Logo Lanfrica

GHEASANDRINEMAWEN/Africa-Population-Trend-and-Health-Analysis-UN-dataset-

Domain:

healthcare

Record type:

project
Creator:
GHE
Host:
# Africa Population Trend and Health Analysis - Assignment 2 ## Predictive Modeling and Hypothesis Testing This project builds upon Assignment 1's exploratory data analysis to test hypotheses about African demographic transitions using supervised machine learning. --- ## πŸ“ Project Structure ``` Africa-Population-Trend-and-Health-Analysis-UN-dataset-/ β”‚ β”œβ”€β”€ src/ # Python modules (OOP implementation) β”‚ β”œβ”€β”€ __init__.py β”‚ β”œβ”€β”€ data_processor.py # DataProcessor class - data loading & cleaning β”‚ β”œβ”€β”€ feature_engineer.py # FeatureEngineer class - feature creation β”‚ β”œβ”€β”€ model_trainer.py # ModelTrainer class - training & tuning β”‚ └── model_evaluator.py # ModelEvaluator class - evaluation & comparison β”‚ β”œβ”€β”€ notebooks/ β”‚ └── main_analysis.ipynb # Main workflow notebook with all analyses β”‚ β”œβ”€β”€ results/ # Auto-generated outputs β”‚ β”œβ”€β”€ figures/ # Model comparison & feature importance plots β”‚ β”œβ”€β”€ tables/ # Performance metrics CSV files β”‚ └── predictions/ # Model predictions for each hypothesis β”‚ β”œβ”€β”€ df_africa_cleaned.xls # Dataset (UN DESA population & health data) β”‚ └── README.md # This file ``` --- ## 🎯 Project Overview ### Research Questions This assignment tests **three hypotheses** about African demographic transitions (2010-2024): 1. **H1:** Does fertility rate predict maternal mortality? (Regression) 2. **H2:** Does life expectancy predict population growth rate? (Regression) 3. **H3:** Does under-five mortality predict life expectancy? (Regression) ### Dataset - **Source:** UN DESA Population Division - **Scope:** 54 African countries, 2010-2024 (annual observations) - **Key Indicators:** - Life expectancy at birth (both sexes) - Total fertility rate - Under-five mortality rate - Maternal mortality ratio - Population annual growth rate --- ## πŸš€ Getting Started ### Prerequisites Required …