Food and Fermentation Industries >

2026 , Vol. 52 >Issue 12: 316 - 323

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.044389

Authenticity assessment of Jiangxiangxing Baijiu by machine learning based on flavour components

  • LIU Yuhao ,
  • SHANG Yuhao ,
  • JIANG Lili ,
  • TIAN Rui ,
  • WANG Liping ,
  • YOU Xiaolong ,
  • SUN Youlan ,
  • HU Jianfeng
Expand
  • 1(Guizhou Xijiu Co.Ltd., Xishui 564622, China)
    2(Key Laboratory of Quality and Safety of Jiangxiangxing Baijiu, State Administration for Market Regulation, Guiyang 550000, China)
    3(Guizhou Province Technology Innovation Center for Jiangxiangxing Baijiu, Xishui 564622, China)

Received date: 2025-09-05

  Revised date: 2025-10-30

  Online published: 2026-07-17

Fold

Abstract

Jiangxiangxing Baijiu, with unique production process, is highly popular in the market.Due to its high profit margins and significant market demand, a wide range of counterfeiting methods have emerged.But the traditional authentication methods are difficult to cope with the increasingly sophisticated counterfeiting methods.Therefore, in this study, the flavour components of jiangxiangxing Baijiu were analyzed by gas chromatography and other technologies, and the authenticity discrimination model was constructed by machine learning.Firstly, the importance of flavour substances was assessed by pearson’s correlation and random forest, and 21 key flavour substances, including furfural, 2-tridecanone and propanol, were selected as discriminators from 96 compounds.Subsequently, comparing the discriminative performance of multilayer perceptron, K-nearest neighbour and logistic regression models, it was found that the logistic regression model performed the best.Finally, with the help of SHAP interpretability analysis, the model of authenticity discrimination was mainly affected by furfural, isobutyraldehyde, ethyl lactate, and other characteristic components.This study not only enriches the database of flavor components of jiangxiangxing Baijiu, but also provides a new technical reference for liquor quality supervision, which has important theoretical significance and practical value.

Key words： Jiangxiangxing Baijiu; flavor substances; machine learning; logistic regression

Cite this article

LIU Yuhao , SHANG Yuhao , JIANG Lili , TIAN Rui , WANG Liping , YOU Xiaolong , SUN Youlan , HU Jianfeng . Authenticity assessment of Jiangxiangxing Baijiu by machine learning based on flavour components[J]. Food and Fermentation Industries, 2026 , 52(12) : 316 -323 . DOI: 10.13995/j.cnki.11-1802/ts.044389

References

[1] 汪廷彩, 刘嘉飞, 杨中花, 等.酒类产品真假鉴别技术研究现状[J].食品安全质量检测学报, 2019, 10(13):4029-4035.
WANG T C, LIU J F, YANG Z H, et al.Research status of true or false identification technology of liquor products[J].Journal of Food Safety & Quality, 2019, 10(13):4029-4035.
[2] 杨宇. 浅析白酒包装材料防伪和造假技术现状及发展趋势[J].现代食品, 2024, 30(4):72-74.
YANG Y.Analysis on the present situation and development trend of anti-counterfeiting and counterfeiting technology of Baijiu packaging materials[J].Modern Food, 2024, 30(4):72-74.
[3] 杨博, 郭启鹏, 苏正, 等.现代检测技术在白酒真实性鉴别中的应用研究进展[J].食品工业, 2024, 45(2):192-196.
YANG B, GUO Q P, SU Z, et al.Research progress on the application of modern detection technology in Baijiu authenticity identification[J].The Food Industry, 2024, 45(2):192-196.
[4] RÍO SEGADE S, ŠRRAB D, PEZZUTO E, et al.Isomer composition of aroma compounds as a promising approach for wine characterization and differentiation:A review[J].Critical Reviews in Food Science and Nutrition, 2024, 64(2):334-353.
[5] 高云. 基于气、液相色谱技术的干邑白兰地真伪鉴定方法研究[D].广州:华南理工大学, 2018.
GAO Y.Investigation on the potential adulteration determination method of Cognac based on the technique of gas and liquid chromatography[D].Guangzhou:South China University of Technology, 2018.
[6] 孙细珍, 张帆, 杜佳炜, 等.基于毛细管气相色谱法和主成分分析的白酒真假酒判别分析[J].酿酒, 2021, 48(1):42-50.
SUN X Z, ZHANG F, DU J W, et al.Discriminant analysis for Chinese spirits authenticity based on capillary gas chromatography and principal component analysis[J].Liquor Making, 2021, 48(1):42-50.
[7] 杜玲玲. 金门高粱酒真伪鉴别技术研究[D].厦门:集美大学, 2015.
DU L L.Study of the authenticity identification of Kinmen Kaoliang liquor[D].Xiamen:Jimei University, 2015.
[8] DENG Z W, WANG T, ZHENG Y, et al.Deep learning in food authenticity:Recent advances and future trends[J].Trends in Food Science & Technology, 2024, 144:104344.
[9] TRUONG NGOC M, NGUYEN Q T, PHAM V T, et al.Assessing vodka authenticity and origin in Vietnam’s market:An analytical approach using FTIR and ICP-MS with multivariate statistics[J].Journal of Analytical Methods in Chemistry, 2024(1), 2024:5652559.
[10] CALLE J L P, BAREA-SEPÚLVEDA M, RUIZ-RODRÍGUEZ A, et al.Rapid detection and quantification of adulterants in fruit juices using machine learning tools and spectroscopy data[J].Sensors, 2022, 22(10):3852.
[11] HATEGAN A R, DAVID M, PIRNAU A, et al.Fusing 1H NMR and Raman experimental data for the improvement of wine recognition models[J].Food Chemistry, 2024, 458:140245.
[12] STÓJ A, WOJNOWSKI W, PŁOTKA-WASYLKA J, et al.The use of ultra-fast gas chromatography for fingerprinting-based classification of zweigelt and rondo wines with regard to grape variety and type of malolactic fermentation combined with greenness and practicality assessment[J].Molecules, 2024, 29(19):4667.
[13] 王金龙, 程平言, 陆伦维, 等.基于DB-WAX UI色谱柱气相色谱法检测5种香型白酒中45种挥发性风味物质[J].中国酿造, 2023, 42(4):238-243.
WANG J L, CHENG P Y, LU L W, et al.Detection of 45 volatile flavor compounds in 5 flavor types Baijiu by GC based on DB-WAX UI column[J].China Brewing, 2023, 42(4):238-243.
[14] 陆伦维, 孙优兰, 王金龙, 等.液液微萃取结合GC-MS-SIM法检测酱香型白酒中15种微量成分含量[J].酿酒科技, 2022(12):115-120;133.
LU L W, SUN Y L, WANG J L, et al.Determination of fifteen trace components in Jiangxiang Baijiu by liquid-liquid microextraction combined with GC-MS-SIM[J].Liquor-Making Science & Technology, 2022(12):115-120;133.
[15] 孙优兰, 骆红波, 王金龙, 等.酱香型白酒不同轮次基酒风味特征分析[J].食品与发酵工业, 2024, 50(17):343-352;I0003-I0005.
SUN Y L, LUO H B WANG J L, et al.Analysis of flavor characteristics of different rounds base liquor for Jiangxiangxing Baijiu[J].Food and Fermentation Industries, 2024, 50(17):343-352;I0003-I0005.
[16] GOYAL R, SINGHA P, SINGH S K.Spectroscopic food adulteration detection using machine learning:Current challenges and future prospects[J].Trends in Food Science & Technology, 2024, 146:104377.
[17] CHENG K Q, DONG R N, PAN F, et al.Characterization and feature selection of volatile metabolites in Yangxian pigmented rice varieties through GC-MS and machine learning algorithms[J].Frontiers in Nutrition, 2025, 12:1598875.
[18] 苏礼君, 李健, 孔建磊, 等.机器学习技术在食品风味分析中的研究进展[J].食品工业科技, 2024, 45(18):19-30.
SU L J, LI J, KONG J L, et al.Progress in research on machine learning for studies on food flavor analysis[J].Science and Technology of Food Industry, 2024, 45(18):19-30.
[19] 黄晓琛, 张凯利, 刘元杰, 等.机器学习与计算机视觉技术在食品质量评价中的研究进展[J].食品科学, 2024, 45(12):1-10.
HUANG X C, ZHANG K L, LIU Y J, et al.Research progress on machine learning and computer vision technology in food quality evaluation[J].Food Science, 2024, 45(12):1-10.
[20] 贺海洋, 李朋刚, 李亚林, 等.基于机器学习的花岗岩型铀矿含矿潜力分析—以湘中地区为例[J/OL].地学前缘, 2026.DOI:10.13745/j.esf.sf.2025.2.9.
HE H Y, LI P G, LI Y L, et al.Analysis of the mineral-bearing potential of granite-type uranium deposits based on machine learning:A case study of central Hunan area[J/OL].Earth Science Frontiers, 2026.DOI:10.13745/j.esf.sf.2025.2.9.
[21] 韩凤歌, 冷成彪, 陈加杰, 等.基于磷灰石微量元素组成的机器学习方法判别花岗岩成因类型[J].岩石学报, 2025, 41(2):737-750.
HAN F G, LENG C B, CHEN J J, et al.Machine learning method for discriminating granite genetic types based on trace element composition of apatite[J].Acta Petrologica Sinica, 2025, 41(2):737-750.
[22] 胡记磊, 吴文良, 王璟, 等.基于自适应LASSO的逻辑回归砂土液化判别模型[J].三峡大学学报(自然科学版), 2023, 45(2):67-72.
HU J L, WU W L, WANG J, et al.A discriminant model for sand liquefaction based on adaptive LASSO-LR[J].Journal of China Three Gorges University (Natural Sciences), 2023, 45(2):67-72.
[23] 杨婷婷, 黄永光, 左乾程, 等.酱香白酒酿造过程糠醛及糠醇的时空分布规律及溯源分析[J].食品科学, 2025, 46(5):38-47.
YANG T T, HUANG Y G, ZUO Q C, et al.Temporal and spatial distribution patterns and traceability analysis of furfural and furfuryl alcohol in the brewing process of sauce-flavor Baijiu[J].Food Science, 2025, 46(5):38-47.
[24] LUO X, FAN W L, XU Y, et al.Characterization of volatile bitter off-taste compounds in Maotai-flavor Baijiu (Chinese liquor)[J].LWT, 2025, 217:117363.
[25] 刘斐. 基于机器学习算法的酱香型白酒真实性多源数据表征研究[D].湘潭:湘潭大学, 2024.
LIU F.Multi-source data characterization of authenticity of Jiangxiang flavor Baijiu based on machine learning algorithms[D].Xiangtan:Xiangtan University, 2024.
[26] 黄治国, 刘履常, 李彦中, 等.不同香型白酒中的乳酸及乳酸乙酯旋光异构体的分析比较[J].现代食品科技, 2026, 42(3):1-8.
HUANG Z G, LIU L C, LI Y Z, et al.Comparison of the optical isomers of lactic acid and ethyl lactate in Baijiu of different aroma types[J].Modern Food Science and Technology, 2026, 42(3):1-8.
[27] 胡春红, 卢君, 尹学忠, 等.不同轮次酱香型基酒风味特征分析[J].中国酿造, 2025, 44(7):45-50.
HU C H, LU J, YIN X Z, et al.Analysis of flavor characteristics of sauce-flavor base liquor in different rounds[J].China Brewing, 2025, 44(7):45-50.
[28] FAN J Y, CHEN J, DU G C, et al.Mechanism of lactic acid synthesis during strong-flavor Baijiu fermentation revealed by transcriptomic analysis[J].Food Bioscience, 2024, 58:103780.
[29] 宋瑞滨, 邵泽良, 宋军.浓香型白酒发酵过程中多因素对正丁醇生成影响的研究[J].酿酒, 2018, 45(5):62-66.
SONG R B, SHAO Z L, SONG J.Study on the influence of multiple factors on the formation of N-butanol during the fermentation of Luzhou flavor liquor[J].Liquor Making, 2018, 45(5):62-66.
[30] 孔小勇, 曹永楠.浓香型白酒正丁醇生成规律及代谢途径的研究[J].酿酒科技, 2022(7):60-64.
KONG X Y, CAO Y N.Generation laws and metabolic pathways of N-butanol in Nongxiang Baijiu[J].Liquor-Making Science & Technology, 2022(7):60-64.
Options
Outlines

/

〈 〉

Powered by Beijing Magtech Co. Ltd,.