Adulteration detection in cactus seed oil: Integrating analytical chemistry and machine learning approaches
Economically motivated adulteration threatens both consumer rights and market integrity, particularly with high-value cold-pressed oils like cactus seed oil (CO). This study proposes a machine learning model that integrates analytical measurements, data simulations, and classification techniques to...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Current Research in Food Science |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2665927125000176 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832576451286663168 |
|---|---|
| author | Said El Harkaoui Cristina Ortiz Cruz Aaron Roggenland Micha Schneider Sascha Rohn Stephan Drusch Bertrand Matthäus |
| author_facet | Said El Harkaoui Cristina Ortiz Cruz Aaron Roggenland Micha Schneider Sascha Rohn Stephan Drusch Bertrand Matthäus |
| author_sort | Said El Harkaoui |
| collection | DOAJ |
| description | Economically motivated adulteration threatens both consumer rights and market integrity, particularly with high-value cold-pressed oils like cactus seed oil (CO). This study proposes a machine learning model that integrates analytical measurements, data simulations, and classification techniques to detect adulteration of CO with refined sunflower oil (SO) and determine the detectable limit of adulteration without measuring a huge number of different mixtures. First, pure CO and SO samples were analyzed for their fatty acid, triacylglycerol, and tocochromanol content using HPLC or GC. The resulting oil composition data served as the foundation for further simulations. Monte Carlo (MC) simulations outperformed Conditional Tabular Generative Adversarial Networks (CTGAN) in simulating realistic oil compositions, with MC yielding lower Kullback-Leibler Divergence values compared to CTGAN. The MC-simulated data were then used to simulate larger datasets, a critical step for training and testing two classification models: Random Forest (RF) and Neural Networks (NN), as robust training cannot be achieved with small sample sizes. Both models achieved good classification accuracies, with RF achieving higher accuracy than NN, reaching 94% on simulated datasets and 90% on real-world samples with detectable adulteration levels as low as 1%. RF also offers better interpretability and is computational less demanding as compared to NN which makes it advantageous for authenticity verification in this study. Therefore, combining MC simulation with RF as a robust method for detecting CO adulteration is proposed. The proposed method, coded in Python and available as open-source, offers a flexible framework for continuous adaptation with new data. |
| format | Article |
| id | doaj-art-fd340d4ca1c246baa08f07ca3b02c819 |
| institution | Kabale University |
| issn | 2665-9271 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Current Research in Food Science |
| spelling | doaj-art-fd340d4ca1c246baa08f07ca3b02c8192025-01-31T05:12:23ZengElsevierCurrent Research in Food Science2665-92712025-01-0110100986Adulteration detection in cactus seed oil: Integrating analytical chemistry and machine learning approachesSaid El Harkaoui0Cristina Ortiz Cruz1Aaron Roggenland2Micha Schneider3Sascha Rohn4Stephan Drusch5Bertrand Matthäus6Max Rubner-Institut, Federal Research Institute for Nutrition and Food, Department for Safety and Quality of Cereals, Schützenberg 12, 32756, Detmold, Germany; Department of Food Chemistry and Analysis, Institute of Food Technology and Food Chemistry, Technische Universität Berlin, Berlin, Germany; Department of Food Technology and Food Material Science, Institute of Food Technology and Food Chemistry, Technische Universität Berlin, Berlin, Germany; Corresponding author. Max Rubner-Institut, Federal Research Institute for Nutrition and Food, Department for Safety and Quality of Cereals, Schützenberg 12, 32756, Detmold, Germany.Max Rubner-Institut, Federal Research Institute for Nutrition and Food, Zentralabteilung, Haid-und-Neu-Str. 9, 76131, Karlsruhe, Germany; BMEL Project KIDA, AI consultancy, GermanyMax Rubner-Institut, Federal Research Institute for Nutrition and Food, Zentralabteilung, Schützenberg 12, 32756, Detmold, Germany; BMEL Project KIDA, AI consultancy, GermanyJohann Heinrich von Thünen Institute - Federal Research Institute for Rural Areas, Forestry and Fisheries, Bundesallee 50, 38116, Braunschweig, Germany; BMEL Project KIDA, AI consultancy, GermanyDepartment of Food Chemistry and Analysis, Institute of Food Technology and Food Chemistry, Technische Universität Berlin, Berlin, GermanyDepartment of Food Technology and Food Material Science, Institute of Food Technology and Food Chemistry, Technische Universität Berlin, Berlin, GermanyMax Rubner-Institut, Federal Research Institute for Nutrition and Food, Department for Safety and Quality of Cereals, Schützenberg 12, 32756, Detmold, GermanyEconomically motivated adulteration threatens both consumer rights and market integrity, particularly with high-value cold-pressed oils like cactus seed oil (CO). This study proposes a machine learning model that integrates analytical measurements, data simulations, and classification techniques to detect adulteration of CO with refined sunflower oil (SO) and determine the detectable limit of adulteration without measuring a huge number of different mixtures. First, pure CO and SO samples were analyzed for their fatty acid, triacylglycerol, and tocochromanol content using HPLC or GC. The resulting oil composition data served as the foundation for further simulations. Monte Carlo (MC) simulations outperformed Conditional Tabular Generative Adversarial Networks (CTGAN) in simulating realistic oil compositions, with MC yielding lower Kullback-Leibler Divergence values compared to CTGAN. The MC-simulated data were then used to simulate larger datasets, a critical step for training and testing two classification models: Random Forest (RF) and Neural Networks (NN), as robust training cannot be achieved with small sample sizes. Both models achieved good classification accuracies, with RF achieving higher accuracy than NN, reaching 94% on simulated datasets and 90% on real-world samples with detectable adulteration levels as low as 1%. RF also offers better interpretability and is computational less demanding as compared to NN which makes it advantageous for authenticity verification in this study. Therefore, combining MC simulation with RF as a robust method for detecting CO adulteration is proposed. The proposed method, coded in Python and available as open-source, offers a flexible framework for continuous adaptation with new data.http://www.sciencedirect.com/science/article/pii/S2665927125000176Cactus seed oilAuthenticityMachine learningConditional generative adversarial networkMonte-CarloRandom Forest |
| spellingShingle | Said El Harkaoui Cristina Ortiz Cruz Aaron Roggenland Micha Schneider Sascha Rohn Stephan Drusch Bertrand Matthäus Adulteration detection in cactus seed oil: Integrating analytical chemistry and machine learning approaches Current Research in Food Science Cactus seed oil Authenticity Machine learning Conditional generative adversarial network Monte-Carlo Random Forest |
| title | Adulteration detection in cactus seed oil: Integrating analytical chemistry and machine learning approaches |
| title_full | Adulteration detection in cactus seed oil: Integrating analytical chemistry and machine learning approaches |
| title_fullStr | Adulteration detection in cactus seed oil: Integrating analytical chemistry and machine learning approaches |
| title_full_unstemmed | Adulteration detection in cactus seed oil: Integrating analytical chemistry and machine learning approaches |
| title_short | Adulteration detection in cactus seed oil: Integrating analytical chemistry and machine learning approaches |
| title_sort | adulteration detection in cactus seed oil integrating analytical chemistry and machine learning approaches |
| topic | Cactus seed oil Authenticity Machine learning Conditional generative adversarial network Monte-Carlo Random Forest |
| url | http://www.sciencedirect.com/science/article/pii/S2665927125000176 |
| work_keys_str_mv | AT saidelharkaoui adulterationdetectionincactusseedoilintegratinganalyticalchemistryandmachinelearningapproaches AT cristinaortizcruz adulterationdetectionincactusseedoilintegratinganalyticalchemistryandmachinelearningapproaches AT aaronroggenland adulterationdetectionincactusseedoilintegratinganalyticalchemistryandmachinelearningapproaches AT michaschneider adulterationdetectionincactusseedoilintegratinganalyticalchemistryandmachinelearningapproaches AT sascharohn adulterationdetectionincactusseedoilintegratinganalyticalchemistryandmachinelearningapproaches AT stephandrusch adulterationdetectionincactusseedoilintegratinganalyticalchemistryandmachinelearningapproaches AT bertrandmatthaus adulterationdetectionincactusseedoilintegratinganalyticalchemistryandmachinelearningapproaches |