Tag: class imbalance

A leakage-free benchmark of ten chronic-disease cohorts finds that no outlier-removal method beats doing nothing, while data-cleaning steps silently delete ...

Researchers in China have developed CADEE, an online ensemble learning framework that simultaneously handles multi-class imbalance, concept drift, and limited ...

Researchers in Hyderabad have built a deep learning pipeline that predicts fault-prone software modules with 99.3 percent accuracy in under ...

Researchers report the first trained stance-detection models and benchmark for Sorani Kurdish, using data augmentation and QLoRA fine-tuning of KuBERT ...

A new deep learning model called CWT-AuxNet converts wavelet-based time-frequency representations of uterine electrical recordings into individualized preterm birth risk ...

A rigorous evaluation of 126 machine learning pipelines on data from more than 3,000 Golden Retrievers shows that routine laboratory ...

Researchers at Manipur University have developed EF²RB, an ensemble feature selection framework that boosts IoT intrusion detection accuracy to between ...

A Kazakhstani research team shows that physically structured features and calibrated baselines, not model architecture, drive a tenfold improvement in ...

Researchers have developed a geometry-aware oversampling method that uses neural density fields and meta-learning to generate synthetic minority-class samples adapted ...

A new baseline study shows that deep learning can learn meaningful microstructural information from severely imbalanced ultra-high carbon steel images, ...