I get the following result:
How can I use this information, for example, to create a ranking system and weigh the factors (the factors/nodes)?
Weight = Coefficient * 10000 if your normalization method is Rank
1 Like
insights are incredible if credible
This is the top 50 with the highest Coeff. Equal Weight.
A fairly good result, but I thought it would be better:
1 Like
You can try to average of equal weights and the weights implied by the parameters given by the system.
Edit: Weight = (Coefficient + Average Coefficient) * 10000
Thank you. I will post the results when I have run the simulation.
How do you handle the negative part of the "importance"?
Change the direction of the factors/nodes and use the absolute coefficients.
1 Like