Figure 5From: An AUC-based permutation variable importance measure for random forestsDistribution of AUC-values for AUC-based (filled) and error-rate-based (unfilled) permutation VIMs for different class imbalances derived from 100 modified datasets from C-to-U conversion data. The AUC is used to assess the ability of a VIM to discriminate between associated predictors and predictors not associated with the response.Back to article page