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.

Purpose We investigated the energy usage, economics, and global warming potential (GWP) of spring rice production via farm sizes in Nepal. Methods Seventy farmers were selected via purposive sampling methods, and data were collected by interviewing the farmers on site. Results It was found that 22,987 mega joules of energy per hectare (MJ/ha) was invested in order to yield 77,355 MJ/ha worth of outputs (5262 kg/ha). The analysis shows that the highest share of energy input was contributed by machinery (33.50%) and least by chemicals (0.96%). It was further found that spring rice farming in Nepal is less energy productive (0.23 kg/MJ) and less energy efficient (3.37) than that of neighboring countries-such as India, China, and Pakistan. The total GWP is estimated about 720.56 kg CO 2 eq/ha. The average cost of production was found to equate to USD810.24/ha (or USD0.154/kg of production), and that profit only reached USD73.93/ha. Conclusion This study found that energy use, cost of production, and yield per ha all generally decrease as the size of the landholding increases, whereas energy use efficiency and profits increase to an optimum level of land size and inputs. Low profits could be the result of the small size of landholdings and low levels of mechanization. This can be improved by increasing energy inputs and practicing community-farming.