Table 2 Performance evaluation of selected algorithms
Algorithm | Hyper-parameters | Sensitivity | Specificity | Accuracy | F-score |
|---|---|---|---|---|---|
RF | Maximum of depth = “6” Maximum number of iterations = “50” Number of execution slots = “1” Bag size percent = “100” Break Tie randomly = “True” | 0.95 ± 0.01 | 0.94 ± 0.01 | 0.94 ± 0.005 | 0.94 ± 0.003 |
Bagging | Bag size percent = “100” Classifier = “REP-Tree” Maximum number of iterations = “15” Number of decimal places = “2” Number of execution slots = “1” | 0.84 ± 0.02 | 0.84 ± 0.01 | 0.84 ± 0.01 | 0.84 ± 0.02 |
AdaBoost | Batch size = “100” Classifier = “Decision Stump” Maximum number of iterations = “20” Weight threshold = “50” Minimum number of instances per leafs = “1” | 0.88 ± 0.01 | 0.86 ± 0.02 | 0.88 ± 0.01 | 0.86 ± 0.01 |
XG-Boost | Maximum depth = “8” Classifier = “Decision Stump” Base score = “4” Min child weight = “1” Booster = “gb-tree” | 0.90 ± 0.01 | 0.88 ± 0.02 | 0.89 ± 0.015 | 0.88 ± 0.01 |
MLP | Number of hidden layers “10” Learning rate = “0.25” Momentum = “0.2” Validation threshold = “20” Maximum number of iterations = “20” Normalize numeric values and attributes = “True” | 0.77 ± 0.03 | 0.75 ± 0.04 | 0.76 ± 0.035 | 0.76 ± 0.035 |
SVM | Kernel type = “RBF” Regularization parameters (C) = “10” Gamma = “10” RBF gamma = “0.1” Degree for increasing dimensions = “3” | 0.80 ± 0.03 | 0.79 ± 0.03 | 0.79 ± 0.03 | 0.79 ± 0.03 |
J-48 | Confidence factor = “0.2” Minimum number of objects = “2” Number of folds = “3” Binary split = “false” Reduced error pruning = ” True” | 0.72 ± 0.04 | 0.72 ± 0.04 | 0.70 ± 0.04 | 0.72 ± 0.04 |
NB | Use Kernel Classifier = “true” Use Supervise discretization = “true” Batch size = “100” Number of decimal places = “100” | 0.68 ± 0.04 | 0.65 ± 0.05 | 0.69 ± 0.045 | 0.66 ± 0.04 |