This project focuses on predicting severe earthquake damage in Nepal using logistic regression and decision trees. The goal is to classify whether the damage is severe (1) or not severe (0) based on various building attributes.
The dataset used for this project contains the following features:
- Building ID
- Count of floors before the earthquake
- Count of floors after the earthquake
- Age of the building
- Plinth area in square feet
- Height of the building before the earthquake
- Height of the building after the earthquake
- Land surface condition
- Foundation type
- Roof type
- Ground floor type
- Type of other floors
- Building position
- Plan configuration
- Post-earthquake building condition
- Superstructure type
- Damage grade (target variable)
nepal_dataset.csv: Contains the dataset files.Nepal_Damage_Prediction.ipynb: Jupyter notebooks for data preprocessing, model training, and evaluation.