Fig. 3From: Identification of parameters and formulation of a statistical and machine learning model to identify Babesia canis infections in dogs using available ADVIA hematology analyzer dataDecision tree classifier. The top line shows the condition for descending the tree. Blue leaves imply the model predicts a positive B. canis infection, whereas orange leaves predict no infection. Samples refers to the total number of samples from the training set that end up in a particular leaf. Values are the weighted samples in a leaf, where the first entry corresponds to the negative samples (which have a weight of ~ 0.54) and the second entry to the positive samples (with a weight of ~ 7.07). Whichever value is largest determines the leaf label. Note that the complete right branch only contains one positive sample in the train set. As such, the parameter abs_lymphs(× 10E9 cells/l) is plausibly of lesser importance, despite it being high up in the treeBack to article page