Banks often depend on credit score prediction models to approve or deny a loan request. A good prediction model is necessary for a bank so that they can provide maximum credit without exceeding the risk threshold. This data science project uses credit score dataset which has fairly large volume of data (250K). The predictive models will be build following various approaches - random forests, graident boosting and logistic regression. At the end of the project you will build a predictive model that will automatically score each applicant with a credit score which is human readable and easy to interpret.