Near-infrared spectroscopy coupled with machine learning algorithms based on L1-norm and L21-norm to identify the geographical origins of Chinese wolfberry.
Gespeichert in:
| Titel: | Near-infrared spectroscopy coupled with machine learning algorithms based on L1-norm and L21-norm to identify the geographical origins of Chinese wolfberry. |
|---|---|
| Autoren: | Zhu X; School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, China., Wu X; School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, China; High-tech Key Laboratory of Agricultural Equipment and Intelligence of Jiangsu Province, Jiangsu University, Zhenjiang 212013, China. Electronic address: wxh419@ujs.edu.cn., Shen J; School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, China., Sun J; School of Electrical and Information Engineering, Jiangsu University, Zhenjiang 212013, China., Wu B; Department of Information Engineering, Chuzhou Polytechnic, Chuzhou 239000, China; School of Computer Science and Engineering, Southeast University, Nanjing 211189, China. Electronic address: wubin@chzc.edu.cn. |
| Quelle: | Food chemistry [Food Chem] 2025 Nov 30; Vol. 493 (Pt 2), pp. 145863. Date of Electronic Publication: 2025 Aug 07. |
| Publikationsart: | Journal Article; Evaluation Study |
| Sprache: | English |
| Info zur Zeitschrift: | Publisher: Elsevier Applied Science Publishers Country of Publication: England NLM ID: 7702639 Publication Model: Print-Electronic Cited Medium: Internet ISSN: 1873-7072 (Electronic) Linking ISSN: 03088146 NLM ISO Abbreviation: Food Chem Subsets: MEDLINE |
| Imprint Name(s): | Publication: Barking : Elsevier Applied Science Publishers Original Publication: Barking, Eng., Applied Science Publishers. |
| MeSH-Schlagworte: | Lycium*/chemistry , Lycium*/classification , Machine Learning* , Fruit*/chemistry , Fruit*/classification, Spectroscopy, Near-Infrared/methods ; China ; Discriminant Analysis ; Algorithms ; Geography ; Support Vector Machine |
| Abstract: | Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. The nutritional value of Chinese wolfberry varies depending on different geographical origins, even at the regional level. Therefore, a non-destructive and effective method has important implications for identifying the geographical origins of Chinese wolfberry. To address the issue of traditional feature extraction methods being sensitive to outliers and lacking robustness in processing NIR spectral data, inspired by L21-norm based linear discriminant analysis (L21-LDA), this study proposed an L21-norm based robust linear discriminant analysis (L21-RLDA), which was compared with traditional LDA, L21-LDA and L1-NSLDA. Finally, K-nearest neighbor (KNN) and support vector machine (SVM) were applied for classification. L1-NSLDA, L21-LDA and L21-RLDA all achieved significant results with high classification accuracies. Moreover, the proposed L21-RLDA reached the highest classification accuracy of 99.3 %. This study demonstrates the feasibility of NIR spectroscopy in combination with feature extraction methods based on L1-norm and L21-norm to discriminate Chinese wolfberry origins rapidly, accurately, and robustly. (Copyright © 2025 Elsevier Ltd. All rights reserved.) |
| Contributed Indexing: | Keywords: Chinese wolfberry; Feature extraction; L21-norm; Near-infrared spectroscopy; Origin traceability |
| Entry Date(s): | Date Created: 20250812 Date Completed: 20250915 Latest Revision: 20250915 |
| Update Code: | 20250916 |
| DOI: | 10.1016/j.foodchem.2025.145863 |
| PMID: | 40795554 |
| Datenbank: | MEDLINE |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstract: | Competing Interests: Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.<br />The nutritional value of Chinese wolfberry varies depending on different geographical origins, even at the regional level. Therefore, a non-destructive and effective method has important implications for identifying the geographical origins of Chinese wolfberry. To address the issue of traditional feature extraction methods being sensitive to outliers and lacking robustness in processing NIR spectral data, inspired by L21-norm based linear discriminant analysis (L21-LDA), this study proposed an L21-norm based robust linear discriminant analysis (L21-RLDA), which was compared with traditional LDA, L21-LDA and L1-NSLDA. Finally, K-nearest neighbor (KNN) and support vector machine (SVM) were applied for classification. L1-NSLDA, L21-LDA and L21-RLDA all achieved significant results with high classification accuracies. Moreover, the proposed L21-RLDA reached the highest classification accuracy of 99.3 %. This study demonstrates the feasibility of NIR spectroscopy in combination with feature extraction methods based on L1-norm and L21-norm to discriminate Chinese wolfberry origins rapidly, accurately, and robustly.<br /> (Copyright © 2025 Elsevier Ltd. All rights reserved.) |
|---|---|
| ISSN: | 1873-7072 |
| DOI: | 10.1016/j.foodchem.2025.145863 |