Hybrid approaches address such trade-offs by combining oversampling with post-resampling cleaning strategies that remove noisy or duplicate samples. A classic example is SMOTE combined with Edited Nearest Neighbors (SMOTE+ENN), which, after oversampling minority-class samples through synthetic generation techniques, performs neighbourhood cleaning to remove uncertain samples, thereby increasing both class separability and instability \cite{Fernandez18}. Hybrid approaches are typical for biomedical machine learning, as high-throughput data is often noisy, and the value of a feature often depends on other features

Hybrid methods tackle these challenges by merging oversampling techniques with cleaning processes that eliminate noisy or redundant data points after resampling. SMOTE paired with Edited Nearest Neighbors (SMOTE+ENN) serves as a well-known illustration of this approach. This combination first generates synthetic samples to boost minority class representation, then applies neighborhood-based cleaning to eliminate ambiguous data points, which enhances class distinction while reducing model stability. These hybrid strategies are commonly employed in biomedical machine learning applications, where high-throughput datasets frequently contain noise and feature interactions play a crucial role in determining individual feature importance.