Sensory Characterization of Mezcal Using Free Choice Profiling and Machine Learning Tools

Authors

DOI:

https://doi.org/10.34190/icair.5.1.4381

Keywords:

consumer description, agave spirits, intelligent systems, sensory evaluation

Abstract

This study characterized the sensory profiles of four regional mezcals by integrating free-choice profiling with machine learning. A total of 1,148 consumers across four Mexican cities described the samples using their own vocabulary and rated attribute intensities. Machine learning analysis revealed significant correlations between sensory descriptors and demographic variables—such as age, gender, and origin—identifying detailed consumer preference patterns. The findings enable producers to customize products for specific market segments, enhancing their commercial strategies. By supporting traditional producers, this research also promotes sustainable economic growth in rural communities, aligning with the objectives of SDG 8.

Author Biographies

Sergio Erick García Barrón, Universidad Panamericana

Biochemical engineer specializing in food from the Technological Institute of Durango, he holds a Master's and a Doctorate in Science and Technology with a specialization in Sensory Evaluation and Consumer Science from the PICYT program at the CIATEJ campus. Currently, he is a postdoctoral researcher at the School of Hospitality of the Universidad Panamericana, Mexico City campus, conducting research on the perception of gluten-free products among the Mexican population. Previously, he worked as a postdoctoral researcher at the Industrial Biotechnology Unit of CIATEJ (2022-2024), analyzing public perception of traditional Mexican fermented beverages and supporting various projects requiring sensory evaluation tools and statistical data analysis. He is a member of the National System of Researchers (SNI) at the Candidate level. He is also a member of the International Network for Research and Studies in Gastronomy, coordinated by the Technological University of Nayarit. He has collaborated nationally and internationally with various higher education institutions on scientific and technological projects, focusing on sensory evaluation and consumer science. He is the author or co-author of several book chapters and scientific articles. His areas of interest include the sensory characterization of food, the study of consumer behavior, statistical analysis of information, experimental design, market research, and the correlation of instrumental and sensory measurements.

Claudia Ariadna Acero-Ortega, Universidad Panamericana

Research Professor - Universidad Panamericana - ESDAI, Mexico City. Pharmaceutical Chemist and Biologist, Universidad Colegio Mayor de Cundinamarca, Colombia, with a Master's and Doctorate in Science specializing in Food Science, National School of Biological Sciences - IPN, Mexico. Currently, she is a research professor at ESDAI. An academic with over 15 years of experience in food science, she has taught at both public universities in Colombia and private universities in Mexico City, in the Food Chemistry and Gastronomy undergraduate program, where she has supervised approximately 15 undergraduate theses. She worked as Coordinator of the National Laboratory Network in the area of ​​epidemiological surveillance within the InDRE (National Institute of Epidemiological Diagnosis and Reference) of the Ministry of Health. She is currently part of the international COST initiative as a member of the ACRYRED project, which aims to reduce consumer exposure to acrylamide through a cereal supply chain approach focused on asparagine. Her research also includes obtaining and using plant extracts with potential antimicrobial and/or functional activity, as well as applying predictive microbiology to food to extend shelf life while ensuring food safety. She has also worked with traditional fermented foods and the isolation, identification, and characterization of the beneficial bacteria present in them.

Downloads

Published

2025-12-04

Issue

Vol. 6 No. 1 (2025): Proceedings of the 6th International Conference on AI Research (ICAIR 2025)

Section

Academic