,0.8,0,0,0,0,0,0,0,0,0,0,0,0,0,0], and so on and so forth
But I get the error:
File “”, line 1
adjacency_matrix = [[ ,b,c,d,e,f,g,h,l,j,k,l,m,n,o,p,q,r,s,t,u,v,w,x,y],
^
SyntaxError: invalid syntax
What am I doing wrong?