Machine learning combined with GC-FID for discrimination of different categories of maotai-flavor baijiu

IF 6.5 1区农林科学 Q1 CHEMISTRY, APPLIED

Food Chemistry: X Pub Date : 2025-05-01 DOI:10.1016/j.fochx.2025.102555

Liang Yang , Chun Xian , Shuai Li , Ye Wang , Xinying Wu , Qingcai Chen , Wenwu Zhao , Cheng Zhao , Xiaobo Li , Junjun He , Renyuan Chen , Chunlin Zhang

{"title":"Machine learning combined with GC-FID for discrimination of different categories of maotai-flavor baijiu","authors":"Liang Yang , Chun Xian , Shuai Li , Ye Wang , Xinying Wu , Qingcai Chen , Wenwu Zhao , Cheng Zhao , Xiaobo Li , Junjun He , Renyuan Chen , Chunlin Zhang","doi":"10.1016/j.fochx.2025.102555","DOIUrl":null,"url":null,"abstract":"<div><div>Maotai-flavor Baijiu, a traditional Chinese liquor produced via solid-state fermentation, exhibits diverse base Baijiu types due to variations in fermentation rounds, styles, and grades. While crucial for flavor complexity, current manual identification methods hinder blending efficiency and quality control. This study employed GC-FID and machine learning to analyze 410 base Baijiu samples. Decision Tree (74.36 %), XGBoost (92.9 %), and Random Forest (62.3 %) emerged as optimal classifiers for fermentation rounds, typical styles, and Chuntian grades, respectively. SHAP analysis revealed: (1) esters as primary markers for fermentation rounds, (2) ester-trimethylbutanol combinations for grade differentiation, and (3) multi-compound signatures (butyric acid, tetramethylpyrazine, 2-butanol et al.,) for style discrimination. Notably, marker compounds' flavor properties - beyond mere concentration - critically influenced their discriminative power, as evidenced by correlations with nine sensory dimensions.</div></div>","PeriodicalId":12334,"journal":{"name":"Food Chemistry: X","volume":"28 ","pages":"Article 102555"},"PeriodicalIF":6.5000,"publicationDate":"2025-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Food Chemistry: X","FirstCategoryId":"97","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S259015752500402X","RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CHEMISTRY, APPLIED","Score":null,"Total":0}

引用次数: 0

Abstract

Maotai-flavor Baijiu, a traditional Chinese liquor produced via solid-state fermentation, exhibits diverse base Baijiu types due to variations in fermentation rounds, styles, and grades. While crucial for flavor complexity, current manual identification methods hinder blending efficiency and quality control. This study employed GC-FID and machine learning to analyze 410 base Baijiu samples. Decision Tree (74.36 %), XGBoost (92.9 %), and Random Forest (62.3 %) emerged as optimal classifiers for fermentation rounds, typical styles, and Chuntian grades, respectively. SHAP analysis revealed: (1) esters as primary markers for fermentation rounds, (2) ester-trimethylbutanol combinations for grade differentiation, and (3) multi-compound signatures (butyric acid, tetramethylpyrazine, 2-butanol et al.,) for style discrimination. Notably, marker compounds' flavor properties - beyond mere concentration - critically influenced their discriminative power, as evidenced by correlations with nine sensory dimensions.

查看原文本刊更多论文

机器学习与GC-FID相结合用于茅台味白酒的鉴别

茅台味白酒是一种通过固态发酵生产的中国传统白酒，由于发酵周期、风格和等级的不同，茅台味白酒呈现出多种基础白酒类型。虽然对风味复杂性至关重要，但目前的人工识别方法阻碍了混合效率和质量控制。本研究采用气相色谱- fid和机器学习技术对410份白酒样品进行了分析。决策树（74.36%）、XGBoost（92.9%）和随机森林（62.3%）分别成为发酵轮、典型风格和春天等级的最佳分类器。SHAP分析显示：(1)酯是发酵回合的主要标记，(2)酯-三甲基丁醇组合用于等级区分，(3)多化合物特征（丁酸、四甲基吡嗪、2-丁醇等）用于风格区分。值得注意的是，标记化合物的风味特性——不仅仅是浓度——严重影响了它们的辨别能力，与九个感官维度的相关性证明了这一点。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Food Chemistry: X CHEMISTRY, APPLIED-

CiteScore

4.90

自引率

6.60%

发文量

315

审稿时长

55 days

期刊介绍： Food Chemistry: X, one of three Open Access companion journals to Food Chemistry, follows the same aims, scope, and peer-review process. It focuses on papers advancing food and biochemistry or analytical methods, prioritizing research novelty. Manuscript evaluation considers novelty, scientific rigor, field advancement, and reader interest. Excluded are studies on food molecular sciences or disease cure/prevention. Topics include food component chemistry, bioactives, processing effects, additives, contaminants, and analytical methods. The journal welcome Analytical Papers addressing food microbiology, sensory aspects, and more, emphasizing new methods with robust validation and applicability to diverse foods or regions.