Table 3 Evaluating classical ML models’ accuracy using different training/testing methods without feature selection
From: Classification of lung cancer severity using gene expression data based on deep learning
|  | Training – testing method | SVM | RF | KNN | Adaboost |
|---|---|---|---|---|---|
LUAD | 10-fold cross-validation | 0.57 | 0.54 | 0.58 | 0.54 |
5-fold cross-validation | 0.57 | 0.57 | 0.58 | 0.56 | |
80–20 splitting | 0.61 | 0.58 | 0.57 | 0.60 | |
70–30 splitting | 0.63 | 0.60 | 0.59 | 0.56 | |
LUSC | 10-fold cross-validation | 0.56 | 0.53 | 0.50 | 0.52 |
5-fold cross-validation | 0.55 | 0.52 | 0.51 | 0.55 | |
80–20 splitting | 0.61 | 0.59 | 0.59 | 0.57 | |
70–30 splitting | 0.58 | 0.57 | 0.54 | 0.50 |