Near infrared (NIR) spectroscopy as a rapid and non-destructive technique was used with principal component analysis and linear discriminant analysis to develop an alternative technique for oil palm fresh fruit ripeness classification. The fruit samples were broadly classified into four main ripeness grades: unripe, under-ripe, ripe and over-ripe. A total of 3870 samples of individual fruits were taken at different stages of ripeness between 6 and 27 weeks after anthesis. Right after field collection, fresh samples were sent immediately to the laboratory for NIR and wet chemistry analysis. Two parameters, oil and moisture content in mesocarp were used as prior reference to determine the ripeness grades of oil palm fresh fruit. The NIR spectra of the ripe and over-ripe grades had higher absorbance compared to the unripe and under-ripe grades. The overall average accuracy of the classification by ripeness grades using this technique for both the training and the validation set samples was larger than 81%. In practical application, the rapid and non-destructive NIR technique has potential use for classifying oil palm fresh fruit into ripeness grades.
The extraction of relevant wavelengths from a large dataset of Near Infrared Spectroscopy (NIRS) is a significant challenge in vibrational spectroscopy research. Nonetheless, this process allows the improvement in the chemical interpretability by emphasizing the chemical entities related to the chemical parameters of samples. With the complexity in the dataset, it may be possible that irrelevant wavelengths are still included in the multivariate calibration. This yields the computational process to become unnecessary complex and decreases the accuracy and robustness of the model. In multivariate analysis, Partial Least Square Regression (PLSR) is a method commonly used to build a predictive model from NIR spectral data. However, in the PLSR method and common commercial chemometrics software, there is no standard wavelength selection procedure applied to screen the irrelevant wavelengths. In this study, a new robust wavelength selection procedure called the modified VIP-MCUVE (mod-VIP-MCUVE) using Filter-Wrapper method and input scaling strategy is introduced. The proposed method combines the modified Variable Importance in Projection (VIP) and modified Monte Carlo Uninformative Variable Elimination (MCUVE) to calculate the scale matrix of the input variable. The modified VIP uses the orthogonal components of Partial Least Square (PLS) in investigating the informative variable in the model by applying the amount of variation both in and , simultaneously. The modified MCUVE uses a robust reliability coefficient and a robust tolerance interval in the selection procedure. To evaluate the superiority of the proposed method, the classical VIP, MCUVE, and autoscaling procedure in classical PLSR were also included in the evaluation. Using artificial data with Monte Carlo simulation and NIR spectral data of oil palm (Elaeis guineensis Jacq.) fruit mesocarp, the study shows that the proposed method offers advantages to improve model interpretability, to be computationally extensive, and to produce better model accuracy.
Multivariate statistical analysis such as partial least square regression (PLSR) is the common data processing technique used to handle high-dimensional data space on near-infrared (NIR) spectral datasets. The PLSR is useful to tackle the multicollinearity and heteroscedasticity problem that can be commonly found in such data space. With the problem of the nonlinear structure in the original input space, the use of the classical PLSR model might not be appropriate. In addition, the contamination of multiple outliers and high leverage points (HLPs) in the dataset could further damage the model. Generally, HLPs contain both good leverage points (GLPs) and bad leverage points (BLPs); therefore, in this case, removing the BLPs seems relevant since it has a significant impact on the parameter estimates and can slow down the convergence process. On the other hand, the GLPs provide a good efficiency in the model calibration process; thus, they should not be eliminated. In this study, robust alternatives to the existing kernel partial least square (KPLS) regression, which are called the kernel partial robust GM6-estimator (KPRGM6) regression and the kernel partial robust modified GM6-estimator (KPRMGM6) regression are introduced. The nonlinear solution on PLSR was handled through kernel-based learning by nonlinearly projecting the original input data matrix into a high-dimensional feature mapping that corresponded to the reproducing kernel Hilbert spaces (RKHS). To increase the robustness, the improvements on GM6 estimators are presented with the nonlinear PLSR. Based on the investigation using several artificial dataset scenarios from Monte Carlo simulations and two sets from the near-infrared (NIR) spectral dataset, the proposed robust KPRMGM6 is found to be superior to the robust KPRGM6 and non-robust KPLS.
With the complexity of Near Infrared (NIR) spectral data, the selection of the optimal number of Partial Least Squares (PLS) components in the fitted Partial Least Squares Regression (PLSR) model is very important. Selecting a small number of PLS components leads to under fitting, whereas selecting a large number of PLS components results in over fitting. Several methods exist in the selection procedure, and each yields a different result. However, so far no one has been able to determine the more superior method. In addition, the current methods are susceptible to the presence of outliers and High Leverage Points (HLP) in a dataset. In this study, a new automated fitting process method on PLSR model is introduced. The method is called the Robust Reliable Weighted Average—PLS (RRWA-PLS), and it is less sensitive to the optimum number of PLS components. The RRWA-PLS uses the weighted average strategy from multiple PLSR models generated by the different complexities of the PLS components. The method assigns robust procedures in the weighing schemes as an improvement to the existing Weighted Average—PLS (WA-PLS) method. The weighing schemes in the proposed method are resistant to outliers and HLP and thus, preserve the contribution of the most relevant variables in the fitted model. The evaluation was done by utilizing artificial data with the Monte Carlo simulation and NIR spectral data of oil palm (Elaeis guineensis Jacq.) fruit mesocarp. Based on the results, the method claims to have shown its superiority in the improvement of the weight and variable selection procedures in the WA-PLS. It is also resistant to the influence of outliers and HLP in the dataset. The RRWA-PLS method provides a promising robust solution for the automated fitting process in the PLSR model as unlike the classical PLS, it does not require the selection of an optimal number of PLS components.
26 January 2026 in BAREKENG: Jurnal Ilmu Matematika dan Terapan
Indonesia, as a tropical country, is characterized by two primary seasons: the rainy season and the dry season. It is evident that meteorological shifts can exert considerable influence on the agricultural sector, a notable example being the cultivation of palm oil. Consequently, the ability to predict rainfall has emerged as a pivotal element in the broader endeavor to mitigate the adverse effects of climate change. This study employs statistical downscaling using the Principal Component Regression (PCR) approach to model rainfall predictions. The issue of multicollinearity, a common occurrence in Global Circulation Model (GCM) data, is addressed through the use of Principal Component Regression (PCR). This method has been demonstrated to stabilize the model structure and reduce variance in the regression coefficients. The data utilized encompass observed rainfall from LIBO Estate, which is owned by PT SMART Tbk (SMART Research Institute), for the period from 2013 to 2022. This data serves as the response variable, while the CMIP6 GCM simulation output data functions as the predictor variable. The findings indicated that the initial PCR model exhibited an RMSE value ranging from 97.06 to 131.69, along with an R² value ranging from 14.25% to 20.49%. The incorporation of dummy variables into the model resulted in a substantial enhancement in its performance, as evidenced by a decline in RMSE to 24.46–35.83 and an increase in R² to 89.02%–90.24%. The findings indicate that the use of PCR with dummy variables is an effective approach for enhancing the accuracy of rainfall modeling through statistical downscaling.
A common used as solution for local kernel nonparametric regression problem is given using polynomial regression. In this study, we demonstrated the estimator and properties using maximum likelihood estimator for a non-polynomial basis such B-spline to replacing the polynomial basis. This estimator allows for flexibility in the selection of a bandwidth and a knot. The best estimator was selected by finding an optimal bandwidth and knot through minimizing the famous generalized validation function.