Tag: class imbalance

A new integrated framework combining quantum-inspired optimization and deep learning achieves high accuracy in detecting diverse cyberattacks across major cloud ...

Researchers developed a streamlined Pi–Sigma neural network paired with correlation-based feature filtering that detected 98 percent of software defects in ...

A systematic evaluation reveals that EfficientNet-B0 achieves 91.30 percent average accuracy on real landfill waste imagery, though explainability analyses expose ...

Researchers have introduced ADAPT ClickNet, an end-to-end graph learning framework that significantly improves rare mobile install attribution and probability calibration ...

Researchers developed LogISEL, an interpretable stacked machine learning framework that forecasts slope movement classes ten minutes ahead across diverse Himalayan ...

Researchers developed a long short-term memory intrusion detection framework that achieved 92 percent accuracy on network traffic after utilizing generative ...

Researchers developed EARS-Net, a hybrid deep learning model combining generative oversampling, dual attention, and transformers that achieves superior facial expression ...

Researchers evaluating a multi-branch deep learning model on a 500-student physical education dataset found that the architecture failed to outperform ...

A cross-modal framework coupling multi-prompt language descriptions with class-aware contrastive learning improves minority bankruptcy detection across Polish and American corporate ...

A partitioned feature selection framework improves the interpretability and predictive performance of questionnaire-based depression severity models by grouping interrelated symptoms ...