Sophiajackrich/Bachelors_Final_Project

This is a natural disaster project specifically to analyze flood disaster in Africa # Bachelors_Final_Project **Flood Disaster Analysis in Africa using Supervised Machine Learning** **Project Overview** This repository contains the code and documentation for my Bachelor's thesis project, which explores the application of supervised machine learning techniques to analyze flood disasters in Africa. The goal is to develop a predictive model that can identify areas prone to flood disasters, enabling proactive measures to mitigate the impact on affected communities. **Data Source**: EM-DAT **Features**: - Geographical information (magnitude,magnitude scale) - Climate data (oigin) -Data period (start date, end date) - Historical flood data ('Origin','No. Affected', 'Total Deaths', 'No. Homeless') - Target Variable: - Flood severity (Magnitude) **Methodology** 1. Data Preprocessing: - Handling missing values - Feature scaling and normalization - Data splitting (training, testing, validation) 2. Model Selection and Training: - Supervised learning algorithms: Random Forest,Gradient Boosting, DecisionTreeRegressor, LinearRegression - Hyperparameter tuning using GridSearchCV 3. Model Evaluation: - Performance metrics: accuracy, precision, recall, F1-score, mean squared error (MSE) **Conclusion** This project demonstrates the potential of supervised machine learning in analyzing flood disasters in Africa. The developed model can support disaster risk reduction efforts by identifying high-risk areas and informing early warning systems. **Future Work** - Expand the dataset to include more features and flood events - Explore unsupervised and reinforcement learning techniques for flood disaster analysis - Collaborate with stakeholders to deploy the model in real-world scenarios **Acknowledgments** - Supervisors and mentors - fellow researchers and colleagues **Link_of_Published_work** ijeast.com