Adaptive Framework for Network Traffic Classification using Dimensionality Reduction and Clustering

Information security has become a very important topic especially during the last years. Web services are becoming more complex and dynamic. This offers new possibilities for attackers to exploit vulnerabilities by inputting malicious queries or code. However., these attack attempts are often recorded in server logs. Analyzing these logs could be a way to detect intrusions either periodically or in real time. We propose a framework that preprocesses and analyzes these log Ides. HTTP queries are transformed to numerical matrices using n-gram analysis. The dimensionality of these matrices is reduced using principal component analysis and diffusion map methodology. Abnormal log lines can then be analyzed in more detail. We expand our previous work by elaborating the cluster analysis after obtaining the low-dimensional representation. The framework was tested with actual server log data collected from a large web service. Several previously unknown intrusions were found. Proposed methods could be customized to analyze any kind of log data. The system could be used as a real-time anomaly detection system in any network where sufficient data is available.