Predicting Damage Based on Gorkha Earthquake Dataset

Abstract

Earthquakes like the 2015 Gorkha earthquake in Nepal can cause massive damage to buildings in immediate and surrounding areas. It is important to people in those areas to be prepared for the possibility of such destruction to their structures. One way for those people to be prepared is to be able to predict the degree that buildings will be affected. We set out to predict the variable damage_grade in the dataset, which represents a level of damage to the building that was hit by the earthquake – low, medium, or complete destruction. Based on aspects of building location and construction, we developed a predictive model based on an initial exploratory analysis using descriptive statistics. We utilized R Studio, Python, and other software to construct our predictive model, create visuals, and compile a report of our findings.