The Global Hunger & Dietary Insights project aims to analyze global malnutrition trends, food consumption patterns, and nutritional availability using structured datasets.
This project processes and models data related to:
- Dietary Composition – analyzing per capita food consumption across countries and regions.
- Nutritional Composition – mapping food types to macronutrient content (protein, carbohydrates, fats).
- Global Hunger Index (GHI) – measuring hunger severity per country.
- Children’s Malnutrition Rates – tracking stunting, wasting, and underweight prevalence.
For efficiency, scalability, and accuracy, the project followed data modelling process involving normalisation, data cleaning, and proper schema design.
The data was normalized down to the Boyce-Codd Normal Form (BCNF) standard, achieved by addressing repeating groups (1NF), partial dependencies (2NF), and transitive dependencies (3NF).
-
First Normal Form (1NF): The raw dataset contained repeating groups where multiple food types were stored as separate columns rather than rows. These columns were unpivoted into rows.
-
Second Normal Form (2NF): Partial dependencies were removed by splitting dimension attributes into separate tables, leaving quantitative metrics and references in the fact tables.
Third Normal Form (3NF): Transitive dependencies were eliminated by creating an additional nutrition_dim table to separate food nutrient values from food_dim.