Results 171 to 180 of about 6,034 (237)

Rebalancing Software Defect Datasets via Mutation: Performance Insights From Prediction Models Based on Software Measures

Software Testing, Verification and Reliability, Volume 36, Issue 5, August 2026.
A mutation‐based approach (MBA) to rebalance defect datasets improves recall, particularly in cross‐project prediction, but increases false alarms and does not consistently enhance MCC or AUC. These findings highlight both the potential and limitations of mutation‐based rebalancing in software defect prediction.
Dinçer Güner, Görkem Giray, Onur Demirörs   +2 more
wiley   +1 more source

Closed-Set Heterogeneous Domain Adaptation for IoT Intrusion Detection: An Anchor-Based Benchmark Across Single- and Multi-Source Transfer. [PDF]

Sensors (Basel)
Chizari M, Khan Ali Mirza Q, Alam A, Chizari H.   +3 more
europepmc   +1 more source

PETIMOT: a novel framework for inferring protein motions from sparse data using SE(3)‐equivariant graph neural networks

Acta Crystallographica Section D, Volume 82, Issue 8, Page 862-885, August 2026.
We present a new formulation for protein flexibility and learn protein motions from sparse experimental data.Proteins move and deform to ensure their biological functions. Despite significant progress in protein structure prediction, approximating conformational ensembles under physiological conditions remains a fundamental open problem.
Valentin Lombard, Julien Nguyen Van, Sergei Grudinin, Elodie Laine   +3 more
wiley   +1 more source

Revocable and Traceable Decentralized ABE for P2P Networks. [PDF]

Entropy (Basel)
Gao D, Xu H, Qian S.
europepmc   +1 more source

Are we there yet? Reliable occupancy modeling from AI‐labeled trail camera data

Conservation Science and Practice, Volume 8, Issue 8, August 2026.
Even low false‐positive rates in AI‐labeled camera‐trap data can bias occupancy estimates, particularly for rare species. Our results show that false positive informed occupancy models can substantially improve inference, underscoring the need to evaluate AI classifiers according to the assumptions of downstream ecological analyses.
Mohamed Khalil Meliane, Joseph M. Guthrie, E. Hance Ellington   +2 more
wiley   +1 more source

Hybrid Multimodal Surrogate Modeling and Uncertainty-Aware Co-Design for L-PBF Ti-6Al-4V with Nanomaterials-Informed Morphology Proxies. [PDF]

Nanomaterials (Basel)
Hossain RB, Pan X, Chang G, Su X, Tao Y, Han X.   +5 more
europepmc   +1 more source

Miners' Reward Elasticity and Stability of Competing Proof‐of‐Work Cryptocurrencies

International Economic Review, Volume 67, Issue 3, Page 957-976, August 2026.
ABSTRACT Proof‐of‐Work cryptocurrencies employ miners to sustain the system through algorithmic reward adjustments. We develop a stochastic model of the multicurrency mining and identify conditions for stable transaction speeds. Bitcoin's algorithm requires hash supply elasticity <$<$1 for stability, while ASERT remains stable for any elasticity and ...
Kohei Kawaguchi, Junpei Komiyama, Shunya Noda   +2 more
wiley   +1 more source

Data-based clustering in prediction of cervical cancer DNA methylation using pan-cancer genetic and clinical data. [PDF]

Bioinform Adv
Pai N, Rao JS.
europepmc   +1 more source

Medical Knowledge Integration Into Reinforcement Learning Algorithms for Dynamic Treatment Regimes

International Statistical Review, Volume 94, Issue 2, Page 382-411, August 2026.
Summary The goal of precision medicine is to provide individualised treatment at each stage of chronic diseases, a concept formalised by dynamic treatment regimes (DTR). These regimes adapt treatment strategies based on decision rules learned from clinical data to enhance therapeutic effectiveness.
Sophia Yazzourh, Nicolas Savy, Philippe Saint‐Pierre, Michael R. Kosorok   +3 more
wiley   +1 more source

BAAR: A framework for blockchain-based anonymous and revocable user authentication scheme. [PDF]

PLoS One
Ahmed M, Ahmad A, Zeshan F, Akram S.
europepmc   +1 more source