Feature or Attribute Extraction for Intrusion Detection System using Gain Ratio and Principal Component Analysis (PCA)

Abstract

Intrusion detection systems (IDS) refer to a category of defense tools that is used to provide warnings indicating that a system is under attack or intrusion. The IDS monitors activities within a network and alerts security administrators of suspicious activities. This paper extracted significant or highly relevant features or attributes of the Knowledge Discovery and Data Mining 1999 (KDD ‟99) dataset, which is a standard benchmark dataset for all intrusion problems using two features extraction techniques: Gain Ratio for discrete attributes and Principal Component Analysis (PCA) for continuous attributes. C# Programming language was used for the implementation. Also, Microsoft Excel was used to depict the result of the extraction. The result shows that thirteen (13) attributes were highly relevant and significant.