Development of an AI-Driven Computational Framework for Integrated Dietary Pattern Assessment: A Simulation-Based Proof-of-Concept Study - Code and Data

This repository contains all code and data required to reproduce the analysis presented in the manuscript "Development of an AI-Driven Computational Framework for Integrated Dietary Pattern Assessment: A Simulation-Based Proof-of-Concept Study." The computational framework integrates random forest classification, dimensionality reduction (t-SNE, PCA), and multi-objective optimization (NSGA-II) to evaluate dietary patterns across nutritional adequacy, environmental sustainability, and economic accessibility dimensions. Repository Contents: Code: Python scripts implementing simulation data generation, machine learning classification, nutrient adequacy prediction, dimensionality reduction, and multi-objective dietary optimization. Data: Synthetic dietary intake dataset for 1,500 simulated individuals across four dietary patterns (Mediterranean, Western, Plant-based, Mixed), including demographic characteristics, macronutrient and micronutrient intakes, environmental footprints (greenhouse gas emissions, water consumption), and economic costs. Documentation: README files with installation instructions, usage examples, and data dictionaries. All analyses use deterministic random seed (42) ensuring full reproducibility.