GitHub - vickytr44/Predicting-EMI-defaulter: Problem statement and approach:A banking customer needed to predict EMI defaulters based on 2395 variables. Data set had 17521 rows x 2395 columns with 25% missing values. Elimination and interpolation was performed to cleanse the data. Following cleaning, normalization was done with standard scaler. Finally, classifier model was developed using SVM with Gaussian kernel. The output accuracy from the model was 84%.

vickytr44 / Predicting-EMI-defaulter Public

Notifications You must be signed in to change notification settings
Fork 0
Star 1

Problem statement and approach:A banking customer needed to predict EMI defaulters based on 2395 variables. Data set had 17521 rows x 2395 columns with 25% missing values. Elimination and interpolation was performed to cleanse the data. Following cleaning, normalization was done with standard scaler. Finally, classifier model was developed using…