Prediction of home loan approval with machine learning

In recent years, due to rapidly increasing housing prices worldwide, people from various income levels are applying to banks for mortgage loans to afford housing. Banks aim to keep their profit margins high and minimize risks when evaluating loan applications. Therefore, historical data becomes crucial in their evaluation process. Methods like machine learning are utilized for swift analysis of this data. This thesis aims to predict whether a customer applying for a mortgage loan from a bank will be approved using different machine learning models. In this study, four machine learning algorithms k-nearest neighbors, random forest, support vector machines, and logistic regression were employed using two different datasets to predict whether a bank will approve a mortgage loan. The results of these studies were compared using performance metrics. The best results were obtained with SVM and RF algorithms. Keywords: K-Nearest Neighbor Algorithm, Random Forest Classifier, Support Vector Machines (SVMs), Logistic Regression, Machine Learning