Loan business is one of the major income sources for bank. However, loan default problem is a major issue for loan business. With the rise of big data era and the development of machine learning techniques, nowadays we have more options for classifying and predicting loan default, other than manual processing. With a real-world dataset from a prestigious international bank, we demonstrate that the AdaBoost model can achieve a 100% accuracy for predicting loan default, outperforming other models including XGBoost, random forest, k nearest neighbors, and multilayer perceptrons. Our result shows the promising application of machine learning techniques in the financial industry.