Leave-one-out cross-validation results.

<p>Recall, precision, and f-measure are calculated for each class. Weighted f-measure, <i>f_W</i>, and average accuracy, <i>Acc</i>, are calculated for all classes of a problem. All assessments are based on leave-one-out cross-validation on the labeled dataset. Shown in bold are the top-performing classifiers for each problem. RF-SL-2F corresponds to the self-learning RF classifier using 2 additional features.