The aim of this study was to evaluate and compare the performance of multivariate classification algorithms, specifically Partial Least Squares Discriminant Analysis (PLS-DA) and machine learning algorithms, in the classification of Monthong durian pulp based on its dry matter content (DMC) and soluble solid content (SSC), using the inline acquisition of near-infrared (NIR) spectra. A total of 415 durian pulp samples were collected and analyzed. Raw spectra were preprocessed using five different combinations of spectral preprocessing techniques: Moving Average with Standard Normal Variate (MA+SNV), Savitzky–Golay Smoothing with Standard Normal Variate (SG+SNV), Mean Normalization (SG+MN), Baseline Correction (SG+BC), and Multiplicative Scatter Correction (SG+MSC). The results revealed that the SG+SNV preprocessing technique produced the best performance with both the PLS-DA and machine learning algorithms. The optimized wide neural network algorithm of machine learning achieved the highest overall classification accuracy of 85.3%, outperforming the PLS-DA model, with overall classification accuracy of 81.4%. Additionally, evaluation metrics such as recall, precision, specificity, F1-score, AUC ROC, and kappa were calculated and compared between the two models. The findings of this study demonstrate the potential of machine learning algorithms to provide similar or better performance compared to PLS-DA in classifying Monthong durian pulp based on DMC and SSC using NIR spectroscopy, and they can be applied in the quality control and management of durian pulp production and storage.

This research aimed to determine the higher heating value, volatile matter, fixed carbon and ash content of ground bamboo using Fourier transform near infrared spectroscopy as an alternative to bomb calorimetry and thermogravimetry. Bamboo culms used in this study had circumferences ranging from 16 to 40 cm. Model development was performed using partial least squares regression. The higher heating value, volatile matter, fixed carbon and ash content were predicted with coefficients of determination (r²) of 0.92, 0.82, 0.85 and 0.51; root mean square error of prediction (RMSEP) of 122 J g⁻¹, 1.15%, 1.00% and 0.77%; ratio of the standard deviation to standard error of validation (RPD) of 3.66, 2.55, 2.62 and 1.44; and bias of 14.4 J g⁻¹, −0.43%, 0.03% and −0.11%, respectively. This report shows that near infrared spectroscopy is quite successful in predicting the higher heating value, and is usable with screening for the determination of fixed carbon and volatile matter. For ash content, the method is not recommended. The models should be able to predict the properties of bamboo samples which are suitable for achieving higher efficiency for the biomass conversion process.