1H NMR-based machine learning methods for rapid authentication and composition profiling of crude palm oil
| dc.contributor.author | Nggofur, Abdul | |
| dc.contributor.author | Sueviriyapan, Natthapong | |
| dc.contributor.author | Nuntawong, Noppadon | |
| dc.contributor.author | Sutthiumporn, Ketsada | |
| dc.contributor.author | Sooknoi, Tawan | |
| dc.contributor.author | Chirachanchai, Suwabun | |
| dc.contributor.author | Jongpatiwut, Siriporn | |
| dc.date.accessioned | 2026-08-06T10:56:20Z | |
| dc.date.available | 2026-08-06T10:56:20Z | |
| dc.date.issued | 2026-09-01 | |
| dc.description.abstract | A rapid analytical workflow for determining geographical origin and predicting fatty acid composition of crude palm oil (CPO) was developed using <sup>1</sup>H NMR, GC-FID, and machine learning. Analyzing CPO samples from Indonesia, Malaysia, the Philippines, and Thailand using unsupervised fingerprinting with principal component analysis (PCA), t-distributed stochastic neighbor embedding (t-SNE), and uniform manifold approximation and projection (UMAP) revealed partial origin-based grouping. Supervised classification, validated via leave-one-out cross-validation (LOOCV) and uncertainty quantification (UQ), reliably discriminated the origins above random chance. Additionally, partial least squares regression (PLSR) accurately predicted oleic, linoleic and myristic acid levels measured by GC-FID, whereas the accuracy decreased for lauric, stearic and palmitic acids. PLSR reliability was rigorously validated using latent variable selection and permutation testing to rule out random correlations. Overall, this integrated <sup>1</sup>H NMR and machine learning approach offers a rapid tool for CPO geographical traceability and compositional evaluation, demonstrating its potential for industrial quality control. | |
| dc.identifier.citation | Food Chemistry, 522, 2026 | |
| dc.identifier.doi | 10.1016/j.foodchem.2026.150052 | |
| dc.identifier.issn | 03088146 | |
| dc.identifier.other | 2-s2.0-105042425057 | |
| dc.identifier.uri | https://dspace.kmitl.ac.th/handle/123456789/18309 | |
| dc.source | Food Chemistry | |
| dc.subject | 1H NMR | |
| dc.subject | Crude palm oil (CPO) | |
| dc.subject | Fatty acids composition | |
| dc.subject | GC-FID | |
| dc.subject | Geographical traceability | |
| dc.subject | Machine learning | |
| dc.title | 1H NMR-based machine learning methods for rapid authentication and composition profiling of crude palm oil | |
| dc.type | Article |