Tag: Machine Learning

A new study explains how variational autoencoders integrate multimodal biomedical data and reveals key limitations of in silico perturbation analyses.

University of Johannesburg researchers have built an AI model that uses routine hospital admission data to flag emergency room patients ...

A federated learning framework combining MRI and clinical data classifies Alzheimer's disease with reported accuracy up to 94.1 percent while ...

Researchers propose a modular multi-agent framework, SoS-ML, that generates evidence-based explanations from internal model representations rather than post-hoc justifications.

A lightweight AI pipeline combining wavelet and texture features classified hemorrhagic stroke, ischemic stroke and normal tissue on brain CT ...

A physics-informed generative adversarial network produced 944 physically consistent synthetic rock-strength records from 944 laboratory measurements, preserving key UCS–porosity–density relationships.

A new open-access study reports that a layered fraud detection framework combining verified explanations, probability calibration and cost-sensitive thresholds outperformed ...

Glue-on tri-axial accelerometers and a machine-learning model classified 13 Eurasian otter behaviours with 53.6% accuracy in zoo trials.

A machine learning ensemble trained on 24 wetland sites estimates that Southeastern U.S. wetlands took up a net 20 teragrams ...

Researchers have developed Riemannian and joint optimization methods that reduce bond dimensions in tensor networks by formulating tensor disentangling as ...

A federated learning framework let three oilfields jointly train a gamma-ray prediction model without sharing raw well data, lowering prediction ...

A new review of 119 studies concludes that somatic cell counts in goat milk are inherently elevated and context-dependent, making ...

A computational study combining transcriptomics, machine learning and single-cell analysis identifies CYP1B1 as a potential functional target of kaempferol in ...

A new review describes how monochromated STEM-VEELS has evolved into a quantitative platform for measuring bandgaps at the nanometer scale, ...

A rank-reduction autoencoder trained on only 421 electromagnetic simulations generated five validated optimized waveguide designs for 6G frequencies that a ...

A bibliometric analysis of 857 publications maps three decades of machine learning research on sepsis prediction, revealing explosive growth since ...

Researchers used geologically informed machine learning to reconstruct missing neutron porosity logs from legacy self-potential and resistivity records in a ...

A combined neural network and gray relational analysis framework identified optimal CO₂ laser cutting settings for PMMA, cutting machining time ...

A systematic review of 78 papers examines how machine learning methods support demand forecasting and inventory decision-making.

A new deep learning framework called RCNNricePPI predicts rice protein-protein interactions from amino acid sequences, achieving an accuracy of 0.8996 ...