Surgical patients are usually at high risk of developing pressure ulcers after their operation. Usually, the pressure ulcers data sets are imbalanced. Therefore, this study aims to examine the real medical case of pressure ulcers with the use of support vector machines (SVMs). SVMs are used for forecasting and are a type of classification techniques. We utilize the measurement of sensitivity and specificity to compare the performances of SVMs with several classification techniques. The results indicated that SVMs performed better than the other classifiers for pressure ulcers prediction. In addition, the classifier of SVMs is a robust and powerful approach when facing the different ratios of training data sets.