Abstract
Presented manuscript described data analysis on near infrared spectroscopy used as adopted and portable technology for cocoa farmers in Aceh Province, Indonesia. The near infrared spectroscopy (NIRS) assisted farmers in post-harvest handling especially for cocoa quality evaluation. This technology was used to determine moisture content (MC) and fat content (FC) of intact cocoa bean samples rapidly and simultaneously. Near infrared spectra data were acquired as absorbance spectrum in wavelength range from 1000 to 2500 nm with co-added of 32 scans for a total of 72 intact bulk cocoa bean samples. Spectra data can be used to predict MC and FC of intact cocoa beans by establishing prediction models and validate with actual MC and FC measured by means of standard laboratory procedures. Prediction performances were evaluated using several statistical indicators: coefficient correlation (r), coefficient of determination (R2), root mean square error (RMSE) and residual predictive deviation (RPD) index. Near infrared spectra data can be enhanced using spectra pre-treatment methods to improve prediction performances. Moreover, prediction models can be developed using principal component regression (PCR), partial least squares regression (PLSR) and other regression approaches. Ideal prediction models should have r and R2 above 0.75, RPD index above 2.0 and RMSE lower than its standard deviation (SD). Dataset were available as raw MS Excel format and The Unscrambler files as *.unsb extension.
Highlights
Data analysis on near infrared spectroscopy as a part of technology adoption for cocoa farmer in Aceh Province, Indonesia
Spectra data can be used to predict moisture content (MC) and fat content (FC) of intact cocoa beans by establishing prediction models and validate with actual MC and FC measured by means of standard laboratory procedures
A total of 72 bulk intact cocoa bean amounted 50g per bulk were obtained from cocoa farmers in Pidie Jaya, Aceh Province, Indonesia
Summary
Near infrared spectral data of intact cocoa bean samples were acquired using a portable near infrared spectroscopy (FTIR PSD i15) in wavelength range from 1000 to 2500 nm. Cocoa beans were fermented for 3, 5 and 7 days in order to obtain different. Spectra data were recorded and saved in two different file formats as *.csv and *.spa. Sample labelling was required automatically before acquisition to differ cocoa bean samples. Predicted MC and FC of intact cocoa bean samples were obtained simultaneously by performing calibration models using principal component regression (PCR) and partial least squares regression (PLSR) followed by cross validation respectively. Data used for testing and validating models were obtained from calibration datasets from which cross validation method is employed [10]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
