Evaluation of the importance of each parameter using machine learning

I am a university student studying machines using the free software Orange.
We are currently dealing with the problem of getting six output values for three input values.
I would like to know as a numerical value which parameters contribute to each output value, but I have no idea which evaluation to use and how to consider.

First of all, (i) I think it's a random forest to evaluate the importance (contribution rate?) of each feature. Also, (ii) I don't think I can set the target function to multiple output values in Orange software. How should I proceed with the evaluation? (iii) () in the first place?

If you have any knowledge of the above, please give us your advice.

machine-learning

2022-09-30 21:37

1 Answers

Orange, I didn't know.There seems to be no answer yet, so I will answer as much as I can.

"Problems with 6 output values for 3 input values"
I don't know that
Is it a regression problem that predicts continuous values?
Or is it a "discrimination" issue that classifies 0/1 or more categories?

The random forest has an output of the importance of each feature, but
if the main focus is to see the contribution of the feature. First of all,

For regression
　　Multiple Regression Analysis, Regression Tree Analysis
For determination
　　logistic regression analysis, decision tree analysis

I think it's better to go in first.The contribution of the feature quantity is easy to understand.
These methods are not accurate enough, so if you want to know the contribution while seeking accuracy, you can use random forests.
I think it will become clearer if we compare the contributions of characteristic quantities.
I checked and found that Orange can do this, so please try it.

2022-09-30 21:37

If you have any answers or tips

Popular Tags

python x 4647

android x 1593

java x 1494

javascript x 1427

c x 927

c++ x 878

ruby-on-rails x 696

php x 692

python3 x 685

html x 656