Cached Neighborhood Evidence Gives kNN Classifiers a Fairer, Faster Way to Handle Rare Classes
A new classifier called CABLE-kNN caches local neighborhood evidence during training and neutralizes class priors, beating ordinary kNN on imbalanced ...
A new classifier called CABLE-kNN caches local neighborhood evidence during training and neutralizes class priors, beating ordinary kNN on imbalanced ...
Researchers at Yanshan University have developed ADPOD, an outlier detection algorithm that improves density-peak clustering with an affinity-based density measure, ...
Brazilian researchers have combined vibration analysis with explainable artificial intelligence to diagnose unbalance, misalignment, and cracks in rotors supported by ...
Researchers at Soongsil University have developed a self-supervised, multi-view defense that detects label-flipping poisoning attacks in tabular machine learning by ...
Researchers in Spain have unveiled a data reduction method combining self-organizing maps and permutation importance that shrinks banking datasets by ...
A new dual-masked graph autoencoder called DM-GAE maps mineral prospectivity in Tibet with high accuracy despite scarce labeled deposits, outperforming ...
Researchers at Yantai University have developed SSAO, an adaptive oversampling method that uses Gaussian mixture model clustering and minimum enclosing ...
© 2026 Scienmag - Science Magazine
© 2026 Scienmag - Science Magazine