Intrusion Detection: Visualizing Attacks in IDS Data

Vast amounts of data are produced by Intrusion Detection Systems (IDS). Security professionals must sift through this data to distinguish between potential and real attacks. Mining raw data is difficult because all of the data can not be viewed at once. One way to overcome this is to display the data in images. Humans are visual creaturs and are able to easily find patterns in well created images. This paper gives a short introduction to image theory and several practical examples of using visual tools for mining IDS data.