Results 201 to 210 of about 38,464 (255)

Machine Learning Modeling of Drill Cuttings Transport Based on Computational Fluid Dynamics Simulation Data

Energy Science &Engineering, EarlyView.
Machine learning surrogates accurately predict drill cuttings transport from computational fluid dynamics data, with Reynolds number dominating system behavior. While tree‐based models excel in accuracy, Gaussian process regression provides physically consistent, smooth predictions essential for reliable drilling simulations.
Kathrin Skinder, Mostafa M. Abdelhafiz, Khizar Shahid, Gunther Brenner   +3 more
wiley   +1 more source

Custom Loss Functions in XGBoost Algorithm for Enhanced Critical Error Mitigation in Drill-Wear Analysis of Melamine-Faced Chipboard. [PDF]

Sensors (Basel)
Bukowski M, Kurek J, Świderski B, Jegorowa A.   +3 more
europepmc   +1 more source

A Genetic Algorithm‐Optimized LSTM Framework for Mechanics‐Conditioned Reconstruction of Stress–Strain Curves in High‐Performance Cementitious Materials

Energy Science &Engineering, EarlyView.
GA‐optimized LSTM for full stress–strain curve prediction of high‐performance cementitious materials. Compression‐test data are preprocessed (cleaning, normalization, interpolation/standardization, and Savitzky–Golay smoothing). A Genetic Algorithm tunes LSTM hyperparameters, while a physics‐informed loss enforces stiffness, peak response ( σ max ...
Mahmoud AlJamal, Qasim Aljamal, Mohammad Q. Al‐Jamal, Bashar S. Khassawneh, Ayoub Alsarhan, Dimah Al‐Fraihat, Abdulmohsen Algarni, Osama Harfoushi   +7 more
wiley   +1 more source

Development of CQ prediction based on XGBoost Algorithm

Journal of the Korea Institute of Information and Communication Engineering
Seungyeop Baek, Hyunbin Han, Yoon-Soo Seo, Keun-Young Lee, Donghyun Kim, Jihoon Park, Gwanghyun Jo   +6 more
openaire   +1 more source

High Precision Prediction Model of Wind Speed for Underground Mine Tunnel by Using ASTF Net

Energy Science &Engineering, EarlyView.
This paper proposes an Adaptive Spectral‐Temporal Fusion Net (ASTF‐Net), which features a novel decomposition‐ensemble framework with adaptive feature fusion. This design deliberately balances spectral and temporal features and effectively boosts prediction accuracy in complex ventilation environments. ABSTRACT An Adaptive Spectral‐Temporal Fusion Net (
Ruo‐Qi Li, Yong‐Qiang Chai, Lian Shi, Fu‐Gang Wang, Mi Liu   +4 more
wiley   +1 more source

DFT and Machine Learning Analysis of Eco‐Friendly Formamidinium Tin Iodide Perovskite Solar Cells: Optimizing Charge Transport Layers Through SCAPS‐1D Simulations

Energy Science &Engineering, EarlyView.
Machine learning using Random Forest, GB, XGBoost, and LightGBM regression algorithms: SHAP plot for PCE. ABSTRACT Tin‐based solar cells are the eco‐friendly alternative to the toxic lead‐based cells that are currently in use. In this context, Formamidinium Tin Iodide (FASnI3) emerges as a promising candidate due to its superior performance, exhibiting
Rabeya Khan, Md Rasidul Islam, I.K. Gusral Ghosh Apurba, Md. Amimul Ihsan, Rahat Ul Nasib, Mohammad Arefin, Md Rifatul Islam, Abu Zahid, Hasnain A. Ziad, Jehan Y. Al‐Humaidi   +9 more
wiley   +1 more source

Forecasting Financial Assets Using Recursive Clustering Approach

Journal of Forecasting, EarlyView.
ABSTRACT Machine learning‐based forecasting has been gaining attention for its flexibility and adaptability. Given the dynamics of temporal data, clustering can be an effective approach for improving forecasting accuracy. While clustering for classification typically considers all observations in the data to effectively divide the data into clusters ...
Jeongwoo Kim
wiley   +1 more source

MARS‐Diff: Guarded Residual Diffusion for Leakage‐Disciplined Probabilistic Portfolio‐Loss Forecasting

Journal of Forecasting, EarlyView.
ABSTRACT This paper develops Masked Asset–Regime Scenario Diffusion (MARS‐Diff), a leakage‐disciplined framework for probabilistic forecasting of multiday portfolio losses. The framework combines a regularized heterogeneous autoregressive model with exogenous predictors (HAR‐X) as its anchor, a train‐only masked representation of a high‐dimensional ...
Çağlar Sözen, Mervenur Sözen
wiley   +1 more source

Beyond Prediction: Data, Baselines, Explanation, and Causation in Machine Learning for Food Insecurity

Food Safety and Health, EarlyView.
The gains from machine learning in nowcasting and forecasting food insecurity are still small and limited and cannot be observed in all countries. This review summarizes the public resources for data, corrects common misconceptions about model requirements, and establishes baseline requirements, explanations, causal inference, and equity in operational
Shabnam Mehboob, Azman Abdullah, Gunesh Kumar, Samreen Memon, Naila Afghan   +4 more
wiley   +1 more source

State‐Dependent Market (In)Efficiency in Cryptocurrency Markets

Journal of Futures Markets, EarlyView.
ABSTRACT This paper develops a meta‐learning framework to test whether cryptocurrency market inefficiency is static or state‐dependent. To resolve competing claims between the Efficient and Adaptive Market Hypotheses (AMH), we construct an adaptive trading strategy using high‐frequency data for 50 major cryptocurrencies.
Sasan Barak, Siavash Razmi, Alireza Mousavi   +2 more
wiley   +1 more source