The absolute values of the calculated feature importance scores were normalized, i.e. scaled to unit norm, in order to provide comparable feature rating across models: For each sample (each of the 50 shuffles) the calculated importance scores were rescaled to be confined within the range [0,1] with their sum equal to one. Then, for each feature the mean and standard deviation over the samples (shuffles) were calculated and reported as the final rating measures.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.