Loan Status Prediction using SVM and Logistic Regression

Over the past few years, credit lines are the primary source of income for banks. The assets of bank are determined by how much "Loans" they have provided to the customer. Behind this credit line, one more important is that "Will the customer be capable of repaying the debt in a given time?". We can consider many attributes such as age, gender, income source & expenditure which limit the customer to pay the loan over a definite period [1]. To maximize the profit of bank, it’s very important to analyze the various models and compare them. One of the very effective models is Logistic regression. This model is widely used because it predicts over a group of independent variables, the categorical dependent variable. We have gathered data from Kaggle for our research and forecasting purpose. The data obtained from here will act as base for loan default prediction because it contains customer personal attribute like age, objective, previous credit utilization, credit extent, and credit period. After evaluating customer on following attribute, the bank should realize to whom they should provide the loan or to whom they shouldn’t. This can be analyzed by Logistic Regression and Support Vector Machine technique for truthful results. Our model doesn’t consider only the rich people should be given loan but it considers customer characteristics who satisfies the following above attribute, which plays an important role in credit choice and forecasting lone defaulters.