[Fast detection of sugar content in fruit vinegar using NIR spectroscopy]

Area of Science:

• Analytical Chemistry
• Spectroscopy
• Chemometrics

Context:

• Accurate quantification of sugar content in fruit vinegar is crucial for quality control.
• Traditional methods for sugar analysis can be time-consuming and destructive.
• Near-infrared (NIR) spectroscopy presents a non-destructive alternative for rapid chemical component measurement.

Purpose:

• To develop a rapid and precise method for determining the sugar content in fruit vinegar.
• To investigate the efficacy of combining Near-infrared (NIR) spectroscopy with Principal Component Analysis (PCA) and Least Squares Support Vector Machines (LS-SVM) for quantitative analysis.

Summary:

• Near-infrared (NIR) transmittance spectra of 300 fruit vinegar samples were acquired.
• Principal Component Analysis (PCA) was employed for spectral data dimensionality reduction, selecting six principal components (PCs).
• A prediction model for sugar content was built using Least Squares Support Vector Machines (LS-SVM) based on the selected PCs, achieving a high correlation coefficient (r=0.9939) and low prediction error (RMSEP=0.363).

Impact:

• The developed PCA-LS-SVM model demonstrates high prediction precision for sugar content determination in fruit vinegar.
• This technique offers a fast, non-destructive, and accurate alternative for fruit vinegar quality assessment.
• The findings support the application of spectroscopic methods combined with machine learning for food analysis.