Classification of In-Situ Solar Wind Data Measured by Solar Orbiter/SWA-PAS and HIS Using Machine Learning DOI
L. Zhao, Henry Han, S. T. Lepri

и другие.

Communications in computer and information science, Год журнала: 2024, Номер unknown, С. 183 - 198

Опубликована: Янв. 1, 2024

Язык: Английский

A Clustering-Based Federated Deep Learning Approach for Enhancing Diabetes Management with Privacy-Preserving Edge Artificial Intelligence DOI Creative Commons
Xinyi Yang, Juan Li

Healthcare Analytics, Год журнала: 2025, Номер unknown, С. 100392 - 100392

Опубликована: Апрель 1, 2025

Язык: Английский

Процитировано

0
Explainable machine learning for high frequency trading dynamics discovery DOI
Henry Han, Jeffrey Yi‐Lin Forrest, Jiacun Wang

и другие.

Information Sciences, Год журнала: 2024, Номер 684, С. 121286 - 121286

Опубликована: Авг. 3, 2024

Язык: Английский

Процитировано

2
Improvement of land cover classification accuracy by training sample clustering DOI Creative Commons
Artem Andreiev,

Leonid Artiushyn

RADIOELECTRONIC AND COMPUTER SYSTEMS, Год журнала: 2024, Номер 2024(2), С. 66 - 72

Опубликована: Апрель 23, 2024

The subject of this article is land cover classification based on geospatial data. supervised methods are appropriate for most the thematic tasks remote sensing because they provide opportunity to set characteristics initial classes in form a training sample set, contrast unsupervised methods. There many approaches processing such set; however, their common disadvantage that do not consider factor separability. This characteristic indicates extent which signatures representing different overlap. A low degree separability inherent high-level mixing. Thus, affects accuracy. One possible ways increase clustering. Considering above, goal study develop clustering technique improve accuracy by increasing samples. work as follows: 1) method assessment; 2) separability; 3) test effectiveness developed applying it experimental classification. In experiments, two classifications were obtained each selected areas (i.e., one before and another after Six defined experiment. samples class. Conclusions. After application technique, an was evidenced index. turn, approach led improvement For first experiment, overall kappa coefficient 20% (from 63 83%) 21% 60% 81%), respectively. second 4% 77% 81%) 5% 66% 71%),

Язык: Английский

Процитировано

1
Dimension Reduction Stacking for Deep Solar Wind Clustering DOI
D. Carpenter, Henry Han, L. Zhao

и другие.

Communications in computer and information science, Год журнала: 2024, Номер unknown, С. 111 - 125

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0
Composition Analysis and Identification of Ancient Glass Products DOI
Xuemei Yang, Yuanyuan Zheng,

Yan‐Yan Xue

и другие.

Communications in computer and information science, Год журнала: 2024, Номер unknown, С. 78 - 94

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0
Classification of In-Situ Solar Wind Data Measured by Solar Orbiter/SWA-PAS and HIS Using Machine Learning DOI
L. Zhao, Henry Han, S. T. Lepri

и другие.

Communications in computer and information science, Год журнала: 2024, Номер unknown, С. 183 - 198

Опубликована: Янв. 1, 2024

Язык: Английский

Процитировано

0