数据挖掘与建模技术在食品嗅觉和味觉感知与情绪认知中的应用  

Application of Data Mining and Modeling Techniques in Food Olfactory and Taste Perception and Emotional Cognition

作　　者：王蓓[1] 王颖[1] 王亚东 刘帅[2] 江滔 Wang Bei;Wang Ying;Wang Yadong;Liu Shuai;Jiang Tao(School of Food and Health,Beijing Technology and Business University,Beijing 100048;School of Mathematics and Statistics,Beijing Technology and Business University,Beijing 100048;Centre for Neuroscience Research,University of Burgundy,Lyon,Bron 69500,France)

机构地区：[1]北京工商大学食品与健康学院,北京100048 [2]北京工商大学数学与统计学院,北京100048 [3]勃艮第大学里昂神经科学研究中心,法国布龙69500

出　　处：《中国食品学报》2024年第7期1-13,共13页Journal of Chinese Institute Of Food Science and Technology

基　　金：国家自然科学基金项目(32072345)。

摘　　要：随着食品嗅觉和味觉感官及情绪认知相关的科学技术水平的不断提升,越来越多的仪器分析方法和实验设备用于上述领域的研究。检测方法的多样化、全面化以及检测精度的提升,也伴随着风味感知相关数据规模的扩增。如何从食品风味仪器分析结果以及消费者的情绪认知行为相关研究中收集的大量数据中获得关键信息,并建立数据间的关联,越来越被研究人员所关注。食品领域的数据挖掘与建模技术是利用食品生产及流通过程中获得的大量数据,实时、准确地监控食品产业链的各个环节中的物理化学变化,并预测这些变化可能对消费者感官特征及情绪认知产生的影响。目前在食品嗅觉和味觉感知领域,数据挖掘与建模技术可为食品科研人员和消费者提供前所未有的洞察力和分析能力。本文在针对经典的机器学习方法中的有监督和无监督的数据挖掘与建模方法以及深度学习方法的基础上,对食品感官属性研究以及情绪认知方面的最新应用进展进行分析,并展望数据挖掘与建模技术在食品嗅觉和味觉感知领域的应用前景,助力食品行业的科技进步和产业升级。With the continuous improvement of science and technology related to food olfactory and taste senses and emotional cognition,more and more instrumental analysis methods and experimental equipment are used in the research of the above fields.The diversification and comprehensiveness of detection methods and the improvement of detection accuracy are accompanied by the expansion of data scale related to flavor perception.How to obtain key information from the analysis results of food flavor instruments and the large amount of data collected in the research of consumers'emotional cognition and behavior,and establish the correlation between the data,has been paid more and more attention by researchers.Data mining and modeling technology in the field of food is to use a large amount of data obtained in the process of food production and circulation to monitor the physical and chemical changes in each link of the food industry chain in real time and accurately,and to predict the impact of these changes on consumers'sensory characteristics and emotional cognition.In the field of food smell and taste perception,data mining and modeling techniques can provide unprecedented insight and analytical power to food researchers and consumers.Based on the supervised and unsupervised data mining and modeling methods and deep learning methods of classical machine learning methods,this paper analyzes the latest application progress in the research of food sensory attributes and emotional cognition,and looks forward to the application prospect of data mining and modeling technology in the field of food olfactory and taste perception.Help the scientific and technological progress and industrial upgrading of the food industry.

关 键 词：数据挖掘与建模 机器学习 风味感知 感官属性 情绪认知 

分 类 号：TP311.13[自动化与计算机技术—计算机软件与理论] TS207.3[自动化与计算机技术—计算机科学与技术]