Fault diagnosis system and method based on ensemble learning and computer readable medium

The present invention is a fault diagnosis system based on ensemble learning and a method thereof. A host performance degradation evaluation device uses a first ensemble learning method to estimate whether the host resources are degraded or not based on a basic monitoring characteristic data of multiple types of hosts. A digital convergence service fault locating device uses a second ensemble learning method to estimate the end-to-end fault estimation area based on an end-to-end characteristic data of multiple convergence services. An intelligent estimation and management device generates intelligent estimation and management information based on the output data of the host performance degradation evaluation device and the digital convergence service fault locating device. The present invention performs ensemble learning calculations after deep learning calculation devices suitable for respective data source styles, and produces estimated data including service host performance evaluation and full path area f