Richter's predictor: Modeling earthquake damage using random forest

This research work deals with Richter’s Predictor, using the datasets which consider 39 features, available at a driven data platform, each row of dataset representing a unique result. By using approaches such as k-nearest neighbor (KNN), Decision Tree, artificial neural networks (ANNs), and Random Forest predict the damage caused by the earthquake in Nepal based upon their level of accuracy and performance of the training model. First, we applied Random Forests using some features. In this work, important features are selected with the help of the Random Forest approach. We obtained the highest accuracy when we selected an important feature by giving numeric values 0.02.