A Data Science capstone project that predicts GDP Growth in Sub-Saharan African countries using the World Bank World Development Indicators (WDI) dataset and machine learning models.
# 🌍 GDP Growth Prediction Using World Bank Development Indicators (WDI)
## AnalystLab Africa Data Science Internship – Capstone Project
## 📖 Project Overview
This project was completed as the final Capstone Project for the **AnalystLab Africa Data Science Internship Program (Batch B)**.
The objective of this project was to develop machine learning models capable of predicting **GDP Growth** in **Sub-Saharan African countries** using socioeconomic indicators from the **World Bank World Development Indicators (WDI)** dataset.
The project demonstrates the complete Data Science workflow, including:
- Data Collection
- Data Cleaning & Preprocessing
- Exploratory Data Analysis (EDA)
- Feature Engineering
- Machine Learning Model Development
- Hyperparameter Tuning
- Model Evaluation
- Deployment using Streamlit
---
## 🎯 Problem Statement
Can GDP Growth be predicted using key socioeconomic indicators such as:
- Electricity Access
- Internet Users
- Inflation
- Population Growth
- Life Expectancy
- Secondary School Enrollment
- Trade (% of GDP)
for countries in **Sub-Saharan Africa**?
---
## 📊 Dataset
**Dataset:** World Bank World Development Indicators (WDI)
**Source:**
datatopics.worldbank.org
The dataset contains global development statistics covering over 200 countries and territories across multiple sectors including:
- Economy
- Education
- Health
- Infrastructure
- Trade
- Technology
- Environment
For this project, the analysis focused on **Sub-Saharan African countries**.
---
## Technologies Used
- Python
- Pandas
- NumPy
- Matplotlib
- Seaborn
- Scikit-learn
- Jupyter Notebook
- Streamlit
- Git & GitHub
---
## Machine Learning Models
The following regression models were developed and evaluated:
- Linear Regression
- Random Forest Regressor
- Gradient Boosting Regressor
- Optimized Random Forest (GridSearchCV)
---
## …