首页|Studies from Chinese Academy of Sciences Yield New Data on Machine Learning [Optimizing Machine Learning Models for Predicting Soil Ph and Total P In Intact Soil Profiles With Visible and Nearinfrared Reflectance (Vnir) Spectroscopy]
Studies from Chinese Academy of Sciences Yield New Data on Machine Learning [Optimizing Machine Learning Models for Predicting Soil Ph and Total P In Intact Soil Profiles With Visible and Nearinfrared Reflectance (Vnir) Spectroscopy]
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-Data detailed on Machine Learning have been presented. According to news reporting from Nanjing, People's Republic of China, by NewsRx journalists, research stated, "Machine learning (ML) models hav e recently been used in visible and near -infrared reflectance (VNIR) spectrosco py applications. However, the predictive performance of ML models is data -speci fic and depends strongly on the selected hyperparameters." Financial supporters for this research include Chinese Academy of Sciences, Scie nce & Technology Fundamental Resources Investigation Program, Nati onal Key Research and Development Pro-gram of China, CAS Key Laboratory of Soil Environment and Pollution Remediation, ISSAS. The news correspondents obtained a quote from the research from the Chinese Acad emy of Sciences, "This study aimed to test the hyperparameter optimization metho ds on the three ML models (cubist regression tree, Cubist; support vector machin e regression, SVMR; and extreme gradient boosting, XGBoost) for predicting the s oil pH and total phosphorus (TP) in intact soil profiles to a depth of 100 +/- 5 cm. The VNIR spectra of nineteen intact soil profiles from several typical soil types in China were recorded. To determine the optimal hyperparameters of these ML models, a new Bayesian optimization (BO) strategy was introduced and compare d to the standard grid search (GS) approach. The accuracy of the models was comp ared with the partial least squares regression (PLSR) model in terms of the root mean square error (RMSE), the coefficient of determination (R2), and Lin ‘ s co ncordance correlation coefficient (LCC). Overall, the results showed that the BO -based models performed similarly to the GS -based models for soil pH and TP pr edictions. However, the BO method was more efficient for tuning the hyperparamet er values and had a considerably lower computational cost than the GS method. Th e tested ML models performed better than the PLSR models in all cases. Among the three ML techniques, the SVMR model achieved the best performance in terms of p redicting soil pH and TP. When the SVMR model was used on the testing set, the R MSE and R2 for soil pH were 0.26-0.27 and 0.97, respectively, while those for TP were 0.06 g kg(-1) and 0.85-0.87, respectively. Both soil properties were predi cted with excellent agreement (LCC >= 0.92)."
NanjingPeople's Republic of ChinaAsi aCyborgsEmerging TechnologiesMachine LearningChinese Academy of Sciences
NETL
NSTL
