Design of Log Analysis System Based on Deep Learning for Operation System Anomaly Detection

With the development of network communication technology, large number of log data will be generated during the operation of systems. The logs of the application system record the execution track of the system and exist in all components of the system, so it is possible to detect the abnormal behaviors of the system by processing logs data. However, it is very inefficient and difficult to detect and evaluate abnormal data by manual method, the existing research on system log analysis is limited and scattered, and there is a lack of unified platform to detect and manage system log files. Meanwhile, for maintenance personnel, the analysis of log data is too specialized and complex, and there is a lack of user-friendly platform for operation and maintenance decision makers to use and analyze. This project aims to build some scientific anomaly detection event models based on deep learning by mining rules from large number of complex data. And then give the design and implementation of a user-friendly abnormal behavior detection platform of operating system log files.