Loan Prediction Using Machine Learning Methods

Credit risk has always been the most important risk faced by commercial banks. Credit risk management has important practical significance for preventing credit risk. With the emerging of machine learning algorithms, numerous frameworks, including linear regression, support vector machine, random forest and decision tree are proposed with satisfying performance and robust accuracy. This paper will focus on predicting credit outcomes and calculating forecast accuracy from a given dataset. This paper adopts three algorithms, decision tree, random forest and logistic regression, to calculate the dataset from the Bank of Portugal separately and obtain relevant conclusions. Finally, the authors evaluate the advantages and disadvantages of the three methods according to the accuracy of the prediction results, and the conclusion is described as follow, First, all three methods have great potential on handling loan prediction task. Second, the logistic regression algorithm is the most accurate, which obtains 86.4% accuracy.