This is the submission for the DataStorm data analytics competition. DataStorm is a data analytics competition organized by the Roteract Club of the University of Moratuwa for university students islandwide.
The final model uses a XGBoost algorithm.This algorithm gave the highest accuracy among catboost, random forests and lightgbm. Feature selection has been done to reduce any overfitting and the training set has been balanced so that the majority class and minority class are equally represented.