Date of Publication
8-2023
Document Type
Bachelor's Thesis
Degree Name
Bachelor of Science in Statistics Major in Actuarial Science
Subject Categories
Statistics and Probability
College
College of Science
Department/Unit
Mathematics and Statistics Department
Thesis Advisor
Rechel G. Arcilla
Defense Panel Chair
Angelo M. Alberto
Defense Panel Member
Karl Man Collado
Abstract/Summary
Coffee and cacao beans are considered as cash crops in the Philippines; however, they are at high risk of product fraud to increase profitability. Principal Component Analysis (PCA) and Partial Least Square-Discriminant Analysis (PLS-DA) address these issues since they can discriminate the harvesting regions of these beans that will allow their authentication. With limited samples gathered from different regions of the Philippines, both multivariate techniques were applied to multi-elements and Carbon-13 (δ13C) and Nitrogen-15 (δ15N) data obtained using X-Ray Fluorescence (XRF) and Isotope-Ratio Mass Spectrometry (IRMS), respectively. Using Parallel Analysis (PA), results show that two PCs are enough to represent the data obtained using (a) XRF only, (b) XRF and δ13C, (c) XRF and δ15N, (d) XRF and IRMS cacao, while three PCs are needed for data obtained using XRF and IRMS methods for coffee. Both PLS1-DA and PLS2-DA have the best model using raw data for both coffee and cacao beans. This indicates that PC scores are not good predictors for the model. Elements K, Mn, Rb, Sr, P, Cu, and δ13C isotope significantly discriminate Regions XI, X, IV-A, VII from Regions CAR, VI, II, XII, XIII, IX, I for coffee since it has the highest 𝑅𝑌2 and 𝑄𝑌2 of 0.8960 and 0.7060, respectively. For cacao, the elements Bi, Hg, Pb, U, Mn, Zn, Ni, and δ15N isotope significantly discriminate Regions XI, X, IX from Regions V, VII, BARMM, IV-A, XII, II with the highest 𝑅𝑌2 and 𝑄𝑌2 of 0.6410 and 0.5560, respectively.
Abstract Format
html
Language
English
Format
Electronic
Keywords
Cacao beans--Philippines; Coffee
Recommended Citation
Tan, R. T., Cano, S., & Chua, J. C. (2023). Discriminating the harvesting regions of Philippine coffee and cacao beans using principal component analysis (PCA) and Partial Least Squares - Discriminant Analysis (PLS-DA). Retrieved from https://animorepository.dlsu.edu.ph/etdb_math/29
Upload Full Text
wf_yes
Embargo Period
8-15-2023