In this study, we investigate the use of various machine learning algorithms for predicting loan approval. We begin by collecting and pre-processing data on past loans, including information on the applicant’s income, employment status, and the outcomes of the loans. There are two datasets, one for training purpose and our task is to predict whether the applicant get loan approval or not from test dataset. We have trained the model first and evaluate a range of machine learning algorithms, including logistic regression, decision trees, random forests, and neural networks. We find that all these algorithms can predict loan approval with good accuracy, with the best performance achieved by the random forest model. This model can accurately classify loan approval and non-approval with an accuracy of over 84.45%. To further improve the accuracy of our predictions, we also consider the use of feature selection techniques in logistic regression be selecting three key features to train our model. We find that feature selection can improve the accuracy of our predictions by 67.56%to 82.11%.