Decision Tree and K-Means Analysis of Raman Spectra for Edible Oils: A Physics-Informed AI Approach
Researchers have developed an artificial intelligence approach to identify edible oils using Raman spectroscopy. They used a combination of machine learning algorithms and physics-based methods to analyze the spectra of five different edible oils in pure form and within a fried potato chip matrix. The study found that decision trees achieved 100% accuracy for pure oils, while a more complex method improved classification for samples contaminated with food particles. The appro
Researchers have developed an artificial intelligence approach to identify edible oils using Raman spectroscopy. They used a combination of machine learning algorithms and physics-based methods to analyze the spectra of five different edible oils in pure form and within a fried potato chip matrix. The study found that decision trees achieved 100% accuracy for pure oils, while a more complex method improved classification for samples contaminated with food particles. The approach reduces the amount of data needed for analysis by 99.44% without losing accuracy.
---
Why it matters: This work matters to AI researchers because it demonstrates how physics-informed artificial intelligence can be used to develop compact and interpretable models that are suitable for edge computing and portable sensing applications.
Source: https://arxiv.org/abs/2608.20440
This article was originally published at: https://arxiv.org/abs/2608.20440