Prediction of Credit Card Approval

Credit risk as the board in banks basically centers around deciding probability of a customer's default or credit decay and how expensive it will end up being assuming happens. It is important to consider major factors predict beforehand consumers defaulting given their conditions. Which where machine learning model comes handy allows financial institutions whether customer, they are giving loan to, not. This project builds with best accuracy possible using python. First we load view dataset. The dataset has combination both mathematical non-mathematical elements, that contains values from various reaches, addition few missing passages. We preprocess guarantee AI pick can make great expectations. After information looking great, some exploratory examination done assemble our instincts. Finally, build if an individual's application for card be accepted. Using tools techniques then try improve model. uses Jupyter notebook python programming Data Analysis Machine Learning, attempted determine most essential parameters obtaining acceptance this project. built gave 86 % predicting approved not, considering mentioned holder. Even though achieved 86%, conducted grid search see could increase performance even further. However, models: random forest logistic regression, get data was percent.