Skip to content
DraconiticResearch
Search Sources API
194,220 studies31 sourcesUpdated 44 min ago
Peer-reviewedCC BYFoods

Rapid Determination of Soybean Protein Content by Near-Infrared Spectroscopy Coupled with Multi-Learner Ensemble Wavelength Selection

  • Weida Wang1
  • Chunqi Wang2
  • Baocheng Zhao1
  • Jiayi Shi1
  • Changan Xu3
  • Jinming Liu1
  1. 1College of Information and Electrical Engineering, Heilongjiang Bayi Agricultural University, Daqing 163319, China
  2. 2College of Food, Heilongjiang Bayi Agricultural University, Daqing 163319, China
  3. 3College of Materials and Energy, South China Agricultural University, Guangzhou 510642, China
Open at sourceDOICite

Abstract

Soybean protein content is a key indicator of nutritional value and quality grade, and its determination is important for quality evaluation and cultivar selection. To overcome the time-consuming and costly limitations of conventional chemical assays, this study proposed a multiple linear learner ensemble importance-score wavelength selection (MLLEISWS) method to identify informative wavelengths from soybean near-infrared spectra and establish a partial least squares (PLS) model. MLLEISWS was compared with competitive adaptive reweighted sampling, successive projections algorithm, and uninformative variable elimination. Shapley additive exPlanations (SHAP) were applied to the MLLEISWS algorithm to interpret the selected wavelengths. Results showed that the PLS model developed using MLLEISWS achieved the best performance. With only 29 selected wavelengths, the coefficients of determination for the training and test sets reached 0.941 and 0.933, respectively. Root mean square errors were 0.490% and 0.514%, relative root mean square errors were 1.32% and 1.37%, and residual predictive deviation was 3.863, indicating predictive accuracy and stability. SHAP analysis showed that the selected wavelengths were located in protein-related spectral regions and corresponded to overtone and combination bands information from functional groups. MLLEISWS effectively reduced variable dimensionality while maintaining model performance.

Related

  1. Dual-band scanned-wavelength infrared laser-induced fluorescence of CO with time-domain modeling for 1 kHz thermometry
    Garrett C. Mathews, Jonathan Rustad, Christopher S. Goldenstein
    Optics ExpressPeer-reviewedPublisher termsFull textPhysical chemistryCrossref
  2. Infrared Spectra of the Water–CO2 Complex in the 4.3–3.6 μm Region and Determination of the Ground State Tunneling Splitting for HDO–CO2
    Travis Gartner, Clément Lauzin, A. R. W. Mckellar +1
    Journal of Physical Chemistry APeer-reviewedLicence not statedFull text3 citedPhysical chemistryGeneral physicsHAL
  3. Wavelength-Resolved Timing of Near-Infrared Scintillation in Gaseous Xenon
    Robert Hammann, Steffen Form, Humaira Nazneen +3
    arXivPreprintCC BY 4.0Full textPhysical chemistryApplied physics & instrumentationarXiv
  4. Short Wavelength Infrared Diimmonium Nano-Agents for Photoacoustic Imaging
    Longcai Liu, Yang Chen, Liya Yu +3
    Nature CommunicationsPeer-reviewedCC BY 4.0Full textMaterials & polymersCrossref
  5. Determination of Hydroquinone Clathrate Stability for Different Guest Molecules Using NPH Ensemble Molecular Dynamics.
    Brais R García, Jesús Algaba, Martín Pérez-Rodríguez +3
    Energy & fuels : an American Chemical Society journalPeer-reviewedCC BYFull textOrganic chemistryEurope PMC

Cite

BibTeXRIS
BibTeX
@article{wang2026rapid,
  title = {{Rapid Determination of Soybean Protein Content by Near-Infrared Spectroscopy Coupled with Multi-Learner Ensemble Wavelength Selection}},
  author = {Weida Wang and Chunqi Wang and Baocheng Zhao and Jiayi Shi and Changan Xu and Jinming Liu},
  journal = {Foods},
  year = {2026},
  month = may,
  doi = {10.3390/foods15101755},
  url = {https://doi.org/10.3390/foods15101755},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Rapid Determination of Soybean Protein Content by Near-Infrared Spectroscopy Coupled with Multi-Learner Ensemble Wavelength Selection
AU  - Weida Wang
AU  - Chunqi Wang
AU  - Baocheng Zhao
AU  - Jiayi Shi
AU  - Changan Xu
AU  - Jinming Liu
PY  - 2026
DA  - 2026/05/15
JO  - Foods
DO  - 10.3390/foods15101755
UR  - https://doi.org/10.3390/foods15101755
AB  - Soybean protein content is a key indicator of nutritional value and quality grade, and its determination is important for quality evaluation and cultivar selection. To overcome the time-consuming and costly limitations of conventional chemical assays, this study proposed a multiple linear learner ensemble importance-score wavelength selection (MLLEISWS) method to identify informative wavelengths from soybean near-infrared spectra and establish a partial least squares (PLS) model. MLLEISWS was compared with competitive adaptive reweighted sampling, successive projections algorithm, and uninformative variable elimination. Shapley additive exPlanations (SHAP) were applied to the MLLEISWS algorithm to interpret the selected wavelengths. Results showed that the PLS model developed using MLLEISWS achieved the best performance. With only 29 selected wavelengths, the coefficients of determination for the training and test sets reached 0.941 and 0.933, respectively. Root mean square errors were 0.490% and 0.514%, relative root mean square errors were 1.32% and 1.37%, and residual predictive deviation was 3.863, indicating predictive accuracy and stability. SHAP analysis showed that the selected wavelengths were located in protein-related spectral regions and corresponded to overtone and combination bands information from functional groups. MLLEISWS effectively reduced variable dimensionality while maintaining model performance.
SN  - 2304-8158
ER  -