This paper outlines a comprehensive technique for the sensing and recognizing of faults in rolling bearings used in rotary machines. The methodology integrates time-domain features with a hybrid EEMD & DWT technique. Signals from both faulty and healthy bearing systems, subjected to a range of operating conditions, are decomposed using EEMD-DWT to extract meaningful features. To further evaluate the effectiveness of this approach, a comparative analysis is conducted by applying the same set of time-domain features with EEMD and the conventional discrete wavelet transform (DWT) methods. After feature extraction based on the reconstructed signals, the resulting feature sets are classified using three machine learning algorithms: support vector machine (SVM), random forest, and XGBoost. The performance of these classifiers in detecting different fault types and sizes is analyzed, and the classification finding across all techniques are assessed to establish the success of the proposed approach.