Algorithm | Best Parameters | Avg Precision | Avg Recall | Avg F1 | Precision Score |
---|---|---|---|---|---|
SVC | {'C': 100, 'gamma': 0.001, 'kernel': 'rbf'} | 0.96 | 0.96 | 0.96 | 0.9701492537313433 |
DecisionTreeClassifier | {'max_depth': 100, 'max_features': 'log2', 'min_samples_leaf': 5, 'min_samples_split': 10} | 0.94 | 0.94 | 0.94 | 0.9411764705882353 |
MLPClassifier | {'activation': 'tanh', 'alpha': 0.0001, 'hidden_layer_sizes': (10,), 'max_iter': 200} | 0.95 | 0.95 | 0.95 | 0.9552238805970149 |
GaussianNB | {} | 0.90 | 0.90 | 0.90 | 0.9242424242424242 |
LogisticRegression | {'fit_intercept': True, 'max_iter': 10, 'penalty': 'l1', 'tol': 0.0001} | 0.96 | 0.96 | 0.96 | 0.9558823529411765 |
KNeighborsClassifier | {'algorithm': 'ball_tree', 'n_neighbors': 10, 'p': 1, 'weights': 'uniform'} | 0.96 | 0.96 | 0.96 | 0.9558823529411765 |
BaggingClassifier | {'max_features': 0.5, 'max_samples': 1.0, 'n_estimators': 20, 'random_state': None} | 0.97 | 0.97 | 0.97 | 0.9705882352941176 |
RandomForestClassifier | {'criterion': 'entropy', 'max_depth': 200, 'max_features': 0.5, 'n_estimators': 20} | 0.96 | 0.96 | 0.96 | 0.9701492537313433 |
AdaBoostClassifier | {'algorithm': 'SAMME', 'learning_rate': 0.8, 'n_estimators': 200, 'random_state': None} | 0.98 | 0.98 | 0.98 | 0.9850746268656716 |
GradientBoostingClassifier | {'loss': 'deviance', 'max_depth': 3, 'max_features': 'log2', 'n_estimators': 200} | 0.97 | 0.97 | 0.97 | 0.9705882352941176 |
XGBClassifier | {'booster': 'gbtree', 'learning_rate': 0.1, 'max_delta_step': 0, 'min_child_weight': 1} | 0.97 | 0.97 | 0.97 | 0.9705882352941176 |
The best algorithm among these was the AdaBoost classifier.