Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
June 7, 2026Food and Bioprocess TechnologyOpen Access

XGBoost and LightGBM for Spectral Authentication of Pomegranate Juice Concentrate: A Comparison with Classical Chemometric Models

View Full Paper
Ask AI
Bookmark
Share

Authors

IYIrem YasaEIElif IlgenÇÇÇağrı Çavdaroğlu

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates advanced spectroscopy methods for detecting juice adulteration, highlighting effective models.

Key Points

  • This research aims to compare advanced algorithms and classical chemometric models in detecting pomegranate juice concentrate adulteration.
  • Analyzed 166 authentic and adulterated pomegranate juice concentrate samples.
  • Utilized FTIR, UV–Vis, and fluorescence spectroscopy for detection of adulterants.
  • Applied gradient boosting algorithms (XGBoost, LightGBM) and traditional chemometric models (PLS-DA, OPLS-DA, PLS, OPLS).
  • FTIR-based models achieved classification rates above 97% for detecting adulteration.
  • XGBoost and FTIR models showed R2 values above 0.97 for predicting adulterant levels.
  • UV–Vis and fluorescence models performed moderately, with fluorescence displaying the most variability.

Cite This Study

Yasa et al. (2026) studied this question.

synapsesocial.com/papers/6a2509de7def13d035e1a357https://doi.org/10.1007/s11947-026-04429-4
View Full Paper
Ask AI
Bookmark
Share