Personalized Food Nutrient Recommendations for Kids using AI and Behavior Analysis

This article introduces a custom-made dietary recommendation system for children that harnesses artificial intelligence (AI) and the Internet of Medical Things (IoMT) to provide adaptable, context-sensitive nutrition plans. It integrates a Decision Tree-based Nutritional Assessment, employs Reinforcement Learning (RL) for dietary modification, and incorporates a Context-Aware Recommendation Model to flexibly modify dietary advice based on individual health profiles and environmental variables such as ambient weather. Constructed using Python and evaluated on a robust computing platform featuring IoT-enabled wearables for real-time monitoring of calorie intake, the model exhibited notable advancements over pre-existing methods, achieving 92% accuracy, 90% precision, 91% recall, and an F1-score of 91%. These metrics affirm its effectiveness in providing pertinent dietary guidance. During a 30-day evaluation, compliance with diet plans rose from 75% to 90%, in contrast to other models with a top adherence of 85%. Furthermore, the context-aware feature of the model offered efficient caloric adjustments that boosted adherence by 10% in cooler climates compared to models lacking contextual inputs. User satisfaction surveys revealed that 80% of participants reported being “Very Satisfied” or “Satisfied,” illustrating the system's versatility and user-oriented approach. These results suggest the model's significant promise for real-world utilization in personalized child nutrition, enhancing dietary compliance, precision, and user involvement by fusi^on of context-driven modifications and RL.