Predicting Earthquake Damage: A Case Study in Nepal

This project focuses on predicting the level of damage to buildings caused by the 2015 Gorkha earthquake in Nepal. The goal is to classify buildings into different damage grades based on their structural characteristics and the intensity of the earthquake.

The process involves several key stages:

1. Data Exploration and Cleaning: The dataset includes information on building structure, ownership, and damage grade. Initial exploration involves understanding the distribution of damage, identifying missing values, and handling categorical features.

2. Feature Engineering and Preparation: New features are engineered from existing ones, such as the age of the building or the number of floors. Categorical features are encoded using techniques like one-hot encoding. The data is then split into training and testing sets.

3. Modeling and Evaluation: Various classification models are trained and evaluated, including but not limited to Logistic Regression, Decision Trees, Random Forests, and Gradient Boosting models. The models are evaluated using metrics appropriate for multi-class classification, such as accuracy, precision, recall, F1-score, and confusion matrices.

4. Results and Conclusion: The best-performing model is identified, and its feature importance is analyzed to understand which building characteristics are most influential in predicting damage. The project aims to provide insights that can help in disaster preparedness and response efforts.