Prediction of Caffeine Content in Java Preanger Coffee Beans by NIR Spectroscopy Using PLS and MLR Method

Budiastra, I Wayan; Sutrisno, Sutrisno; Widyotomo, Sukrisno; Ayu, P C

doi:10.1088/1755-1315/147/1/012004

Cited by 16 publications

(15 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The best result was obtained through the use of 10 selected wavelengths in Kubelka-Munk spectral data with rvalue of 0.82, consistency of 108%, RPD of 2.10 and CV of 2.14%. This result is better than the previous study done by Budiastra, Sutrisno, Widyotomo, & Ayu (2018), which used 10 selected wavelength with rvalue of 0.70, consistency of 90%, RPD of 1.21 and CV of 3.23% in predicting caffeine content in Java Preanger coffee bean. The use of coffee powder in predicting caffeine content showed good accuracy by MLR method.…”

Section: Calibration and Validation Of Liberica Green Bean Powdercontrasting

confidence: 68%

“…The caffeine content of Liberica coffee green beans powder found in this study ranged between 1.12% and 1.35% with the standard deviation of 0.055% for calibration and 0.021% for validation (Tabel 1). Meanwhile previous study conducted by Purningsih (2018) showed that caffeine content in Bondowoso Arabica coffee green bean powder was 1.27% -1.57% , Arabica Java Preanger coffee bean was 1.18 -1.41 % (Budiastra, Sutrisno, Widyotomo, & Ayu, 2018) and Arabica Gayo coffee beans was 0.194 to 0.328% (Rosita, Budiastra & Sutrisno, 2016). Hulupi & Martini (2013) stated that caffeine content of Liberica coffee green bean was 1.1% to 1.3%.…”

Section: Caffeine Content Of Liberica Coffee Green Bean Powdermentioning

confidence: 95%

See 1 more Smart Citation

Prediction of Caffeine Content in Liberica Coffee Green Bean by NIR Spectroscopy Using Kubelka-Munk Model

Kyaw¹,

Budiastra

Sutrisno

et al. 2020

J. tanam. industri dan penyegar

View full text Add to dashboard Cite

<p>Liberica is one of coffee species that is becoming popular and increasingly in demand in present days due to its unique characteristics. Caffeine is one of the important coffee quality parameter which determines the coffee flavor, consumer preference and market price. Caffeine content is usually analyzed by chemical method which is destructive, time consuming, expensive and involving a lot of procedures. NIR Spectroscopy is one of the non-destructive techniques to overcome these disadvantages. This study was conducted at the Department of Mechanical and Biosystem Engineering, IPB University for NIR measurement and the Center of Agro-based Industry (BBIA), Bogor for chemical analysis from August to November 2019. The study aimed to determine the best calibration model for the prediction of caffeine content in Liberica coffee green bean powder. In this study, FT-NIRS in the wavelength of 1000-2500 nm was used for NIR measurement and HPLC tool was used for chemical analysis. Kubelka-Munk (K/S) and Absorbance (Log 1/R) were used as data transformation, whereas Standard Normal Variance (SNV) and Second derivative of Savitzky-Golay (dg2) as data pretreatment. In addition, Partial Least Square (PLS) and Multiple Linear Regression (MLR) were applied for multivariate calibration method. The best calibration model for the prediction of caffeine content of Liberica coffee green bean powder was obtained by the spectral data pretreated with second derivative of Savitzky-Golay (dg2) and Kubelka-Munk data transformation using PLS calibration method with the results of r = 0.90, RPD = 2.24, CV = 2.01%.</p>

show abstract

Section: Calibration and Validation Of Liberica Green Bean Powdercontrasting

confidence: 68%

Section: Caffeine Content Of Liberica Coffee Green Bean Powdermentioning

confidence: 95%

Prediction of Caffeine Content in Liberica Coffee Green Bean by NIR Spectroscopy Using Kubelka-Munk Model

Kyaw¹,

Budiastra

Sutrisno

et al. 2020

J. tanam. industri dan penyegar

View full text Add to dashboard Cite

show abstract

“…Partial least squares regression (PLSR) is a classical and widely used linear method for modeling spectral data. Studies have shown that the prediction accuracies of PLS-derived models tend to be higher than that of multiple regression- (MR-) derived models [ 36 , 37 ]. Our previous study found that the PLSR model had excellent predictive and extrapolation abilities [ 19 , 38 ].…”

Section: Introductionmentioning

confidence: 99%

Rapid Geographical Origin Identification and Quality Assessment of Angelicae Sinensis Radix by FT-NIR Spectroscopy

Zhang

Wang

Yan

et al. 2021

Journal of Analytical Methods in Chemistry

View full text Add to dashboard Cite

Angelicae Sinensis Radix is a widely used traditional Chinese medicine and spice in China. The purpose of this study was to develop a methodology for geographical classification of Angelicae Sinensis Radix and determine the contents of ferulic acid and Z-ligustilide in the samples using near-infrared spectroscopy. A qualitative model was established to identify the geographical origin of Angelicae Sinensis Radix using Fourier transform near-infrared (FT-NIR) spectroscopy. Support vector machine (SVM) algorithms were used for the establishment of a qualitative model. The optimum SVM model had a recognition rate of 100% for the calibration set and 83.72% for the prediction set. In addition, a quantitative model was established to predict the content of ferulic acid and Z-ligustilide using FT-NIR. Partial least squares regression (PLSR) algorithms were used for the establishment of a quantitative model. Synergy interval-PLS (Si-PLS) was used to screen the characteristic spectral interval to obtain the best PLSR model. The coefficient of determination for calibration (R2C) for the best PLSR models established with the optimal spectral preprocessing method and selected important spectral regions for the quantitative determination of ferulic acid and Z-ligustilide was 0.9659 and 0.9611, respectively, while the coefficient of determination for prediction (R2P) was 0.9118 and 0.9206, respectively. The values of the ratio of prediction to deviation (RPD) of the two final optimized PLSR models were greater than 2. The results suggested that NIR spectroscopy combined with SVM and PLSR algorithms could be exploited in the discrimination of Angelicae Sinensis Radix from different geographical locations for quality assurance and monitoring. This study might serve as a reference for quality evaluation of agricultural, pharmaceutical, and food products.

show abstract

“…Due to the sudden outbreak of COVID-19, there is no prediction algorithm for the number of enterprises returning to work under this background, but many models such as multilayer perceptron (MLP) [1][2][3][4][5][6],multiple linear regression model(MLR) [7][8][9][10][11][12], support vector machine(SVM) [13][14] have shown excellent performance in the existing prediction and analysis problems.…”

Section: Related Researchmentioning

confidence: 99%

Research on the prediction algorithm about the number of workers returning to work based on the COVID-19

Gao

Zhai

Wang

et al. 2020

2020 IEEE International Conference on Advances in Electrical Engineering and Computer Applications( AEECA)

View full text Add to dashboard Cite

In order to make rational use of the resources about epidemic prevention such as masks, and prevent the leaders of enterprises from falsely reporting the number of workers back to work, the evaluation of the number of workers back to work in enterprises is transformed into the prediction of the number of workers back to work under different distribution about the number of workers back to work. Based on the analysis of the existing historical data, this paper predicts the number of people who return to work through intelligent algorithm, so that the government can prepare and distribute epidemic prevention materials. Based on the analysis of the daily power consumption data of the enterprise, combined with the existing number of enterprises returning to work, this paper constructs a prediction model of the number of enterprises returning to work based on the power consumption, and completes the prediction of the number of enterprises returning to work in the future. The experimental results based on the simulation data of the number of enterprises returning to work with different distributions show that the intelligent algorithm can effectively predict the number of enterprises returning to work under the background of COVID-19.

show abstract

Prediction of Caffeine Content in Java Preanger Coffee Beans by NIR Spectroscopy Using PLS and MLR Method

Cited by 16 publications

References 7 publications

Prediction of Caffeine Content in Liberica Coffee Green Bean by NIR Spectroscopy Using Kubelka-Munk Model

Prediction of Caffeine Content in Liberica Coffee Green Bean by NIR Spectroscopy Using Kubelka-Munk Model

Rapid Geographical Origin Identification and Quality Assessment of Angelicae Sinensis Radix by FT-NIR Spectroscopy

Research on the prediction algorithm about the number of workers returning to work based on the COVID-19

Contact Info

Product

Resources

About