The rolling bearing is a widely used mechanical part, but the existing technology has limitations on the fault diagnosis of rolling bearing. In order to improve the accuracy of rolling bearing fault diagnosis, a method of confidence rule base is proposed on the basis of fault samples by using historical data. Firstly, the typical faults of rolling bearing are analyzed, and their vibration data are obtained as samples to extract. The timedomain and timefrequency parameters are then fused and symptom parameters are extracted to build the confidence rule base of fault diagnosis. Finally, the abnormal data of rolling bearing are obtained through the experimental platform, which verifies the validity and accuracy of the established confidence rule base.