Dashboard based clustering of toddler nutrition data to support malnutrition prevention

This Malnutrition remains a critical health issue among toddlers, especially in developing countries. This study aims to improve the monitoring of toddler nutritional status by integrating data mining techniques and interactive data visualization into a web-based dashboard. The objective is to segment toddlers based on anthropometric indices and provide accessible visual insights to aid local health officers. The dataset, sourced from a public health center, includes 5,708 toddler records with indicators such as weight for age (W/A), height for age (H/A), weight for height (W/H), and BMI for age (BMI/A). These were normalized using min-max scaling and clustered using the K-Means algorithm to identify patterns in nutritional status. The analysis resulted in three distinct clusters: Cluster 0 reflected normal weight with short stature; Cluster 1 showed severe undernutrition and stunting; and Cluster 2 indicated emerging risks of overnutrition despite normal growth metrics. A series of dashboards were designed using Tableau to visualize demographic trends, anthropometric distributions, and prevalence statistics. The findings show that 49.68% of toddlers were stunted, 41.97% were underweight, and 4.15% experienced wasting, while severe malnutrition affected 0.35%. The dashboard enables health practitioners to identify nutritional issues more effectively and prioritize interventions such as food supplementation, education, and referral care. This study demonstrates the practical value of combining data analytics and visualization in public health. Future research may expand dashboard implementation across broader regions and incorporate predictive features for real-time decision-making support.