Status : Verified
| Personal Name | Sabacan, Ethan Josiah T. |
|---|---|
| Resource Title | Development of a rapid and affordable machine learning-based screening technique for honey adulteration using UV-visible absorption spectra |
| Date Issued | 10 June 2025 |
| Abstract | Honey, a high-value food product, is a common target of food fraud through adulteration with low-cost sweeteners and ingredients. Up to 87% of all local honey products sold in the Philippines are adulterated with as much as 100% substitution. Adulterated honey, typically sold at a low price, potentially poses risks to consumer health. This study explored the feasibility of utilizing ultraviolet-visible (UV-Vis) spectrometry coupled with machine learning (ML) as a rapid and affordable screening technique for honey adulteration. Fifty-nine authentic Philippine honey samples from three bee species were analyzed in their unadulterated and adulterated forms with C3 and C4 syrups at 10% concentration. Exploratory data analysis showed the possibility of grouping samples based on the type of bee species rather than the adulteration status. A supervised machine learning model trained to predict the honey samples’ adulteration status showed 98% accuracy with 100% sensitivity. A grouping function added to the code prompted by a temporal data leakage noted in the initial model resulted in the corrected model with an accuracy of 34% and sensitivity of 34%. Possible issues in machine learning implementations in published journals for honey authentication were also explored because most claim to have a > 90% accuracy. The current state of the model does not perform complete discrimination between non-adulterated and adulterated samples. Further improvements to the study protocol and chemometrics are needed to demonstrate the applicability and reliability of an ML-based UV-Vis spectrometric method for honey authenticity screening. |
| Degree Course | Bachelor of Science in Food Technology |
| Language | English |
| Keyword | UV-Vis spectrometry; Philippine honey; Chemometrics; Food adulteration; machine learning; Authenticity screening; Honey adulteration; Discriminant model; Authentication |
| Material Type | Thesis/Dissertation |
Preliminary Pages
926.52 Kb
Category : P - Author wishes to publish the work personally.
Access Permission : Limited Access