The failure frequency of rotating machinery due to the bearing is high, and it causes the sudden shutdown of the system as well as financial loss. Therefore, researchers are devoted to determining the intelligent fault diagnosis method with a minimum number of features and less computational time. However, the bearing statistical feature space is broad, and identifying the ideal element for fault recognition is a challenging exercise. Thus, this paper presents the feature selection routine for bearing fault diagnosis. The proposed method identifies the ideal feature from feature space by applying the ensemble of feature ranking algorithm. The ideal feature set has trained using ensemble classifier for fault classification. The proposed method is evaluated using vibration data, and the results demonstrate that the proposed method provides a decent performance than the conventional feature selection method.