Results 51 to 60 of about 4,087 (203)

Efficient Hyperparameter Optimization for Reference Evapotranspiration Estimation with Limited Parameters: A Comparison of Optuna and Grid Search in the Doukkala Region, Morocco [PDF]

E3S Web of Conferences
Accurate estimation of reference evapotranspiration (ETo) is essential for irrigation scheduling and water resource management, particularly in semi-arid regions where meteorological data are often limited.
Belarbi Zaid, El Younoussi Yacine
doaj   +1 more source

Interpretable Machine Learning for Bandgap Prediction and Descriptor‐Guided Design Rules of Phosphates

Advanced Intelligent Discovery, EarlyView.
An explainable CatBoost model was trained to predict the bandgaps of 474 phosphate crystals based on composition and density descriptors. SHAP analysis identified two key variables—d‐electron‐count dispersion and atomic‐density dispersion—as the primary drivers of the model's predictions.
Wenhu Wang, Abudukadi Tudi, Ran An, Zhihua Yang   +3 more
wiley   +1 more source

FuseLog: Fusing Symbolic and Temporal Dynamics for Enhanced Log Anomaly Detection

IEEE Access
Log-based anomaly detection is crucial for ensuring the reliability of large-scale distributed systems. DeepLog implemented a sequence-oriented methodology utilizing LSTMs, providing a data-driven substitute for rule-based techniques.
Ahmed Alzamil, Ramzi Guesmi, Tarek Gasmi, Ismail Marouani, Amer Alghadhban, Badr M. Alshammari, Tawfik Guesmi, Mohamed Khalil Fadhlaoui, Noamen Guermazi   +8 more
doaj   +1 more source

Catalyst‐Specialized Chemical Language Model Based on Transformer Variational Autoencoder for Catalyst Design and Discovery

Advanced Intelligent Discovery, EarlyView.
We present CatTransVAE, a catalyst‐specialized chemical language model (CLM) built on a transformer variational autoencoder (VAE), developed through pretraining on general compounds followed by fine‐tuning on diverse catalyst databases. A template‐guided generation framework is introduced to enable controlled catalyst design under structural ...
Apakorn Kengkanna, Masahito Ohue
wiley   +1 more source

Classification of Infected Salmon Using CNN Deep Features and Optuna-Optimized SVM [PDF]

Fish diseases are a major challenge in the aquaculture industry, impacting productivity and the economy, particularly in salmon farming. This study aims to develop an image classification system for infected salmon using Convolution Neural Network (CNN ...
Ayu, Putu Desiana Wulaning, Pradita, Agus Hendra, Hostiadi, Dandy Pramana   +2 more
core   +2 more sources

A Comparative Study of Hyperparameter Optimization for CatBoost: Random Search, Optuna, and Successive Halving

Journal of Applied Informatics and Computing
This study aims to evaluate the effectiveness of three hyperparameter optimization approaches Random Search, Successive Halving, and Optuna in the CatBoost algorithm for modeling individual income using the 2024 SAKERNAS data.
Claudian Tikulimbong Tangdilomban, Kusman Sadik, Aji Hamim Wigena   +2 more
doaj   +1 more source

U‐Net‐Based Prediction of Concealed Volume Loss in AZ31 From Optical Corrosion Imprints and As‐Exposed Surface Topography

Advanced Intelligent Discovery, EarlyView.
How much material loss is concealed beneath a layer of corrosion products? This proof‐of‐concept study demonstrates how concealed volume loss is inferred from the optical appearance and as‐exposed surface topography of AZ31 corrosion imprints. By predicting postcleaning depth maps, a multimodal U‐Net model estimates hidden volume loss without chromic ...
Lars Dammann, Ci Song, Bahram Vaghefinazari, Tim Würger, Anna Lisitsyna, Shadi Albarqouni, Mikhail L. Zheludkevich, Sviatlana V. Lamaka, Christian Feiler   +8 more
wiley   +1 more source

Analysis of Ruddlesden‐Popper and Dion‐Jacobson 2D Lead Halide Perovskites Through Integrated Experimental and Computational Analysis

Battery Energy, Volume 4, Issue 2, March 2025.
Optimized ML framework for predicting RP and Dj phases in perovskite solar cells. ABSTRACT Two‐dimensional (2D) lead halide perovskites (LHPs) have captured a range of interest for the advancement of state‐of‐the‐art optoelectronic devices, highly efficient solar cells, next‐generation energy harvesting technologies owing to their hydrophobic nature ...
Basir Akbar, Kil To Chong, Hilal Tayara
wiley   +1 more source

Optimized Ensemble Learning Using Boosting, Stacking, and Voting for Early Stunting Risk Prediction

Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi)
Stunting remains a pressing health issue in Indonesia’s coastal communities, where uneven access to nutrition services and maternal–child care can limit early prevention efforts.
Munawir Munawir, M. Khairul Anam, Liza Fitria, Nurul Fadillah   +3 more
doaj   +1 more source

A Semi‐Automated Microfluidic Platform Employing Machine Learning Analysis to Study Adhesion Kinetics in Acute Myeloid Leukemia

Microscopy Research and Technique, EarlyView.
We have developed a semi‐automated shear flow platform using bright‐field optics and a machine‐learning analysis algorithm to dissect tumor‐microenvironment interactions. The algorithm quantifies the extent of adhesion at the single‐cell level and delivers consistent results within minutes instead of hours, facilitating high‐throughput analysis ...
Driti Ashok, Dennis Raith, Gabriel Kalweit, Maria Kalweit, Anusha Klett, Justus Duyster, Jonas Bermeitinger, Tanja Nicole Hartmann   +7 more
wiley   +1 more source