Datasets and descriptors used for assessing the similarity between candidate features and the original ones

Datasets and descriptors that were used for the assessment of the first stage during a feature selection process. At this stage, a reduced subset of candidate features is selected by removing irrelevant and redundant features (using either Pearson or Spearman correlation coefficient). Moreover, we included the first 50 principal components of both the reduced and original features that were useful to assess the similarity between both descriptor spaces.<br>