Date of Publication
8-2025
Document Type
Bachelor's Thesis
Degree Name
Bachelor of Science in Chemistry
Subject Categories
Chemistry
College
College of Science
Department/Unit
Chemistry
Thesis Advisor
Emmanuel V. Garcia
Defense Panel Member
Mariafe N. Calingacion
Joan Candice V. Ondevilla
Abstract/Summary
The Philippine coffee industry faces several challenges when it comes to the authenticity of the coffee products being sold in the market. The lack of an accessible geographical identification tool hinders the industry in ensuring that the products being sold are authentic based on their geographical origin. With that, the study aimed to determine the multi-elemental composition of local coffee beans through the use of X-Ray Fluorescence spectroscopy in order to identify its geographic origin using statistical methods and machine learning. The study analyzed the 46 samples of green coffee beans grown in the Philippines, specifically in Regions VI, X, XI, and XII. The samples collected were from the annual Philippine Coffee Quality Competition held in 2024. The data obtained was analyzed using One-Way Analysis of Variance, Post Hoc Tukey’s HSD test, and Welch’s two-sample t-test. With that, it was found that the elements Manganese and Strontium were highly significant elements in regional classification. While Potassium, Manganese, and Copper were the highly significant elements in the varietal classification. The Random Forest model exhibited a 80.85% and 95.74% accuracy for the regional and varietal classification, respectively. With that, the method of integrating multi-elemental analysis with machine learning was successful in determining the geographical and varietal origin of the coffee samples, providing a non-invasive method for coffee classification.
Abstract Format
html
Language
English
Format
Electronic
Keywords
Coffee--Philippines; X-ray spectroscopy; Machine learning
Recommended Citation
Alemania, A. A., & Aranton, E. K. (2025). Regional and varietal traceability of arabica (C. arabica) and robusta (C. canephora) Philippine green coffee beans using x-ray fluorescence spectroscopy based elemental profiling with machine learning. Retrieved from https://animorepository.dlsu.edu.ph/etdb_chem/63
Upload Full Text
wf_yes
Embargo Period
9-28-2027
