Jin Young Kim
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Jin Young Kim has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Byung Hee Kim
Department of Food and Nutrition Sookmyung Women's University Seoul Republic of Korea
2 shared publications - Sangdoo Ahn
Department of Chemistry, Chung-Ang University, Seoul 06974, Republic of Korea
2 shared publications - Seon Yeong Lee
Department of Chemistry Chung‐Ang University Seoul Republic of Korea
2 shared publications - Hyeona Lim
Department of Chemistry Chung‐Ang University Seoul Republic of Korea
1 shared publication - Hyeonjin Kim
Department of Medical Sciences Seoul National University Seoul Republic of Korea
1 shared publication - Hyojin Cho
Department of Chemistry Chung‐Ang University Seoul Republic of Korea
1 shared publication
Latest Publications
- Open Access
- 31 July 2025
- Published by Wiley in Bulletin of the Korean Chemical Society
Discrimination of omega‐3 fatty acid oil forms by combining NMR spectroscopy with artificial intelligence
This study presents an approach for discriminating omega‐3 fatty acid forms using proton nuclear magnetic resonance (1H‐NMR) spectroscopy combined with machine learning and deep learning techniques. A total of 90 samples, comprising triglyceride, re‐esterified triglyceride, and ethyl ester forms, were analyzed. Principal component analysis–linear discriminant analysis, support vector machine (SVM), artificial neural network (ANN), and one‐dimensional convolutional neural network (1D CNN) models were applied using binned spectral data. In contrast, a two‐dimensional convolutional neural network (2D CNN) was constructed using spectral images. To prevent overfitting and optimize model hyperparameters, early stopping, cross‐validation, and Bayesian optimization were used across the different machine learning and deep learning models. The 1D and 2D CNN models both achieved 100% accuracy on the training and test sets, while the SVM and ANN models yielded slightly lower but still excellent performance, with a test accuracy of 94.4%. Model interpretability was enhanced through SHapley Additive exPlanations and Gradient‐weighted Class Activation Mapping, which identified critical spectral regions associated with classification decisions. These results demonstrate that the integration of artificial intelligence techniques with 1H‐NMR spectroscopy enables accurate, interpretable discrimination of omega‐3 fatty acid forms, offering a promising strategy for supplement authentication and quality control.
- 31 July 2025
- Published by Wiley in Bulletin of the Korean Chemical Society
Open AccessDiscrimination of omega‐3 fatty acid oil forms by combining NMR spectroscopy with artificial intelligence
This study presents an approach for discriminating omega‐3 fatty acid forms using proton nuclear magnetic resonance (1H‐NMR) spectroscopy combined with machine learning and deep learning techniques. A total of 90 samples, comprising triglyceride, re‐esterified triglyceride, and ethyl ester forms, were analyzed. Principal component analysis–linear discriminant analysis, support vector machine (SVM), artificial neural network (ANN), and one‐dimensional convolutional neural network (1D CNN) models were applied using binned spectral data. In contrast, a two‐dimensional convolutional neural network (2D CNN) was constructed using spectral images. To prevent overfitting and optimize model hyperparameters, early stopping, cross‐validation, and Bayesian optimization were used across the different machine learning and deep learning models. The 1D and 2D CNN models both achieved 100% accuracy on the training and test sets, while the SVM and ANN models yielded slightly lower but still excellent performance, with a test accuracy of 94.4%. Model interpretability was enhanced through SHapley Additive exPlanations and Gradient‐weighted Class Activation Mapping, which identified critical spectral regions associated with classification decisions. These results demonstrate that the integration of artificial intelligence techniques with 1H‐NMR spectroscopy enables accurate, interpretable discrimination of omega‐3 fatty acid forms, offering a promising strategy for supplement authentication and quality control.
- 23 December 2024
- Published by Wiley in Bulletin of the Korean Chemical Society
- Vol. 46 (2) , 131-137
Comparison of machine learning models for classifying edible oils using Fourier‐transform infrared spectroscopy
Accurate classification and authentication of edible oils are essential for maintaining product quality, ensuring consumer safety, and preserving market integrity. Therefore, this study aims to propose Fourier‐transform infrared (FT‐IR) spectroscopy, combined with advanced machine learning models, as a rapid and non‐destructive technique for classifying edible oils. The FT‐IR spectra of seven edible oil types were analyzed across three spectral regions: the full range, the C‐H stretching range, and the fingerprint region. Both absorbance and second derivative spectra were used to evaluate the influence of spectral preprocessing on classification accuracy. Six machine learning models—principal component analysis followed by linear discriminant analysis (PCA‐LDA), k‐nearest neighbors, decision tree, random forest, eXtreme Gradient Boosting, and support vector machines (SVM)—were employed to classify the oils, achieving training accuracies of 96.4%–100% and testing accuracies of 88.1%–100%. The second derivative spectra enhanced model performance by improving the resolution of overlapping peaks, particularly in the CH and CO stretching regions. Additionally, the SHapley Additive exPlanations analysis further revealed the most critical spectral features influencing model predictions, offering valuable insights into the decision‐making processes. This study demonstrates the effectiveness of combining FT‐IR spectroscopy, second derivative preprocessing, and machine learning techniques for classifying edible oils. The findings highlight the benefits of second derivative spectra in enhancing spectral resolution and the superior classification performance of PCA‐LDA and SVM models. These results offer a robust framework for advancing edible oil analysis and emphasize the potential of artificial intelligence in food authentication and quality control.
- 23 December 2024
- Published by Wiley in Bulletin of the Korean Chemical Society
- Vol. 46 (2) , 131-137
Comparison of machine learning models for classifying edible oils using Fourier‐transform infrared spectroscopy
Accurate classification and authentication of edible oils are essential for maintaining product quality, ensuring consumer safety, and preserving market integrity. Therefore, this study aims to propose Fourier‐transform infrared (FT‐IR) spectroscopy, combined with advanced machine learning models, as a rapid and non‐destructive technique for classifying edible oils. The FT‐IR spectra of seven edible oil types were analyzed across three spectral regions: the full range, the C‐H stretching range, and the fingerprint region. Both absorbance and second derivative spectra were used to evaluate the influence of spectral preprocessing on classification accuracy. Six machine learning models—principal component analysis followed by linear discriminant analysis (PCA‐LDA), k‐nearest neighbors, decision tree, random forest, eXtreme Gradient Boosting, and support vector machines (SVM)—were employed to classify the oils, achieving training accuracies of 96.4%–100% and testing accuracies of 88.1%–100%. The second derivative spectra enhanced model performance by improving the resolution of overlapping peaks, particularly in the CH and CO stretching regions. Additionally, the SHapley Additive exPlanations analysis further revealed the most critical spectral features influencing model predictions, offering valuable insights into the decision‐making processes. This study demonstrates the effectiveness of combining FT‐IR spectroscopy, second derivative preprocessing, and machine learning techniques for classifying edible oils. The findings highlight the benefits of second derivative spectra in enhancing spectral resolution and the superior classification performance of PCA‐LDA and SVM models. These results offer a robust framework for advancing edible oil analysis and emphasize the potential of artificial intelligence in food authentication and quality control.