Bank Credit

July 25, 2016



Report Abuse
How to predict customer commitment to the bank according to his status using their data of cooperation with the bank as features.
Used built-in Azure ML functionality, Python, R and SQL to select the features used for training a machine learning model. Then created, trained, and evaluated a first machine learning model to classify bank customers as good or bad credit risks. This is done by splitting the data to 70% train model and 30% test model, then predict the scores by 2 class decision tree and finally evaluate the result which held accuracy of 77.5% and F1 Score 0.799