Improving Network Security through Traffic Log Anomaly Detection Using Time Series Analysis

Detecting and understanding the different anomalies that may occur in the network is a hard and non-well defined problem. The main propose in this document is to show the results obtained from the application of Data Mining techniques in order to detect aberrant behavior in the network. For that, we focused the detection on time series analysis, an unsupervised learning technique based on network flows that studies the past patterns to obtain future decisions. This approach have shown to be effective in preliminary anomaly detection as a part of bigger log correlation method or anomaly detector.