Results 161 to 170 of about 50,437 (269)

Machine Learning‐Based Correction of Reanalysis Surface Radiation Fluxes Using Ground‐Measured Data in the Southern Brazilian Pampa Biome

International Journal of Climatology, EarlyView.
This study demonstrates that ERA5 provides more accurate surface radiation flux estimates than MERRA2 across the Southern Brazilian Pampa. Machine learning models, particularly Random Forest, further improved the precision of reanalysis data for climate applications.
Olusola Samuel Ojo, Michel Baptistella Stefanello, Alecsander Mergen, Maria Eduarda Oliveira Pinheiro, Vanessa de Arruda Souza, Débora Regina Roberti   +5 more
wiley   +1 more source

Generalizable framework for multi-site bone density prediction using non-dominant wrist optical biomarkers. [PDF]

Biomed Opt Express
Chang WC, Meitei TG, Wang YM, Chen LK, Sun CW.   +4 more
europepmc   +1 more source

High‐Fidelity Synthetic Raman Spectra Generation for Sinter Basicity Prediction Using β$$ \beta $$‐Variational Autoencoders

Journal of Raman Spectroscopy, EarlyView.
A β$$ \beta $$‐variational autoencoder with β$$ \beta $$ = 0.1 generates high‐fidelity synthetic Raman spectra of industrial sinter with a 16‐fold improvement in spectral fidelity over SMOTE‐based augmentation (KL divergence: 0.0075 vs. 0.121), enabling reliable basicity prediction (R2 = 0.83) from limited labeled datasets.
Marjorie Ariele Pereira, Daniel Cruz Cavalieri, Adilson Ribeiro Prado, Cassius Zanetti Resende, Erica Simões Rodrigues   +4 more
wiley   +1 more source

AQ-MultiCal: An Interactive No-Code Machine Learning Platform for Low-Cost Air Quality Sensor Calibration and Comparative Model Analysis. [PDF]

Sensors (Basel)
Taştan M, Karsu Asal EC, Gökozan H.
europepmc   +1 more source

AI/ML Enabled High‐Throughput Design and Synthesis for Energetic Molecules

Materials Genome Engineering Advances, EarlyView.
Novel design methods for a special kind of functional molecules — the energetic molecules—are summarized. Both classic Machine Learning (ML) and Artificial Intelligence (AI) generative models are utilized for the high‐throughput design of these energetic molecules, and a sort of high‐energy low‐sensitivity molecules are obtained.
Wen Qian
wiley   +1 more source

An interpretable machine learning model for predicting symptomatic pelvic lymphocele after pelvic lymphadenectomy in cervical cancer. [PDF]

Front Oncol
Wang Y, Tang N, Zhang W, Zeng Y.
europepmc   +1 more source

Physics‐Informed Generative Machine Learning for Designing Crack‐Free γ′‐Strengthened Ni‐Based Superalloys for Laser Powder Bed Fusion

Materials Genome Engineering Advances, EarlyView.
This research proposes a physics‐informed generative machine learning framework to design SHA800, a crack‐free γ′‐strengthened nickel‐based superalloy for laser powder bed fusion, achieving a 43% γ′ volume fraction and 587 HV0.2 hardness. ABSTRACT Fabricating γ′‐strengthened nickel‐based superalloys via laser powder bed fusion (LPBF) faces significant ...
Kai Guo, Dongwei Guan, Tao Hu, Yaqing Hou, Xiaoqun Li, Zhongnan Bi, Songzhe Xu, Hao Yu, Wei Xu, Weidong Xuan, Chaoyue Chen, Zhongming Ren   +11 more
wiley   +1 more source

Automatic Infant Movement Assessment Using Pose-LBP Features and a Cost-Sensitive Subspace kNN Ensemble. [PDF]

Bioengineering (Basel)
Ari A, Atalan Efkere P, Çangur EY, Uzun Akkaya K, Gurler Ari B, Elbasan B, Sengur A, Tian Y.   +7 more
europepmc   +1 more source

Symbolic Regression‐Guided Feature Engineering for Predicting Magnetization in Cu‐Based Alloys Under Data‐Scarce Conditions

Materials Genome Engineering Advances, EarlyView.
A symbolic regression approach (SISSO) with physics‐informed feature engineering achieves high‐accuracy prediction of magnetic properties in Cu‐based alloys under data‐scarce conditions. The framework offers an interpretable and transferable strategy for accelerated alloy design.
Buyang Ma, Jiejie Li, Qiuju Zhang, Fenghua Chen, Bo Li, Ziqi Tian, Liang Chen   +6 more
wiley   +1 more source