Artificial Intelligence Stack Exchange is a question and answer site for people interested in conceptual questions about life and challenges in a world where "cognitive" functions can be mimicked in purely digital environment. It only takes a minute to sign up.
Sign up to join this community
I have a simple neural network for binary classification.
The input features include age, sex, economic situation, illness, disability, etc. The output is simply 1 and 0.
I would like to order the features from the greatest to least impact it had on the classification.
An example answer could look like this:
Classification: 1
Another example:
Classification: 0
Two popular methods I’ve seen done:
1) For each feature, remove it and run the model and see the impact it has on the result. The idea is that the larger the impact, the more pertinent it was to the result.
2) Look at the gradients magnitude $|\nabla_f {y} |$. You can either look at the raw gradient or look at the guided back-propagation which is just the back props product rule, but you only look at when the nodes positively help trigger a neuron by taking only the positive gradients at each step.
There’s probably also more methods. Hope this helped.
