Prediction of soil arsenic concentration in European soils: a dimensionality reduction and ensemble learning approach

Journal of Hazardous Materials Advances, Journal Year: 2025, Volume and Issue: 17, P. 100604 - 100604

Published: Jan. 15, 2025

Language: Английский

Improved Cd Detection in Rice Grain Using LIBS with Husk-Based XGBoost Transfer Learning

Agriculture, Journal Year: 2024, Volume and Issue: 14(11), P. 2053 - 2053

Published: Nov. 14, 2024

Cadmium (Cd) is a highly toxic metal that difficult to completely eliminate from soil, despite advancements in modern agricultural and environmental technologies have successfully reduced Cd levels. However, rice remains key source of exposure for humans. Even small amounts absorbed by can pose potential health risk the human body. Laser-induced breakdown spectroscopy (LIBS) has advantages simple sample preparation fast analysis, which, combined with transfer learning method, expected realize real-time rapid detection low-level heavy metals rice. In this work, 21 groups naturally matured samples potentially Cd-contaminated environments were collected. These processed into husk, brown rice, polished groups, reference content was measured ICP-MS. The XGBoost algorithm, known its excellent performance handling high-dimensional data nonlinear relationships, applied construct both base model XGBoost-based predict By pre-training on husk data, learn abundant information available improve quantification grain. For achieved RC2 0.9852 RP2 0.8778, which improved 0.9885 0.9743, respectively, model. case 0.9838 0.8683, while enhanced these 0.9883 0.9699, respectively. results indicate method not only improves capability low but also provides new insights food safety detection.

Language: Английский

Prediction of copper contamination in soil across EU using spectroscopy and machine learning: handling class imbalance problem

Smart Agricultural Technology, Journal Year: 2024, Volume and Issue: unknown, P. 100728 - 100728

Published: Dec. 1, 2024

Language: Английский