The highly reliable rotating equipment can prevent power plants from unplanned maintenance, which causes incidental costs. Vibration analysis can be the solution for maintaining rotating equipment's reliability. The failure symptoms can be extracted from the equipment's vibration data. The study has been done before to classify vibration symptoms using one angle of a bearing. However, bearing consists of horizontal, vertical, and axial angles. Therefore, this study uses combined horizontal, vertical, and axial angles to classify the vibration symptoms. In this study, the classification process was performed using the Decision Tree algorithm in two steps. The proposed method uses seven parameters extracted from the spectrum of vibration data as input for the first step classification and three parameters for the second step classification. This study resulted in 100% accuracy for classifying all five cases in the first step classification and 100% accuracy for unbalanced, parallel misalignment, and shaft bending cases in second step classification. This study found that the decision tree can be used to classify vibration symptoms.