Detection of Dietary Self-Regulation Patterns Using Agglomerative Hierarchical Clustering (AHC) for Ultra-Processed Foods Among Schoolchildren
Open Access
- 21 September 2026
- journal article
- Published by Editorial Académica Dragón Azteca in International Journal of Combinatorial Optimization Problems and Informatics
- Vol. 18 (1) , 372-384
Abstract
This research analyzed food self-control patterns regarding ultra-processed foods (UPF) in 238 schoolchildren (aged 8–12) using a cross-sectional design. A pictorial instrument assessed health, taste, and preference for 50 foods. Calculating a Self-Control Index and applying AI (Agglomerative Hierarchical Clustering with Canberra distance and full linkage) alongside Multidimensional Scaling, results showed significantly lower self-control for UPF versus healthy foods (Wilcoxon test). Clustering identified three profiles: high UPF attraction (n=189), balanced/healthy preference (n=45), and dissociation between knowledge and behavior (n=2). Findings highlight children's vulnerability to UPF, the need for differentiated interventions, and demonstrate AI-integrated indices as valuable for early risk profiling and precision nutrition. AcknowledgmentsThe authors express their sincere gratitude to the administration, teaching staff, and administrative personnel of the elementary school “Josefa Ortiz de Dominguez” in Ciudad del Carmen, Campeche, for the support provided for this study. Their invaluable collaboration and willingness to assist were fundamental to administering the research instruments to the students. Likewise, deep gratitude is extended to all the students who participated voluntarily and enthusiastically in this project. Spanish-language metadata / Metadatos en españolTítulo en español:Detección de patrones de autorregulación alimentaria mediante agrupamiento jerárquico aglomerativo (AHC) para alimentos ultraprocesados en escolares Resumen:Esta investigación analizó los patrones de autocontrol alimentario relacionados con los alimentos ultraprocesados (AUP) en 238 escolares de entre 8 y 12 años mediante un diseño transversal. Se utilizó un instrumento pictórico para evaluar la percepción de salubridad, el sabor y la preferencia respecto a 50 alimentos. Tras calcular un Índice de Autocontrol y aplicar técnicas de inteligencia artificial —agrupamiento jerárquico aglomerativo con distancia de Canberra y enlace completo— junto con escalamiento multidimensional, los resultados mostraron un autocontrol significativamente menor frente a los AUP en comparación con los alimentos saludables (prueba de Wilcoxon). El análisis de agrupamiento identificó tres perfiles: alta atracción por los AUP (n = 189), preferencia equilibrada/saludable (n = 45) y disociación entre conocimiento y comportamiento (n = 2). Los hallazgos ponen de manifiesto la vulnerabilidad de los niños frente a los AUP y la necesidad de intervenciones diferenciadas. Asimismo, demuestran el valor de los índices integrados con inteligencia artificial para la identificación temprana de perfiles de riesgo y la nutrición de precisión. Palabras Claves:Autocontrol alimentario, alimentos ultraprocesados, estadística no paramétrica, algoritmo de agrupamiento jerárquico. Smart citations:SciteAI. Dimensions.Open Alex.Keywords
This publication has 20 references indexed in Scilit:
- An investigation into food choices among 5–12 years children in relation to sensory, nutritional, and healthy product cuesFood Quality and Preference, 2023
- Individual differences in the influence of taste and health impact successful dietary self-control: A mouse tracking food choice study in childrenPhysiology & Behavior, 2020
- Hierarchical Clustering2020
- Development and body mass inversely affect children’s brain activation in dorsolateral prefrontal cortex during food choiceNeuroImage, 2019
- Unhealthy foods taste better among children with lower self-controlAppetite, 2019
- Statistical Analysis with Missing Data, Third Edition2019
- The individual environment, not the family is the most important influence on preferences for common non-alcoholic beverages in adolescenceScientific Reports, 2017
- Pattern classification with missing data: a reviewNeural Computing and Applications, 2009
- Inference and missing dataBiometrika, 1976
- Computer Programs for Hierarchical Polythetic Classification ("Similarity Analyses")The Computer Journal, 1966