In this work, an architecture consisting entirely Self-Organizing Feature Maps is developed for network based intrusion detection. The principle interest is to analyze just how far such an approach can be taken in practice. To do so, the KDD benchmark dataset from the International Knowledge Discovery and Data Mining Tools Competition is employed. In this work, no content based feature is utilized. Experiments are performed on two-level and three-level hierarchies, training set biases and the contribution of features to intrusion detection. Results show that a hierarchical SOM intrusion detection system is as good as the other learning based approaches that use content based features. |