Examination of Dimension Reduction Performances of PLSR and PCR Techniques in Data with Multicollinearity

Güven, Gamze; Şamkar, Hatice

doi:10.1007/s40995-018-0565-1

Cited by 13 publications

(7 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Least Squares (LS), Partial Least Squares (PLS), and Principal Component Regression (PCR) are approaches that can be applied to multicollinearity problems. In brief, compared to PCR, the PLS technique gives better results in solving a large number of independent variables [ 89 ]. These statistical approaches help in gather information and speed up the processing of analytical data.…”

Section: Rapid Methods For Non-nutritive Sweeteners Determinationmentioning

confidence: 99%

Analytical Methods for Determination of Non-Nutritive Sweeteners in Foodstuffs

et al. 2021

View full text Add to dashboard Cite

Sweeteners have been used in food for centuries to increase both taste and appearance. However, the consumption of sweeteners, mainly sugars, has an adverse effect on human health when consumed in excessive doses for a certain period, including alteration in gut microbiota, obesity, and diabetes. Therefore, the application of non-nutritive sweeteners in foodstuffs has risen dramatically in the last decade to substitute sugars. These sweeteners are commonly recognized as high-intensity sweeteners because, in a lower amount, they could achieve the same sweetness of sugar. Regulatory authorities and supervisory agencies around the globe have established the maximum amount of these high-intensity sweeteners used in food products. While the regulation is getting tighter on the market to ensure food safety, reliable analytical methods are required to assist the surveillance in monitoring the use of high-intensity sweeteners. Hence, it is also necessary to comprehend the most appropriate method for rapid and effective analyses applied for quality control in food industries, surveillance and monitoring on the market, etc. Apart from various analytical methods discussed here, extraction techniques, as an essential step of sample preparation, are also highlighted. The proper procedure, efficiency, and the use of solvents are discussed in this review to assist in selecting a suitable extraction method for a food matrix. Single- and multianalyte analyses of sweeteners are also described, employing various regular techniques, such as HPLC, and advanced techniques. Furthermore, to support on-site surveillance of sweeteners’ usage in food products on the market, non-destructive analytical methods that provide practical, fast, and relatively low-cost analysis are widely implemented.

show abstract

Section: Rapid Methods For Non-nutritive Sweeteners Determinationmentioning

confidence: 99%

Analytical Methods for Determination of Non-Nutritive Sweeteners in Foodstuffs

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Compared to other machine-learning models, PLSR exhibits higher accuracy in SOC estimation for various reasons. The main factor is handling multicollinearity, with PLSR being particularly adept at handling scenarios in which there is multicollinearity between independent variables [107]. PLSR is an appropriate method to effectively take these interactions into account [108].…”

Section: Soc Prediction Using Plsrmentioning

confidence: 99%

“…The third reason is that although the SVR model is capable of modeling The PLSR model demonstrates superior accuracy in estimating SOC because of different reasons. The main reason is that the PLSR is able to effectively manage the multicollinearity relations between the SOC laboratory or measured data and DRIFT-FTIR spectra [107,108]. The second reason is that PLSR is able to reduce the data dimensionality (caused by a high number of DRIFT-FTIR spectral data compared to the number of soil samples) and obtain the most effective information from the original data by creating several components [109].…”

Section: Svrmentioning

confidence: 99%

Using Various Models for Predicting Soil Organic Carbon Based on DRIFT-FTIR and Chemical Analysis

Thabit,

Negim,

AbdelRahman

et al. 2024

Soil Systems

View full text Add to dashboard Cite

Soil organic carbon (SOC) is a crucial factor influencing soil quality and fertility. In this particular investigation, we aimed to explore the possibility of using diffuse reflectance infrared fourier transform spectroscopy (DRIFT-FTIR) in conjunction with machine-learning models, such as partial least squares regression (PLSR), artificial neural networks (ANN), support vector regression (SVR) and random forest (RF), to estimate SOC in Sohag, Egypt. To achieve this, we collected a total of ninety surface soil samples from various locations in Sohag and estimated the total organic carbon content using both the Walkley-Black method and DRIFT-FTIR spectroscopy. Subsequently, we used the spectral data to develop regression models using PLSR, ANN, SVR, and RF. To evaluate the performance of these models, we used several evaluation parameters, including root mean square error (RMSE), coefficient of determination (R2), and ratio of performance deviation (RPD). Our survey results revealed that the PLSR model had the most favorable performance, yielding an R2 value of 0.82 and an RMSE of 0.006%. In contrast, the ANN, SVR, and RF models demonstrated moderate to poor performance, with R2 values of 0.53, 0.27, and 0.18, respectively. Overall, our study highlights the potential of combining DRIFT-FTIR spectroscopy with multivariate analysis techniques to predict SOC in Sohag, Egypt. However, additional studies and research are needed to improve the accuracy or predictability of machine-learning models incorporated into DRIFT-FTIR analysis and to compare DRIFT-FTIR analysis techniques with conventional soil chemical measurements.

show abstract

“…Throughout the processes of hyperspectral imaging technology, the core is to build precise and robust detection models. Up to now, there have been many classical modeling methods for employment, such as principal component regression (PCR), partial least squares regression (PLSR), 10 backpropagation neural network (BPNN), and radial basis function neural network (RBFNN). Among these classical methods, RBFNN and BPNN have the stronger ability to mine the nonlinear relationship between the input and output variables.…”

Section: Introductionmentioning

confidence: 99%