Random Forest

Ensemble learning is a machine learning technique where multiple models, often called weak learners or base models, are combined to produce a stronger, more accurate model. The primary aim of ensemble learning is to reduce overfitting, improve generalization, and achieve better predictive performance compared to individual models.

Random Forest is a bagging-based ensemble method that uses Decision Trees as base learners. It introduces random feature selection to make the decision trees more diverse. Random Forest Worked Better, because it has ability to handle imbalanced data

Conlclusion

Random Forest's accuracy of 87.0% is a significant improvement over the Linear SVM (~82.4%). This shows that Random Forest is better at handling the class imbalance and the feature relationships. The confusion matrix highlights that B1 (minority class) still faces some misclassifications. But Random Forest reduces misclassifications compared to SVM but struggles slightly due to the underrepresentation of B1.