Advanced Intelligent Discovery, EarlyView.Several simulation techniques are used to explore static and dynamic behavior in polyanion sodium cathode materials. The study reveals that universal machine learning interatomic potentials (MLIPs) struggle with system‐specific chemistry, emphasizing the need for tailored datasets.
Martin Hoffmann Petersen +5 more
wiley +1 more source
An arithmetic method algorithm optimizing k-nearest neighbors compared to regression algorithms and evaluated on real world data sources. [PDF]
Sci RepAnagnostopoulos T +4 more
europepmc +1 more source
A Comprehensive Assessment and Benchmark Study of Large Atomistic Foundation Models for Phonons
Advanced Intelligent Discovery, EarlyView.We benchmark six large atomistic foundation models on 2429 crystalline materials for phonon transport properties. The rapid development of universal machine learning potentials (uMLPs) has enabled efficient, accurate predictions of diverse material properties across broad chemical spaces.
Md Zaibul Anam +5 more
wiley +1 more source
Ultrasound and microwave assisted extraction of bioactives from food wastes: An overview on their comparative analysis towards commercialization. [PDF]
Ultrason SonochemNayak A +6 more
europepmc +1 more source
Advanced Intelligent Discovery, EarlyView.This work establishes a correlation between solvent properties and the charge transport performance of solution‐processed organic thin films through interpretable machine learning. Strong dispersion interactions (δD), moderate hydrogen bonding (δH), closely matching and compatible with the solute (quadruple thiophene), and a small molar volume (MolVol)
Tianhao Tan, Lian Duan, Dong Wang
wiley +1 more source
Stacking ensemble machine learning for predicting photodetector performance under varying illumination intensities. [PDF]
Sci RepÖter A +4 more
europepmc +1 more source
Topology‐Aware Machine Learning for High‐Throughput Screening of MOFs in C8 Aromatic Separation
Advanced Intelligent Discovery, EarlyView.We screened 15,335 Computation‐Ready, Experimental Metal–Organic Frameworks (CoRE‐MOFs) using a topology‐aware machine learning (ML) model that integrates structural, chemical, pore‐size, and topological descriptors. Top‐performing MOFs exhibit aromatic‐enriched cavities and open metal sites that enable π–π and C–H···π interactions, serving as ...
Yu Li, Honglin Li, Jialu Li, Wan‐Lu Li
wiley +1 more source
BEmXRD-Nets framework for novel machine learning models to predict crystal energy with diversity structures. [PDF]
Sci RepBoonpan S +4 more
europepmc +1 more source
Advanced Intelligent Discovery, EarlyView.A physics‐guided machine learning framework estimates Young's modulus in multilayered multimaterial hyperelastic cylinders using contact mechanics. A semiempirical stiffness law is embedded into a custom neural network, ensuring physically consistent predictions. Validation against experimental and numerical data on C.
Christoforos Rekatsinas +4 more
wiley +1 more source
Cross-sectional analysis of accuracy versus interpretability in Medicare Advantage risk adjustment. [PDF]
Health Policy OpenLozinski M.
europepmc +1 more source