OPENAIRE OpenAire Datacite Sygma Ghent University Academic Bibliography ZENODO Crossref Microsoft Academic Graph Europe PubMed Central European Union Open Data Portal Environmental Pollution Open Access 2021-09-09 Previous studies have mostly focused on using visible-to-near-infrared spectral technique to quantitatively estimate soil cadmium (Cd) content, whereas little attention has been paid to identifying soil Cd contamination from a perspective of spectral classification. Here, we developed a framework to compare the potential of two spectral transformations (i.e., raw reflectance and continuum removal [CR]), three optimization strategies (i.e., full-spectrum, Boruta feature selection, and synthetic minority over-sampling technique [SMOTE]), and three classification algorithms (i.e., partial least squares discriminant analysis, random forest [RF], and support vector machine) for diagnosing soil Cd contamination. A total of 536 soil samples were collected from urban and suburban areas located in Wuhan City, China. Specifically, Boruta and SMOTE strategies were aimed at selecting the most informative predictors and obtaining balanced training datasets, respectively. Results indicated that soils contaminated by Cd induced decrease in spectral reflectance magnitude. Classification models developed after Boruta and SMOTE strategies out-performed to those from full-spectrum. A diagnose model combining CR preprocessing, SMOTE strategy, and RF algorithm achieved the highest validation accuracy for soil Cd (Kappa = 0.74). This study provides a theoretical reference for rapid identification of and monitoring of soil Cd contamination in urban and suburban areas. DIFFUSE-REFLECTANCE SPECTROSCOPY HUMAN HEALTH PREDICTION POTENTIALLY TOXIC ELEMENTS Boruta algorithm 01 natural sciences Visible-to-near-infrared spectroscopy NIR SPECTROSCOPY Soil ORGANIC-CARBON Machine learning 11. Sustainability Soil Pollutants Least-Squares Analysis 0105 earth and related environmental sciences Spectroscopy, Near-Infrared RANDOM FOREST Urban and suburban soil Cd contamination 04 agricultural and veterinary sciences 15. Life on land QUANTITATIVE-ANALYSIS 6. Clean water RIVER DELTA 13. Climate action Earth and Environmental Sciences Synthetic minority over-sampling technique 0401 agriculture, forestry, and fisheries HEAVY-METAL CONCENTRATIONS Cadmium Hang Cheng, Ruili Shen, Zhou Shi, Zushuai Wei, Long Guo, Songchao Chen, Abdul Mounem Mouazen, Yaolin Liu, Jian Yang, Yiyun Chen, Gang Xu, Yongsheng Hong, Yongsheng Hong, 2021-12-01 journalpaper Previous studies have mostly focused on using visible-to-near-infrared spectral technique to quantitatively estimate soil cadmium (Cd) content, whereas little attention has been paid to identifying soil Cd contamination from a perspective of spectral classification. Here, we developed a framework to compare the potential of two spectral transformations (i.e., raw reflectance and continuum removal [CR]), three optimization strategies (i.e., full-spectrum, Boruta feature selection, and synthetic minority over-sampling technique [SMOTE]), and three classification algorithms (i.e., partial least squares discriminant analysis, random forest [RF], and support vector machine) for diagnosing soil Cd contamination. A total of 536 soil samples were collected from urban and suburban areas located in Wuhan City, China. Specifically, Boruta and SMOTE strategies were aimed at selecting the most informative predictors and obtaining balanced training datasets, respectively. Results indicated that soils contaminated by Cd induced decrease in spectral reflectance magnitude. Classification models developed after Boruta and SMOTE strategies out-performed to those from full-spectrum. A diagnose model combining CR preprocessing, SMOTE strategy, and RF algorithm achieved the highest validation accuracy for soil Cd (Kappa = 0.74). This study provides a theoretical reference for rapid identification of and monitoring of soil Cd contamination in urban and suburban areas. Diagnosis of cadmium contamination in urban and suburban soils using visible-to-near-infrared spectroscopy 10.1016/j.envpol.2021.118128 https://doi.org/10.1016/j.envpol.2021.118128 818346