Skip to main content
. 2020 Aug 14;21(16):5847. doi: 10.3390/ijms21165847

Table 1.

Classification report containing precision, recall and F1 score for DNN and baselines on test data. 0–3 correspond to the four classes “non-responsive”, “antagonist”, “agonist”, and ”mixed-response”. Avg = weighted average for each metric according to number of samples for each class. MCC = Matthew’s correlation coefficient.

DNN RF SVM
P R F1 P R F1 P R F1
0 1.0 0.86 0.92 0.83 0.71 0.77 1.0 0.57 0.73
1 0.9 0.88 0.89 0.93 0.87 0.90 0.70 0.72 0.71
2 0.5 0.25 0.33 0 0 0 0 0 0
3 0.56 0.71 0.62 0.5 0.71 0.59 0.32 0.43 0.35
Avg 0.80 0.79 0.79 0.75 0.75 0.75 0.62 0.61 0.6
MCC 0.654 0.601 0.336