NIRS-based fresh grape ripeness prediction with SPA-LASSO spectral feature selection
Jiayue Hu, Zhuo-Kang Wang, Yuyu Wang, Yuhao Wu, Haicheng Wei, Jing Zhao, Yang Liu, Yu-zhe Tan, Zi-Long Deng, Zhi-Jie Xiang, Ziyi Wang, Xintong Zhao
Abstract
= 0.944, RMSEC = 2.347, and RMSEP = 2.618. Overall, SPA-LASSO proved effective in feature selection, enhancing model generalization for spectroscopic screening in non-destructive grape maturity assessment.
Topics & Concepts
Partial least squares regressionRipenessLasso (programming language)Feature selectionMathematicsRipeningMean squared errorFeature (linguistics)StatisticsBiological systemChemistryArtificial intelligenceComputer scienceFood scienceBiologyPhilosophyLinguisticsWorld Wide WebSpectroscopy and Chemometric AnalysesFermentation and Sensory AnalysisHorticultural and Viticultural Research