Abstract: Banks are making major part of profits through loans. Loan approval is a very important process for banking organizations. It is very difficult to predict the possibility of payment of loan by the customers because there is an increasing rate of loan defaults and the banking authorities are finding it more difficult to correctly access loan requests and tackle the risks of people defaulting on loans. In the recent years, many researchers have worked on prediction of loan approval systems. Machine learning technique is very useful in predicting outcomes for large amount of data. In this paper, four algorithms are used such as Random Forest algorithm, Decision Tree algorithm, Naive Bayes algorithm, Logistic Regression algorithm to predict the loan approval of customers. All the four algorithms are going to be used on the same dataset and going to find the algorithm with maximum accuracy to deploy the model. Henceforth, we develop bank loan prediction system using machine learning techniques, so that the system automatically selects the eligible candidates to approve the loan. Keywords: Loan approval, Loan Default, Random Forest algorithm, Decision Tree algorithm, Naive Bayes algorithm, Logistic Regression algorithm, Loan prediction, Machine learning.