作者: Ming Jing , Wensheng Cai , Xueguang Shao
DOI: 10.1080/00032711003686973
关键词:
摘要: Multiblock partial least squares (MB-PLS) are applied for determination of corn and tobacco samples by using near-infrared diffuse reflection spectroscopy. In the model, spectra separated into several sub-blocks along wavenumber, different latent variable number was used each sub-block. Compared with ordinary PLS, importance contribution sub-block can be balanced super-weights usage numbers. Therefore, prediction obtained MB-PLS model is superior to that especially large data sets a variables.