Abstract

V-Detector is real-valued negative selection algorithm designed to detecting anomalies in datasets. Many of previous experiments were focused on analysis usability of this algorithm to detection intruders in computer network. Intrusion Detection Systems (IDS) should be efficient and reliable due to large number of network connection and their diversity. Additionally, every connection is described as a record containing tens of numerical and symbolic attributes. Choosing proper structures to store samples of normal connections and detectors can meaningfully speed up the processes of learning and classification of observations; further reduced utilization of processor can be gained. In this paper we examine usefulness of some tree-based structures to represent data in an IDS.

Citations