Grade Identification of Strong-flavor Raw Liquor Based on GC-MS Combined with Spearman-KPCA Feature Extraction

Ni Fu; Xianguo Tuo; Wei Zhang

doi:10.54097/cwntjb65

Authors

Ni Fu
Xianguo Tuo
Wei Zhang

DOI:

https://doi.org/10.54097/cwntjb65

Keywords:

Strong-flavor base liquor; gas chromatography-mass spectrometry; Spearman's rank correlation coefficient; kernel principal component analysis; grade identification model.

Abstract

Taking different grades of strong-flavor raw liquor as the research object, gas chromatography-mass spectrometry (GC-MS) technology was used to obtain the volatile components mapping data of raw liquor, and Spearman correlation coefficient (Spearman) combined with principal component analysis (PCA) and kernel principal component analysis (KPCA) was used to realize the secondary feature extraction of the GC-MS data, and then combined with the support vector machine (SVM), extreme gradient boosting (XGBoost), and BP neural network to establish the raw liquor grades identification model, respectively. The results show that the prediction accuracy of the grade identification model based on Spearman-KPCA dimensionality reduction data is better, in which the Spearman-KPCA-BP neural network model has the best classification effect, and the accuracy of the correction set and prediction set reaches 99.44% and 96.10%, respectively. Research shows that the principal components extracted based on Spearman-KPCA secondary features can better characterize the characteristic information of different grades of raw liquor. Combined with the BP neural network model, it can effectively realize the identification of different grades of raw liquor. It is an effective method for identifying the grade of raw liquor.

Downloads

Download data is not yet available.

References

[1] SUN J, ZHAO D, ZHANG F, et al. Joint direct injection and GC–MS chemometric approach for chemical profile and sulfur compounds of sesame‑flavor Chinese Baijiu (Chinese liquor)[J/OL]. European Food Research and Technology, 2018, 244(1): 145-160.

[2] HONG J, TIAN W, ZHAO D. Research progress of trace components in sesame-aroma type of baijiu[J/OL]. Food Research International, 2020, 137: 109695.

[3] LIU Q R, ZHANG X J, ZHENG L, et al. Machine learning based age-authentication assisted by chemo-kinetics: Case study of strong-flavor Chinese Baijiu[J/OL]. Food Research International, 2023, 167: 112594.

[4] QIAN Y, ZHANG L, SUN Y, et al. Differentiation and classification of Chinese Luzhou‐flavor liquors with different geographical origins based on fingerprint and chemometric analysis[J/OL]. Journal of Food Science, 2021, 86(5): 1861-1877.

[5] Zhou Xuan. Research on volatile composition analysis and grade identification of base wine of strong aromatic liquor [D/OL]. Jiangsu University, 2019.

[6] CAMARA J S, MEDINA S, PERESTRELO R. Recent Developments in the Applications of Fingerprinting Technology in the Food Field [J/OL]. FOODS, 2022, 11(14): 2006.

[7] LIU Fei. Discussion on the application of gas chromatography-mass spectrometry in food analysis[J/OL]. Modern Food, 2020(11): 167-168.

[8] Han Yuncui. Aroma modeling and automated wine picking for strong-flavored base wines [D/OL]. Qilu University of Technology, 2023.

[9] Liu Qingru, Meng Lianjun, Zhang Xiaojuan, et al. Identification of storage time of Lu-type base wine based on GC-MS fingerprinting and XGBoost machine learning[J]. Food Science, 2022, 43(24): 310-317.

[10] QIAN Yu, HU Xue, SUN Yue, et al. Classification of strongly flavored liquor based on fingerprinting and chemometrics[J]. China Brewing, 2021, 40(6): 152-156.

[11] ZHU Kaixian, HU Xue, DENG Jing, et al. Discriminative analysis of different aromatic liquors based on GC-MS technology[J]. China Brewing, 2023, 42(1): 213-218.

[12] FAN Sanshuan, TANG Jie, LUO Xianxuan, et al. Classification of small-square clear-flavored wine based on HS-SPME-Arrow-GC-MS and chemometrics[J/OL]. Food and Fermentation Industry, 2021, 47(13): 254-260.

[13] LIU Y, QIAO Z, ZHAO Z, et al. Comprehensive evaluation of Luzhou-flavor liquor quality based on fuzzy mathematics and principal component analysis[J/OL]. FOOD SCIENCE & NUTRITION, 2022, 10(6): 1780-1788.

[14] YAO Y, MENG H, GAO Y, et al. Linear dimensionality reduction method based on topological properties[J/OL]. Information Sciences, 2023, 624: 493-511.

[15] ELHENAWY M, MASOUD M, GLASER S, et al. A New Approach to Improve the Topological Stability in Non-Linear Dimensionality Reduction[J/OL]. IEEE ACCESS, 2020, 8: 33898-33908.

[16] REN Yulan, TIAN Mi, LI Chunyan, et al. Gas chromatographic analysis of trace components in liquor[J]. China Brewing, 2011(7): 177-179.

[17] YAN Y, CHEN S, NIE Y, et al. Quantitative Analysis of Pyrazines and Their Perceptual Interactions in Soy Sauce Aroma Type Baijiu[J/OL]. Foods, 2021, 10(2): 441.

[18] LAN Wenbao, CHE Chang, TAO Chengyun. Spearman rank correlation-based single-acting spectral component selection and its application to SAR target identification[J/OL]. Journal of Radio Science, 2020, 35(3): 414-421.

[19] CHEN X, HOU Y, XI P. Parameter estimation of the structured illumination pattern based on principal component analysis (PCA): PCA-SIM[J/OL]. LIGHT-SCIENCE & APPLICATIONS, 2023, 12(1): 41.

[20] ZHAI Shuang, TOU Xiangguo, ZHANG Guiyu, et al. Rapid discrimination of white spirit base wine based on FT-NIR spectroscopy combined with KPCA-MD-SVM[J/OL]. Modern Food Science and Technology, 2022, 38(4): 248-253.

[21] HE Y, YE L, ZHU X, et al. Feature extraction based on PSO-FC optimizing KPCA and wear fault identification of planetary gear[J/OL]. Journal of Mechanical Science and Technology, 2021, 35(6): 2347-2357.

[22] ZHANG K, ZHANG K, BAO R. Prediction of gas explosion pressures: A machine learning algorithm based on KPCA and an optimized LSSVM[J/OL]. Journal of Loss Prevention in the Process Industries, 2023, 83: 105082.

[23] CHEN H, TAN C, WU T, et al. Discrimination between authentic and adulterated liquors by near-infrared spectroscopy and ensemble classification[J/OL]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2014, 130: 245-249.

[24] LIU G, WANG L, LIU D, et al. Hyperspectral Image Classification Based on Non-Parallel Support Vector Machine[J/OL]. Remote Sensing, 2022, 14(10): 2447.

[25] GÜNDOĞDU S. Efficient prediction of early-stage diabetes using XGBoost classifier with random forest feature selection technique[J/OL]. Multimedia Tools and Applications, 2023.

[26] CHEN R, JIA B, MA L, et al. Marine Radar Oil Spill Extraction Based on Texture Features and BP Neural Network[J/OL]. JOURNAL OF MARINE SCIENCE AND ENGINEERING, 2022, 10(12): 1904.

[27] YANG Y, LIU H, GU Y. A Model Transfer Learning Framework With Back-Propagation Neural Network for Wine and Chinese Liquor Detection by Electronic Nose[J/OL]. IEEE ACCESS, 2020, 8: 105278-105285.

[28] Tang, Jianqing. Quantitative investment based on BP neural network [D/OL]. Soochow University, 2019.

[29] XU Y, ZHAO J, LIU X, et al. Flavor mystery of Chinese traditional fermented baijiu: The great contribution of ester compounds[J/OL]. Food Chemistry, 2022, 369: 130920.