Table 5 Proposed Model with Different Training-validation-testing Split
From: A multimodal graph neural network framework for cancer molecular subtype classification
Model | Training set ratio | |||||
|---|---|---|---|---|---|---|
70% | 60% | 50% | ||||
Accu.a | F1 | Accu.a | F1 | Accu.a | F1 | |
Proposed w/ GAT | \(82.5\% \pm 1.5\%\) | \(0.82 \pm 0.02\) | \(79.9\% \pm 4.0\%\) | \(0.78 \pm 0.06\) | \(74.2\% \pm 7.5\%\) | \(0.71 \pm 0.10\) |
Proposed w/ GCN | \(77.9\% \pm 1.2\%\) | \(0.76 \pm 0.02\) | \(76.7\% \pm 0.4\%\) | \(0.75 \pm 0.01\) | \(77.3\% \pm 2.5\%\) | \(0.76 \pm 0.03\) |