cc_default_ML

I made a ML project to predict the possibilities of a particular credit card holder being a default. This project was a hackathon problem solution for the hackathon named "CONVOLVE" organised by IIT GUWAHATI. A CSV file for development data was attached which consisted of random sample of 96,806 Credit Cards along with the validation data of random sample of 41,792 Credit Card details. The machine had to be trained using the development data which already consisted of a variable named “bad_flag”,where “bad_flag=1” suggested that the given account holder is a default. I have also contributed to making some additional observations such as outlier analysis ,missing value analysis,feature importance and credit card prediction which consisted of account number and predicted probability of that particular account number being a default. All the neccessary csv files of these observations have been attached along with the validation data. Also a complete documentation of the project consisting of its observations and unique characteristics of data including the algorithms and libraries used has been attached alonged too.