Scholarlink

<p>Edible oil adulteration in edible oils is a serious food safety issue that affects customer confidence, market integrity, and public health. In this study, the edible oils are collected from the market for adulteration detection. This study proposes a method of radio frequency (RF) spectroscopy combined with machine learning (ML) algorithms to accurately detect adulteration in edible oil. The mustard oil is treated as oil under test (OuT) and other edible oils include palm, soya refined (soy), olive considered as the adulterants under test in the OuT. The dielectric parameters gathered from the RF sensor is analysed using ML to detect the adulteration in the mustard oil. In this study, LightGBM and XGBoost models are adopted to correctly classify pure and adulterated oil samples. XGBoost achieved an overall accuracy of 88% while LightGBM achieved around 85% accuracy. The research result shows that RF when combined with ML has great potential in accurately detecting adulteration in edible oils.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Integration of RF Sensing and ML Algorithms for Enhanced Detection and Classification of Edible Oil

  • 期刊论文

  • 发表日期：

    2026年2月25日

查看全文
  • Nitika Dhingra,
  • Adish Garg

摘要

Edible oil adulteration in edible oils is a serious food safety issue that affects customer confidence, market integrity, and public health. In this study, the edible oils are collected from the market for adulteration detection. This study proposes a method of radio frequency (RF) spectroscopy combined with machine learning (ML) algorithms to accurately detect adulteration in edible oil. The mustard oil is treated as oil under test (OuT) and other edible oils include palm, soya refined (soy), olive considered as the adulterants under test in the OuT. The dielectric parameters gathered from the RF sensor is analysed using ML to detect the adulteration in the mustard oil. In this study, LightGBM and XGBoost models are adopted to correctly classify pure and adulterated oil samples. XGBoost achieved an overall accuracy of 88% while LightGBM achieved around 85% accuracy. The research result shows that RF when combined with ML has great potential in accurately detecting adulteration in edible oils.