Root Mean Square Error Of Prediction Values Research Articles

A colorimetric microwell method using a desktop scanner for biochemical assays

A method for rapid, inexpensive and sensitive simultaneous analysis of glucose, creatinine, triglycerides, total cholesterol and total protein is needed to analyze blood. The proposed method is based on the production of a specific color after reaction. The method was adapted to a 64-microwell plate format, and it uses the transparency scanner feature of a commercially available desktop scanner. Each microwell plate had an 8×8 array of flat-bottomed 250μL microwells, and these microwells were used to simultaneously house the solutions for clinical assay. The scanned image was saved in TIFF format in a portable computer and then processed using a Graphic User Interface (GUI) designed in our laboratory to obtain analytical curves and to automate the mathematics and statistics calculations. This automation improved the analytical frequency of the method. The results showed that it is possible to measure a few microliters of solution with exactitude and precision better than 5.30%. The measured concentration ranges of glucose, triglycerides, creatinine, total cholesterol and total protein were 0.781 to 100, 1.56 to 200, 0.031 to 4.0, 1.56 to 200mgdL−1 and 0.031 to 4.0gdL−1, respectively. The limits of detection were 16.2, 51.7, 0.12, 41.5mgdL−1 and 0.62gdL−1 for glucose, triglycerides, creatinine, total cholesterol and total protein, respectively. The recoveries were from 98.7% to 101.3% for total cholesterol, 98.7% to 124.9% for triglycerides, 54.2% to 98.3% for total protein, 89.6% to 101% for glucose and 65.7% to 115.4% for creatinine. The results provided by the scanner were compared with those obtained with a commercial photometer and did not show significant differences at a confidence level of 95%. Good results were obtained for the correlation coefficient and Root Mean Square Error of Prediction (RMSEP) values for the five parameters, especially the total cholesterol and creatinine. The RMSEP values for glucose, creatinine, triglyceride, total cholesterol and total protein were 8.05, 0.28, 7.69, 1.41mgdL−1 and 2.2gdL−1, respectively.

Development and Evaluation of a Near‐Infrared Instrument for Single‐Seed Compositional Measurement of Wheat Kernels

ABSTRACTA single‐kernel, near‐infrared reflectance instrument was designed, built, and tested for its ability to measure composition and traits in wheat kernels. The major objective of the work was targeted at improving an existing design concept of an instrument used for larger seeds such as soybeans and corn but in this case designed for small seeds. Increases in throughput were sought by using a vacuum to convey seeds without compromising measurement accuracy. Instrument performance was evaluated by examining measurement accuracy of wheat kernel moisture, protein content, and kernel mass. Spectral measurements were obtained on individual wheat kernels as they were conveyed by air through an illuminated tube. Partial least squares (PLS) prediction models for these constituents were then developed and evaluated. PLS single‐kernel moisture predictions had a root mean square error of prediction (RMSEP) around 0.5% MC wet basis; protein prediction models had an RMSEP near 0.70%. Prediction of mass was not as good but still provided a reasonable estimate of single‐kernel mass, with RMSEP values of 2.8–4 mg. Data showed that kernel mass and protein content were not correlated, in contrast to some previous research. Overall, results showed the instrument performed comparably to other single‐seed instruments or methods based on accuracy but with an increased throughput at a rate of at least 4 seeds/s.

Application of Time Series Hyperspectral Imaging (TS-HSI) for Determining Water Content within Tea Leaves during Drying

Abstract. This research investigated the feasibility of using time series hyperspectral imaging (TS-HSI) for rapid and nondestructive determination of water content in tea leaves. Hyperspectral images of tea leaves were obtained at different periods of drying across the wavelength region of 380 to 1030 nm. The reflectance value of the region of interest (ROI) was extracted with ENVI 4.7 software. Different preprocessing methods were applied to determine the best method based on the root mean square error of calibration (RMSEC), root mean square error of prediction (RMSEP), residual predictive deviation (RPD), and coefficient of determination (R 2 ) of the partial least squares regression (PLSR) model. The successive projections algorithm (SPA) was then used to identify the most important wavelengths and reduce the high dimensionality of the spectral data. On the basis of the four effective wavelengths (542, 709, 752, and 971 nm), an SPA-PLSR model was established. Among the different PLSR models, the multiplicative scatter correction-PLSR (MSC-PLSR) model performed best with the highest values of R 2 cal , R 2 val , R 2 pre , and RPD (0.979, 0.961, 0.968, and 5.616, respectively) and the lowest values of RMSEC, RMSEV, and RMSEP (0.033, 0.045, and 0.040, respectively). However, the SPA-PLSR mode, with only four input variables, was considered to be preferable for determining water content in tea leaves. The SPA-PLSR model obtained R 2 cal , R 2 val , R 2 pre , and RPD values of 0.938, 0.935, 0.946, and 4.292, respectively, and RMSEC, RMSEV, and RMSEP values of 0.055, 0.057, and 0.052, respectively. The results showed that the TS-HSI technique has potential to be a rapid and nondestructive method to detect water content in tea leaves at different drying periods.

Analysis and Prediction of Unreacted Starch Content in Corn Using FT-NIR Spectroscopy

<abstract> <bold><sc>Abstract.</sc></bold> When corn is processed in a conventional dry-grind ethanol process, a portion of the corn starch is not readily converted into ethanol. The amount of unconverted, or unreacted, starch varies according to several factors, including storage time and processing conditions. The current method for determining the amount of unreacted starch is based on an enzyme assay, which is time-consuming and does not lend itself to on-line measurements of corn in processing plants. A rapid method for determining the unreacted starch in corn would be advantageous so that the mix of enzymes and processing conditions could be adjusted to ensure maximum ethanol yield. In this study, we demonstrated the feasibility of using Fourier transform near-infrared (FT-NIR) spectroscopy in developing predictive models of unreacted starch in corn. FT-NIR spectra of corn starch blends and ground corn from 4000 to 10000 cm^-1 were calibrated against unreacted starch content, determined enzymatically, using various spectral preprocessing techniques such as multiplicative scatter correction (MSC), Savitzky-Golay (SG) derivative algorithms, and partial least squares regression (PLSR). Results showed that the unreacted starch content in blends can be predicted with a low root mean square error of prediction (RMSEP) ranging from 1.29% to 1.95%, a coefficient of regression (R²) of 0.97 to 0.98, and a ratio of performance to deviation (RPD) of 4.82 to 7.28. PLS regression models for unreacted starch content in dry and wet ground corn were equally promising, with low RMSEP values of 1.13% to 2.23%, R² values of 0.83 and 0.94, and RPD values of 1.55 to 2.16. These models are a valuable tool for high-throughput monitoring of unreacted starch during corn storage, handling, and processing.

Determination of API gravity, kinematic viscosity and water content in petroleum by ATR-FTIR spectroscopy and multivariate calibration

In this work, API gravity, kinematic viscosity and water content were determined in petroleum oil using Fourier transform infrared spectroscopy with attenuated total reflectance (FT-IR/ATR). Support vector regression (SVR) was used as the non-linear multivariate calibration procedure and partial least squares regression (PLS) as the linear procedure. In SVR models, the multiplication of the spectra matrix by support vectors resulted in information about the importance of the original variables. The most important variables in PLS models were attained by regression coefficients. For API gravity and kinematic viscosity these variables correspond to vibrations around 2900cm−1, 1450cm−1 and below to 720cm−1 and for water content, between 3200 and 3650cm−1, around 1650cm-1 and below to 900cm−1. The SVR model produced a root mean square error of prediction (RMSEP) of 0.25 for API gravity, 22mm2s−1 for kinematic viscosity and 0.26%v/v for water content. For PLS models, the RMSEP values for API gravity was 0.38mm2s−1, for kinematic viscosity was 27mm2s−1 and for water content was 0.34%. Using the F-test at 95% of confidence it was concluded that the SVR model produced better results than PLS for API gravity determination. For kinematic viscosity and water content the two methods were equivalent. However, a non-linear behavior in the PLS kinematic viscosity model was observed.

Estimating Deoxynivalenol Content of Ground Oats Using VIS‐NIR Spectroscopy

ABSTRACTThe potential of VIS‐NIR spectroscopy as a rapid screening method for resistance of Fusarium‐inoculated oats to replace the costly chemical measurements of deoxynivalenol (DON) was investigated. Partial least squares (PLS) regression was conducted on second‐derivative spectra (400–2,350 nm) of 166 DON‐contaminated samples (0.05–28.1 ppm, mean = 13.06 ppm) with separate calibration and test set samples. The calibration set had 111 samples, and the test set had 55 samples. The best model developed had three PLS components and a root mean square error of prediction (RMSEP) of 3.16 ppm. The residual predictive deviation (RPD) value of the prediction model was 2.63, an acceptable value for the purpose of rough screening. Visual inspection and the VIS spectra of the samples revealed that high‐DON samples tended to be darker in color and coarser in texture compared with low‐DON samples. The second‐derivative spectra showed that low‐DON samples tended to have more water and fat content than high‐DON samples. With an RMSEP value of 3.16 and RPD of value of 2.63, it seems possible to use VIS‐NIR spectroscopy to semiquantitatively estimate DON content of oats and discard the worst genotypes during the early stages of screening.

In-line monitoring of the drug content of powder mixtures and tablets by near-infrared spectroscopy during the continuous direct compression tableting process

Continuous manufacturing methods offer economic and quality advantages when compared with batch manufacturing methods. In continuous manufacturing, one requires real time assurance of quality of product via the implementation of PAT tools. This study focuses on an in-line near-infrared (NIR) spectroscopic method for determining the drug content of powder mixtures and tablets during a continuous tableting process. Tablets consisting of acetaminophen (20–30%), lactose (69.07–78.93%) and magnesium stearate (0.93–1.07%) were prepared in a continuous direct compression line that consisted of two loss-in-weight feeders, one for acetaminophen and one for premixed lactose and magnesium stearate, and a continuous mixer followed by a rotary tablet press. NIR spectroscopy was applied to the continuous mixer and tablet press to perform a 100% product check at full tableting speed. The UV-spectrophotometric method was used as an off-line reference method to determine the acetaminophen content in the samples. The powder mixture and tablet samples were taken during the process for the calibration of continuous mixer and tablet press, respectively. For the continuous mixer, model creation with the PLS method yielded R-Square and RMSEC (root mean square error of calibration) values of 0.975% and 0.56%, respectively. For the tablet press, the corresponding R-Square and RMSEC values were 0.943% and 0.75%, respectively. A test run demonstrated good predictability in the estimation of the API content in the powder mixtures and tablets during the continuous tableting process. For the continuous mixer and tablet press, the RMSEP (root mean square error of prediction) values were 0.96% and 1.37%, respectively. This study demonstrates that an NIR instrument capable of fast spectra acquisition can be a valuable tool for the in-line monitoring of the continuous mixing and tableting processes.

Predicting Soil Organic Carbon at Field Scale Using a National Soil Spectral Library

Visible and near infrared diffuse reflectance (vis-NIR) spectroscopy is a low-cost, efficient and accurate soil analysis technique and is thus becoming increasingly popular. Soil spectral libraries are commonly constructed as the basis for estimating soil texture and properties. In this study, partial least squares regression was used to develop models to predict the soil organic carbon (SOC) content of 35 soil samples from one field using (i) the Danish soil spectral library (2688 samples), (ii) a spiked spectral library (a combination of 30 samples selected from the local area and the spectral library, 2718 samples) and (iii) three sub-sets selected from the spectral library. In an attempt to improve prediction accuracy, sub-sets of the soil spectral library were made using three different sample selection methods: those geographically closest (84 samples), those with the same landscape and parent material (96 samples) and those with the most alike spectra to spectra from the field investigation (100 samples). These sub-sets were used to develop three calibration models and in predictions of SOC content. The results showed that the geographically closest model, which used the fewest number of samples, gave the lowest root mean square error of prediction ( RMSEP) of 0.19% and the highest ratio of performance to deviation ( RPD) of 3.7, followed by the spiked library, same parent material, the spectral library and the most alike spectra. The spiked library model also gave a low RMSEP value of 0.19% and high RPD value of 3.7% and performed markedly better than the model without spiking, despite using 30 samples for library spiking. The accuracy of the model developed using a sub-set from a spectral library was highly dependent on geographical location, soil parent material and landscape.

Rapid determination of Panax ginseng by near-infrared spectroscopy

A rapid method for the simultaneous determination of the main panaxadiol saponins, panaxatriol saponins and ginseng polysaccharide in Panax ginseng was developed using near-infrared spectroscopy (NIRS). The regression models were developed using multivariate regression methods with the results from conventional chemical analytical methods as reference values. Multivariate regression methods were discussed and Partial Least Squares Regression (PLSR) was more suitable. Multiplicative scatter correction (MSC), second derivative and Savitzky–Golay smoothing were adopted together for the spectral preprocessing. The final models were evaluated by the correlation coefficient (R2) and the root mean square error of prediction (RMSEP). The optimum RMSEP values of the panaxadiol saponins, panaxatriol saponins and ginseng polysaccharide were 0.091, 0.044 and 2.20 respectively, and the R2 values were 0.994, 0.989 and 0.965 respectively. The results demonstrated that NIR spectroscopy could be used for the rapid determination of Ginseng Radix et Rhizoma.

Developing and Evaluating NIR Calibration Models for Multi-Species Herbaceous Perennials

Abstract Near-infrared (NIR) calibration models for lignin, glucan, and xylan were developed with 55 samples of individual and mixed biomass species containing wheatgrass (Thinopyrum spp.), switchgrass (Panicum virgatum L.), wild rye (Leymus spp.), big bluestem (Andropogon gerardii), alfalfa (Medicago sativa L.), and sweetclover (Melilotus officinalis L.) planted in five locations across North Dakota. The models were developed using a diode array (DA) 7200 NIR spectrometer (950–1,650 nm) and GRAMS statistical software. Models were validated with 10 samples, and the root mean square errors of prediction (RMSEP) were 0.54, 1.12, and 0.97 for lignin, glucan, and xylan, respectively. Lignin, glucan, and xylan models had R2 values of 0.873, 0.843, and 0.902, respectively. The RMSEP and R2 values indicate that it is possible to develop acceptable calibration models to predict chemical composition for mixtures of herbaceous perennials. NIR instrument-based component predictions were consistent over a period of h...

Improvement of spectral calibration for food analysis through multi-model fusion

Near-infrared (NIR) spectroscopy will present a more promising tool for quantitative analysis if the predictive ability of the calibration model is further improved. To achieve this goal, a new ensemble calibration method based on uninformative variable elimination (UVE)-partial least square (PLS) is proposed, which is named as ensemble PLS (EPLS), meaning a fusion of multiple PLS models. In this method, different calibration sets are first generated by bootstrap and different PLS models are obtained. Then, the UVE is used to shrink the original variable space into a specific subspace. By repeating this process, a fixed number of candidates PLS member models are obtained. Finally, a smaller part of candidate models are integrated to produce an ensemble model. In order to verify the performance of EPLS, three NIR spectral datasets from food industry were used for illustration. Both full-spectrum PLS and UVEPLS of single models were used as reference. It was found that the proposed method could lead to lower RMSEP (root mean square error of prediction) value than PLS and UVEPLS and such an improvement is statistically significant according to a paired t-test. The results showed that the method is of value to enhance the predictive ability of PLS-based calibration involving complex NIR matrices in food analysis.

Determining Frying Oil Degradation by Near Infrared Spectroscopy Using Chemometric Techniques

AbstractA rapid method for the determination of some important physicochemical properties in frying oils has been developed. Partial least square regression (PLS) calibration models were applied to the physicochemical parameters and near infrared spectroscopy (NIR) spectral data. PLS regression was used to find the NIR region and the data pre‐processing method that give the best prediction of the chemical parameters. Calibration and validation were appropriated by leave one out cross validation and test set validation techniques for predicting free fatty acids (FFA), total polar materials (cTPM; measured by chromatographic method and iTPM measured by an instrumental method), viscosity and smoke point of the frying oil samples. For PLS models using the cross validation techniques, the best correlations (r) between NIR predicted data and the standard method data for iTPM in oils were 93.79 and root mean square error of prediction (RMSEP) values were 5.53. For PLS models using the test set validation techniques, the best correlations (r) between NIR predicted data and standard method data for FFA, cTPM, viscosity and smoke point in oils were 92.58, 94.61, 81.95 and 84.07 and RMSEP values were 0.121, 3.96, 22.30 and 8.74, respectively. In conclusion, NIR technique with chemometric analysis was found very effective in predicting frying oil quality changes.

Influence of the number of samples on prediction error of visible and near infrared spectroscopy of selected soil properties at the farm scale

This paper reports on the influence of the number of samples used for the development of farm‐scale calibration models for moisture content (MC), total nitrogen (TN) and organic carbon (OC) on the prediction error expressed as root mean square error of prediction (RMSEP) for visible and near infrared (vis‐NIR) spectroscopy. Fresh (wet) soil samples collected from four farms in the Czech Republic, Germany, Denmark and the UK were scanned with a fibre‐type vis‐NIR, AgroSpec spectrophotometer with a spectral range of 305–2200 nm. Spectra were divided into calibration (two thirds) and prediction (one third) sets and the calibration spectra were subjected to a partial least squares regression (PLSR) with leave‐one‐out cross‐validation using Unscrambler 7.8 software. The RMSEP values of models with a large sample number (46–84 samples from each farm) were compared with those of models developed with a small sample number (25 samples selected from the large sample set of each farm) for the same variation range. Both large‐set and small‐set models were validated by the same prediction set for each property. Further PLSR analysis was carried out on samples from the German farm, with different sample numbers of the calibration set of 25, 50, 75 and 100 samples. Results showed that the large‐size dataset models resulted in smaller RMSEP values than the small‐size dataset models for all the soil properties studied. The results also demonstrated that with the increase in sample number used in the calibration set, RMSEP decreased in almost linear fashion, although the largest decrease was between 25 and 50 samples. Therefore, it is recommended that the number of samples should be chosen according to the accuracy required, although 50 soil samples is considered appropriate in this study to establish calibration models of TN, OC and MC with smaller expected prediction errors as compared with smaller sample numbers.

Robust preprocessing and model selection for spectral data

To calibrate spectral data, one typically starts with preprocessing the spectra and then applies a multivariate calibration method such as principal component regression or partial least squares regression. In the model selection step, the optimal number of latent variables is determined in order to minimize the prediction error. To protect the analysis against the harmful influence of possible outliers in the data, robust calibration methods have been developed. In this paper, we focus on the preprocessing and the model selection step. We propose several robust preprocessing methods as well as robust measures of the root mean squared error of prediction (RMSEP). To select the optimal preprocessing method, we summarize the results for the different RMSEP values by means of a desirability index, which is a concept from industrial quality control. These robust RMSEP values are also used to select the optimal number of latent variables. We illustrate our newly developed techniques through the analysis of a real data set containing near‐infrared measurements of samples of animal feed. Copyright © 2012 John Wiley &amp; Sons, Ltd.

Near infrared spectroscopy and multivariate calibration for simultaneous determination of glucose, triglycerides and high-density lipoprotein in animal plasma

The quantitative analysis of glucose, triglycerides and high-density lipoprotein (HDL) in rat plasma without sample pre-treatment using direct near-infrared spectroscopy was studied. Comparison was made of several multivariate calibration techniques and algorithms for data pre-processing and variable selection, including partial least squares (PLS), interval partial least squares (iPLS), genetic algorithm (GA) and successive projections algorithm (SPA). Variable selection yielded good results for the correlation coefficient and Root Mean Square Error of Prediction (RMSEP) values for the three parameters, especially triglycerides. The RMSEP values for glucose, triglycerides and HDL produced by the PLS model were 6.08, 16.07 and 2.03 mg dl(-1), respectively. F tests and t-tests were performed to compare the results of the models with each other and with a reference method. These results suggests that the PLS method can be used to simultaneously determine the concentrations of glucose, triglycerides and HDL in complicated biological fluids with NIR spectroscopy, offering an alternative analysis in animals.

Determination of octane numbers in gasoline by distillation curves and partial least squares regression

In this work, distillation curves (ASTM D86) associated with multivariate calibration (PLS-Partial Least Squares) were used to predict MON (motor octane number) and RON (research octane number) parameters of automotive fuel RMSEC getting values (Root Mean Square Error of Calibration) and RMSEP (Root Mean Square Error of Prediction) smaller than the literature. RMSEC values obtained were 0.051 and 0.078, and RMSEP values were 0.063 and 0.085 for MON and RON, respectively. These low values along with high accuracy, when compared to standard test methods (ASTM D2700 and ASTM D2999), indicated that the PLS models were efficient to predict MON and RON values and can be used as an alternative to control quality for automotive gasoline.

Variable selection in visible and near-infrared spectra: Application to on-line determination of sugar content in pears

Informative variable (or wavelength) selection plays an important role in quantitative analysis by visible and near-infrared (Vis–NIR) spectroscopy. Four different variable selection methods, namely, stepwise multiple linear regression (SMLR), genetic algorithm-partial least squares regression (GA-PLS), interval PLS (iPLS), and successive projection algorithm-multiple linear regression combined with GA (GA-SPA-MLR), were studied to determine the sugar content of pears. The results derived by these techniques were then compared. The calibration model built using GA-SPA-MLR on 18 selected wavelengths (2% of the total number of variables) exhibited higher coefficient of determination (R2)=0.880 and root mean square error of prediction (RMSEP)=0.459°Brix for the validation set. Results show that the accuracy of the quantitative analysis conducted by Vis–NIR spectroscopy can be improved through appropriate wavelength selection. Despite the RMSEP value of GA-SPA-MLR was a slightly higher than that of GA-PLS, considering that this model was simpler and easier to interpret, GA-SPA-MLR can be used for industrial applications.

Calibration of visible and near infrared spectroscopy for soil analysis at the field scale on three European farms

The development of accurate calibration models for selected soil properties is a crucial prerequisite for successful implementation of visible and near infrared (Vis-NIR) spectroscopy for soil analysis. This paper compares the performance of calibration models developed for individual farms with that of general models valid for three farms in three European countries. Fresh soil samples collected from farms in the Czech Republic, Germany and Denmark were scanned with a fibre-type Vis-NIR spectrophotometer. After dividing spectra into calibration (70%) and validation (30%) sets, spectra in the calibration set were subjected to partial least squares regression (PLSR) with leave-one-out cross-validation to establish calibration models of soil properties. Except for the Czech Republic farm, individual farm models provided successful calibration for total carbon (TC), total nitrogen (TN) and organic carbon (OC), with coefficients of determination (R2) of 0.85–0.93 and 0.74–0.96 and residual prediction deviations (RPD) of 2.61–3.96 and 2.00–4.95 for the cross-validation and independent validation respectively. General calibration models gave improved prediction accuracies compared with models of farms in the Czech Republic and Germany, which was attributed to larger ranges in the variation of soil properties in general models compared with those in individual farm models. The results revealed that larger standard deviations (SDs) and wider variation ranges have resulted in larger R2 and RPD, but also larger root mean square errors of prediction (RMSEP). Therefore, a compromise solution, which also results in small RMSEP values, should be found when selecting soil samples for Vis-NIR calibration to cover a wide variation range.

Multivariate Calibration and Instrument Standardization for the Rapid Detection of Diethylene Glycol in Glycerin by Raman Spectroscopy

The transfer of a multivariate calibration model for quantitative determination of diethylene glycol (DEG) contaminant in pharmaceutical-grade glycerin between five portable Raman spectrometers was accomplished using piecewise direct standardization (PDS). The calibration set was developed using a multi-range ternary mixture design with successively reduced impurity concentration ranges. It was found that optimal selection of calibration transfer standards using the Kennard-Stone algorithm also required application of the algorithm to multiple successively reduced impurity concentration ranges. Partial least squares (PLS) calibration models were developed using the calibration set measured independently on each of the five spectrometers. The performance of the models was evaluated based on the root mean square error of prediction (RMSEP), calculated using independent validation samples. An F-test showed that no statistical differences in the variances were observed between models developed on different instruments. Direct cross-instrument prediction without standardization was performed between a single primary instrument and each of the four secondary instruments to evaluate the robustness of the primary instrument calibration model. Significant increases in the RMSEP values for the secondary instruments were observed due to instrument variability. Application of piecewise direct standardization using the optimal calibration transfer subset resulted in the lowest values of RMSEP for the secondary instruments. Using the optimal calibration transfer subset, an optimized calibration model was developed using a subset of the original calibration set, resulting in a DEG detection limit of 0.32% across all five instruments.

Determination of biodiesel adulteration with raw vegetable oil from ATR-FTIR data using chemometric tools

Three different biodiesel sources (cotton, castor and palm) were adulterated with raw soybean oil at concentrations ranging from 1-40% (m/m). These samples were analyzed by infrared spectrometry (MIR) and their spectra were studied at three different spectral ranges: full spectrum (4000-665 cm-1), and the spectral ranges of 1800-1700 cm-1 and 1800-1000 cm-1. To determine the source of biodiesel used in the adulterated system, the data were analyzed by PCA (principal components analysis) spectral analysis tool and the best segregation of the sources was obtained at the range of 4000-665 cm-1. The explained variance was of 99% for the first three components. To quantify the raw soybean oil partial least squares (PLS) tool was applied. The best results were obtained for the spectral range of 1800-1000 cm-1, with values of RMSEP (root mean square error of prediction) varying from 1.10 to 1.47% (m/m).